# Language Engineering

Language engineering is a computing discipline involving computer systems and software techniques for processing, recognizing, understanding, interpreting, and generating human language.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## NL2AGBench: Benchmarking LLM Auto-Formalization for AlphaGeometry

DevFeed: [NL2AGBench: Benchmarking LLM Auto-Formalization for AlphaGeometry](<https://devfeed.tech/articles/nl2agbench-benchmarking-llm-auto-formalization-for-alphageometry-145365.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2608.28481>)

Author: Samuel Xiao, Judy Song, Rory Hu, Ziliang Zong

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [domain-specific language](<https://devfeed.tech/topics/domain-specific-language.md>), [Enhancing AI model performance](<https://devfeed.tech/topics/enhancing-ai-model-performance.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [machine-translation](<https://devfeed.tech/topics/machine-translation.md>)

Tags: [arxiv](<https://devfeed.tech/tags/arxiv.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [domain-specific-language](<https://devfeed.tech/tags/domain-specific-language.md>), [error](<https://devfeed.tech/tags/error.md>), [execution](<https://devfeed.tech/tags/execution.md>), [few-shot](<https://devfeed.tech/tags/few-shot.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [llm](<https://devfeed.tech/tags/llm.md>), [open-source-models](<https://devfeed.tech/tags/open-source-models.md>), [performance](<https://devfeed.tech/tags/performance.md>), [source](<https://devfeed.tech/tags/source.md>), [taxonomy](<https://devfeed.tech/tags/taxonomy.md>), [translation](<https://devfeed.tech/tags/translation.md>)

### AI overview

NL2AGBench evaluates how well LLMs translate English geometry problems into AlphaGeometry's specialized DSL, using execution within AlphaGeometry to assess whether translations work rather than relying only on textual similarity. Across ten open and closed models at multiple scales, leading closed models exceed 80% executable translation, while even the largest open models struggle to preserve geometric constraints and produce valid formalizations. The benchmark distinguishes syntax from logic errors and reports measurable improvements from few-shot prompting, fine-tuning, and human-guided hints across multiple model families.

### Source excerpt

arXiv:2608.28481v2 Announce Type: replace-cross Abstract: Recent advances in large language models (LLMs) have demonstrated strong capabilities in natural language understanding and mathematical reasoning. However, their ability to translate informal mathematical problems into formal representations remains underexplored. This limitation is particularly important for neuro-symbolic geometry systems such as AlphaGeometry, whose theorem-proving engine requires inputs in a specialized domain-specific language (DSL). Although AlphaGeometry achieves near-IMO gold-medalist performance, manually converting natural-language problems into its formal syntax remains a significant usability bottleneck. To address this challenge, we introduce the Natural Language to AlphaGeometry Benchmark (NL2AGBench), which evaluates LLMs in translating English geometry problems into AlphaGeometry-compatible formal representations. NL2AGBench uses execution-based verification within AlphaGeometry to assess translation quality rather than relying solely on textual similarity. We evaluate ten state-of-the-art open- and closed-source LLMs across multiple parameter scales and analyze executable translation accuracy, syntactic correctness, and error characteristics. Our experiments reveal a substantial performance gap between closed- and open-source models: leading closed-source models achieve executable translation rates above 80%, while even the largest open-source models struggle to consistently preserve geometric constraints and produce valid formalizations. We introduce an error taxonomy distinguishing syntax and logic errors and investigate mitigation strategies, including few-shot prompting, fine-tuning, and human-guided hinting, which yield measurable improvements across multiple model families.

## Natural Language Programming: Rethinking Language, Memory, and Cognition in the Era of Generative AI

DevFeed: [Natural Language Programming: Rethinking Language, Memory, and Cognition in the Era of Generative AI](<https://devfeed.tech/articles/natural-language-programming-rethinking-language-memory-and-cognition-in-the-era-of-generative-ai-140121.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23106>)

Author: Giovanni Adorni, Daniele Grosso

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Generative AI](<https://devfeed.tech/topics/generative-ai.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-memory](<https://devfeed.tech/tags/ai-memory.md>), [ai-workflows](<https://devfeed.tech/tags/ai-workflows.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [architectures](<https://devfeed.tech/tags/architectures.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [building](<https://devfeed.tech/tags/building.md>), [but](<https://devfeed.tech/tags/but.md>), [challenge](<https://devfeed.tech/tags/challenge.md>), [cognition](<https://devfeed.tech/tags/cognition.md>), [coherence](<https://devfeed.tech/tags/coherence.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [era](<https://devfeed.tech/tags/era.md>), [generative](<https://devfeed.tech/tags/generative.md>), [generative-ai](<https://devfeed.tech/tags/generative-ai.md>), [language](<https://devfeed.tech/tags/language.md>), [memory](<https://devfeed.tech/tags/memory.md>), [natural-language](<https://devfeed.tech/tags/natural-language.md>), [programming](<https://devfeed.tech/tags/programming.md>)

### AI overview

The paper frames Natural Language Programming (NLPg) as a way to understand prompts as cognitive structures that guide human-AI reasoning, rather than only as inputs for language processing. Prompts can externalize goals, constraints, assumptions, validation criteria, and intermediate reasoning. It proposes "Dialogic Working Memory" for how prompts and conversation states act as temporary external memory supporting continuity and reflection. The abstract maps BTMA, SKGE, and RMR onto episodic, semantic, and metacognitive layers intended to support long-horizon conversational coherence. These are theoretical proposals and mappings; the supplied text does not provide implementation details or evidence comparing the architectures.

### Source excerpt

arXiv:2611.23106v1 Announce Type: cross Abstract: Generative Artificial Intelligence is transforming the role of natural language in computational systems. In classical Natural Language Processing (NLP), language was treated as an object of computation: something to analyse, classify, translate, or generate. Large Language Models (LLMs) challenge this perspective, enabling natural language to function increasingly as an executable cognitive medium through which humans structure reasoning, orchestrate problem solving, and regulate hybrid human/AI workflows. This paper introduces Natural Language Programming (NLPg) as a theoretical and epistemological extension of traditional NLP. Rather than focusing on language understanding alone, NLPg conceptualises prompts as dynamic cognitive artefacts that externalise goals, constraints, assumptions, validation criteria, and intermediate reasoning states. Prompting is thus not merely a technical interaction technique, but a form of dialogic cognitive programming operating within distributed human/AI memory architectures. Building on theories of distributed cognition, extended mind, metacognition, and dialogic learning, the paper proposes the notion of Dialogic Working Memory, describing how prompts and conversational states function as temporary external memory structures supporting reasoning continuity and reflective regulation. We further map existing memory architectures (BTMA, SKGE, RMR) onto this framework, illustrating how episodic, semantic, and metacognitive layers support long-horizon conversational coherence. Finally, implications for computational linguistics and AI/mediated knowledge systems are discussed, suggesting a post symbolic and post interface paradigm in which natural language operates simultaneously as communication medium, executable specification, cognitive scaffold, and distributed memory architecture.

## One Question, Four Voices: How Advice for Alzheimer's Caregiving Differs Between Caregivers, Clinicians, and Large Language Models

DevFeed: [One Question, Four Voices: How Advice for Alzheimer's Caregiving Differs Between Caregivers, Clinicians, and Large Language Models](<https://devfeed.tech/articles/one-question-four-voices-how-advice-for-alzheimer-s-caregiving-differs-between-caregivers-clinicians-and-large-language-models-137300.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18604>)

Author: Congning Ni, Jinkyung Katie Park, Yang Li, Sarvech Qadir, Sean S. Huang, James S. Powers, Julia A. Hiner, Vania Leung, Patricia Commiskey, Lijun Song, Qingxia Chen, Pamela J. Wisniewski, Bradley A. Ma

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Vectors & LLMs](<https://devfeed.tech/topics/vectors-llms.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>)

Tags: [advice](<https://devfeed.tech/tags/advice.md>), [ai](<https://devfeed.tech/tags/ai.md>), [chatgpt](<https://devfeed.tech/tags/chatgpt.md>), [community](<https://devfeed.tech/tags/community.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-hc](<https://devfeed.tech/tags/cs-hc.md>), [emotion](<https://devfeed.tech/tags/emotion.md>), [financial](<https://devfeed.tech/tags/financial.md>), [guidance](<https://devfeed.tech/tags/guidance.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [large-language-models](<https://devfeed.tech/tags/large-language-models.md>), [legal](<https://devfeed.tech/tags/legal.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llms](<https://devfeed.tech/tags/llms.md>), [modeling](<https://devfeed.tech/tags/modeling.md>), [online](<https://devfeed.tech/tags/online.md>), [peer](<https://devfeed.tech/tags/peer.md>), [retrieval](<https://devfeed.tech/tags/retrieval.md>), [scope](<https://devfeed.tech/tags/scope.md>), [structured](<https://devfeed.tech/tags/structured.md>), [style](<https://devfeed.tech/tags/style.md>)

### AI overview

The study compared answers to 85 real Alzheimer's and related dementia caregiving questions from peer caregivers, physicians, ChatGPT, and retrieval-augmented CareGPT. Peers offered time-anchored narratives and broader community-grounded coverage, including emotional, legal, financial, and daily-care concerns; clinicians gave concise, clinically bounded advice; and both LLMs produced formal, verbose, structured responses concentrated on advice. LLMs scored higher than peers on politeness and emotional-support markers and were comparable to clinicians on informativeness and physician alignment, but were less narrative and contextually situated. CareGPT differed little from ChatGPT, suggesting its retrieval and prompt injection did not substantially change its communicative stance. The authors conclude AI may complement, but not replace, human caregiver support.

### Source excerpt

arXiv:2611.18604v1 Announce Type: cross Abstract: Online caregiver forums are an important source of emotional and informational support for Alzheimer's disease and related dementias (ADRD). We compare responses to 85 real caregiver questions from ALZConnected across four responder groups: peer caregivers, physicians, ChatGPT, and a retrieval-augmented CareGPT grounded in peer discussions. Using 28 measures spanning linguistic form, emotional support, informational support, and topic scope, we examine how these voices differ when answering the same questions. Peers provided time-anchored narratives and community grounding; clinicians offered concise, advice-forward, clinically bounded guidance; and LLMs produced a formal, verbose, highly structured consultation style. CareGPT showed minimal measurable separation from ChatGPT, suggesting that retrieval plus prompt injection alone was insufficient to shift communicative stance. LLM outputs scored higher than peers on politeness and emotion-support markers and were comparable to clinicians on informativeness and physician-alignment measures, but were less narrative and less contextually situated. Topic modeling showed that LLM responses concentrated on advice-oriented themes, whereas peers covered a broader range, including emotional processing, legal and financial concerns, and daily care. These findings suggest that AI can complement, but not replace, human contributions to ADRD caregiver support and motivate hybrid systems that combine structured AI delivery with the distinct strengths of human voices.

## LEGAL-UQA: Testing Constitutional Question-Answering Systems for Urdu Speakers with a Parallel English-Urdu Benchmark

DevFeed: [LEGAL-UQA: Testing Constitutional Question-Answering Systems for Urdu Speakers with a Parallel English-Urdu Benchmark](<https://devfeed.tech/articles/legal-uqa-testing-constitutional-question-answering-systems-for-urdu-speakers-with-a-parallel-english-urdu-benchmark-134944.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2410.13013>)

Author: Faizan Faisal, Umair Yousaf, Muhammad Hassnain, Zeerak Babar

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [search-engines](<https://devfeed.tech/topics/search-engines.md>), [test](<https://devfeed.tech/topics/test.md>)

Tags: [benchmark](<https://devfeed.tech/tags/benchmark.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [english](<https://devfeed.tech/tags/english.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [human-review](<https://devfeed.tech/tags/human-review.md>), [legal](<https://devfeed.tech/tags/legal.md>), [parallel](<https://devfeed.tech/tags/parallel.md>), [qa](<https://devfeed.tech/tags/qa.md>), [questions](<https://devfeed.tech/tags/questions.md>), [retrieval](<https://devfeed.tech/tags/retrieval.md>), [testing](<https://devfeed.tech/tags/testing.md>)

### AI overview

LEGAL-UQA provides 619 parallel English-Urdu question-answer pairs linked to passages from Pakistan's Constitution. Its evaluation separates answering Urdu questions with a passage supplied from retrieving the relevant passage. Model rankings varied with the answer-similarity measure, and simple word matching retrieved passages more effectively than the tested embedding methods. At the evaluated cutoff, the automated judge JEV agreed with human grades less often than a baseline that gave every answer full credit. An audit also found generated reference answers that omitted source-supported information. The authors recommend evaluating retrieval and answering separately and using human review to validate judges and references.

### Source excerpt

arXiv:2410.13013v2 Announce Type: replace-cross Abstract: Legal information assistants could help Urdu speakers understand their constitutional rights, but building reliable systems requires test data and evaluation methods suited to the language. We present LEGAL-UQA, a parallel English-Urdu benchmark of 619 question-answer pairs linked to passages from Pakistan's Constitution. Each example includes the question, reference answer, and source passage in both languages. We evaluate two components separately: answering Urdu questions when the source passage is supplied, and retrieving the passage from which each question was generated. We also compare JEV, a decision model used as an automated judge, with human grades. The findings show that model rankings depend on how answer similarity is measured, while a simple word-matching method retrieves source passages more effectively than the embedding methods tested. At the evaluated cutoff, JEV agrees with human grades less often than a baseline that gives every answer full credit. An audit also identifies cases where generated references omit information supported by the source. Together, these findings highlight the need to test retrieval and answering separately, include simple retrieval baselines, and validate automated judges and reference answers with human review when engineering legal QA software for Urdu speakers.

## Typing Matters: Muscle, Language and Typing Behaviour in Live Keystroke Decoding from sEMG

DevFeed: [Typing Matters: Muscle, Language and Typing Behaviour in Live Keystroke Decoding from sEMG](<https://devfeed.tech/articles/typing-matters-muscle-language-and-typing-behaviour-in-live-keystroke-decoding-from-semg-135202.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18464>)

Author: Abhilash Durgam, Nyle Siddiqui, Mubarak Shah, Qiushi Fu

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [Decoding](<https://devfeed.tech/topics/decoding.md>)

Tags: [autoregressive](<https://devfeed.tech/tags/autoregressive.md>), [behaviour](<https://devfeed.tech/tags/behaviour.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [context](<https://devfeed.tech/tags/context.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [decoder](<https://devfeed.tech/tags/decoder.md>), [decoding](<https://devfeed.tech/tags/decoding.md>), [error](<https://devfeed.tech/tags/error.md>), [fixed](<https://devfeed.tech/tags/fixed.md>), [future](<https://devfeed.tech/tags/future.md>), [improvement](<https://devfeed.tech/tags/improvement.md>), [inference](<https://devfeed.tech/tags/inference.md>), [keys](<https://devfeed.tech/tags/keys.md>), [language](<https://devfeed.tech/tags/language.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [live](<https://devfeed.tech/tags/live.md>), [model](<https://devfeed.tech/tags/model.md>), [moving](<https://devfeed.tech/tags/moving.md>), [session](<https://devfeed.tech/tags/session.md>), [signal](<https://devfeed.tech/tags/signal.md>), [typing](<https://devfeed.tech/tags/typing.md>)

### AI overview

The proposed sEMG typing decoder predicts keys live with no future context, carrying session history forward instead of processing isolated fixed windows. A state space model combines muscle signals with an autoregressive language model conditioned on text already typed, including typing behaviors such as typos, backspaces, and line breaks. To train per-character predictions without sEMG-to-character transcripts, the authors retokenize WikiText into human-like typing sequences. On emg2qwerty's cross-user benchmark, the model reports a 25.13% character error rate, a 5.4% relative improvement over the previous best with 6.7 times fewer parameters.

### Source excerpt

arXiv:2611.18464v1 Announce Type: new Abstract: Surface electromyography (sEMG) from the wrists can recover typed text without a keyboard. It is usually posed as a signal-inference problem: characters are read out of the muscle signal and character error rate is reported offline. We argue that this framing removes most of the problem. The decoder is constrained simultaneously by the muscles, by the language being typed, and by typing behaviour such as typos, corrections, backspaces and line breaks, and a usable system must predict keys live while the hand keeps moving. Existing systems fall short on two counts: (1) they process sEMG in fixed windows and carry nothing from one window to the next, discarding context accumulated over a session, and (2) they lack any understanding of the linguistic structure that drives typing, so every keystroke is inferred from the muscle signal alone. We propose a fully causal, zero-lookahead sEMG typing decoder built on a state space model (SSM), in which a single model decodes a whole session without windowed partitions, conditioned on what has already been typed. We achieve this by fusing a traditional typing decoder with an efficient autoregressive language model, instilling both sEMG and linguistic understanding in one model. To train such a model for per-character typing rather than whole-word text, we introduce a retokenization method that rewrites ordinary prose as human-like typing, with keystrokes, typos, backspaces and line breaks included, converting WikiText into a large typing corpus where none existed. Our model, having never seen a single sEMG-to-character transcript in typing, achieves a state-of-the-art character error rate of 25.13% on emg2qwerty's cross-user benchmark, a 5.4% relative improvement over the previous best with 6.7 times fewer parameters, while running strictly causally with no future context.

## Legal interpretation and AI: from expert systems to argumentation and LLMs

DevFeed: [Legal interpretation and AI: from expert systems to argumentation and LLMs](<https://devfeed.tech/articles/legal-interpretation-and-ai-from-expert-systems-to-argumentation-and-llms-131889.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2603.05392>)

Author: V\\'aclav Jane\\v{c}ek, Giovanni Sartor

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [knowledge-engineering](<https://devfeed.tech/topics/knowledge-engineering.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [legal](<https://devfeed.tech/tags/legal.md>), [llms](<https://devfeed.tech/tags/llms.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [system](<https://devfeed.tech/tags/system.md>)

### AI overview

AI and Law research has approached legal interpretation through three methods. Expert systems focus on transferring human-generated interpretations into knowledge bases for consistent application. Argumentation approaches represent interpretive arguments and their dialectical interactions, using frameworks to assess whether interpretive claims are acceptable. Machine learning, including general and specialized language models, generates interpretive suggestions and arguments; such models are increasingly used in legal practice. This overview reflects the supplied abstract.

### Source excerpt

arXiv:2603.05392v2 Announce Type: replace Abstract: AI and Law research has encountered legal interpretation in different ways, in the context of its evolving approaches and methodologies. Research on expert system has focused on legal knowledge engineering, with the goal of ensuring that human-generated interpretations can be precisely transferred into knowledge-bases, to be consistently applied. Research on argumentation has aimed at representing the structure of interpretive arguments, as well as their dialectical interactions, to assess of the acceptability of interpretive claims within argumentation frameworks. Research on machine learning has focused on the automated generation of interpretive suggestions and arguments, through general and specialised language models, now being increasingly deployed in legal practice.

## Cognitive Multi-Agent Search and Rescue: Language-Grounded Hierarchical Reinforcement Learning for Collaborative Robotics

DevFeed: [Cognitive Multi-Agent Search and Rescue: Language-Grounded Hierarchical Reinforcement Learning for Collaborative Robotics](<https://devfeed.tech/articles/cognitive-multi-agent-search-and-rescue-language-grounded-hierarchical-reinforcement-learning-for-collaborative-robotics-128263.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.13161>)

Author: Sepideh Ravanbakhsh, Hamidreza Moradi

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [Machine Intelligence](<https://devfeed.tech/topics/machine-intelligence.md>)

Tags: [ablation](<https://devfeed.tech/tags/ablation.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [adapting](<https://devfeed.tech/tags/adapting.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-behavior](<https://devfeed.tech/tags/agent-behavior.md>), [agent-framework](<https://devfeed.tech/tags/agent-framework.md>), [agents](<https://devfeed.tech/tags/agents.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [autonomous-systems](<https://devfeed.tech/tags/autonomous-systems.md>), [based](<https://devfeed.tech/tags/based.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [coherence](<https://devfeed.tech/tags/coherence.md>), [combined](<https://devfeed.tech/tags/combined.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ce](<https://devfeed.tech/tags/cs-ce.md>), [language](<https://devfeed.tech/tags/language.md>), [learning](<https://devfeed.tech/tags/learning.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [search](<https://devfeed.tech/tags/search.md>)

### AI overview

The proposed LG-HRL framework coordinates UAV and UGV teams in simulated search and rescue by having an LLM turn human instructions into semantic cues for strategy selection and role assignment, alongside hierarchical policies and a shared cognitive workspace. A formal LG-MAHDP models how language grounding, hierarchical control, and distributed coordination interact. In disaster simulations, the authors report 2.3x faster learning, 37% greater coordination coherence, 46% higher subgoal accuracy, 31% better hazard avoidance, and 20-30% higher victim rescue rates. The system showed emergent role specialization and adapted to verbal updates; ablations indicate that both language grounding and the shared workspace contribute to performance. Results are from simulation.

### Source excerpt

arXiv:2611.13161v1 Announce Type: new Abstract: Effective search and rescue (SAR) operations require autonomous systems capable of coordinating multiple heterogeneous agents, adapting to dynamic hazards, and aligning with human mission intent. Existing language-guided reinforcement learning (RL) methods primarily focus on single-agent instruction following and lack mechanisms for team-level coordination in complex, partially observable environments. This paper presents a Cognitive Multi-Agent Framework based on Language-Grounded Hierarchical Reinforcement Learning (LG-HRL) that integrates natural-language understanding, hierarchical policy decomposition, and a shared cognitive workspace to achieve cohesive multi-agent behavior in SAR scenarios. Human instructions are interpreted by a large language model (LLM) and transformed into structured semantic cues that guide high-level strategy selection and role assignment across UAV and UGV agents. A formal Language-Grounded Multi-Agent Hierarchical Decision Process (LG-MAHDP) is introduced to model the interplay between semantic grounding, hierarchical control, and distributed coordination. Comprehensive experiments in a simulated disaster environment show that LG-HRL improves learning speed by 2.3 times, increases coordination coherence by 37%, and enhances subgoal accuracy and hazard avoidance by 46% and 31%, respectively. The system achieves 20-30% higher victim rescue rates, demonstrates emergent role specialization, and adapts immediately to human verbal updates. Ablation studies confirm that both the shared cognitive workspace and language grounding are critical to overall system performance. Overall, this work establishes a new paradigm for language-guided multi-agent SAR robotics, demonstrating that semantic grounding combined with hierarchical reasoning enables safer, more efficient, and more interpretable autonomous rescue operations.

## StakeBench: Evaluating Language Understanding Grounded in Market Commitment

DevFeed: [StakeBench: Evaluating Language Understanding Grounded in Market Commitment](<https://devfeed.tech/articles/stakebench-evaluating-language-understanding-grounded-in-market-commitment-128116.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2605.26074>)

Author: Yunhua Pei, Jingyu Hu, Yiwei Shi, Hongnan Ma, Weiru Liu, John Cartlidge

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [LLM evaluation / benchmarking](<https://devfeed.tech/topics/llm-evaluation-benchmarking.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>)

Tags: [annotation](<https://devfeed.tech/tags/annotation.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [data](<https://devfeed.tech/tags/data.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [failures](<https://devfeed.tech/tags/failures.md>), [financial](<https://devfeed.tech/tags/financial.md>), [incentives](<https://devfeed.tech/tags/incentives.md>), [language](<https://devfeed.tech/tags/language.md>), [llms](<https://devfeed.tech/tags/llms.md>), [markets](<https://devfeed.tech/tags/markets.md>), [metrics](<https://devfeed.tech/tags/metrics.md>), [nlp](<https://devfeed.tech/tags/nlp.md>), [performance](<https://devfeed.tech/tags/performance.md>), [polymarket](<https://devfeed.tech/tags/polymarket.md>), [position](<https://devfeed.tech/tags/position.md>), [q-fin-gn](<https://devfeed.tech/tags/q-fin-gn.md>), [scale](<https://devfeed.tech/tags/scale.md>), [supervision](<https://devfeed.tech/tags/supervision.md>), [tasks](<https://devfeed.tech/tags/tasks.md>), [trading](<https://devfeed.tech/tags/trading.md>)

### AI overview

StakeBench evaluates language understanding against observable market commitments rather than labels based on outside perceptions. It links 560,876 comments across 2,261 resolved Polymarket and Manifold markets to verified positions, subsequent trading actions, and odds trajectories, using these records as supervision for four tasks: detecting commitment, identifying a revealed side, anticipating future action, and projecting collective odds. Across 15 LLMs and 18 topic and platform settings, models partially identify position sides (Directed Accuracy 0.506-0.599) but struggle with later tasks: 10 models collapse to one or two action labels, and none consistently beats a naive odds-direction baseline. The abstract also reports no correlation between model scale and performance, no improvement in revealed-side identification from finance-domain tuning, and strong influence from platform incentives on higher-order results. The authors emphasize that observed commitment signals should not be equated with latent belief or causal market impact.

### Source excerpt

arXiv:2605.26074v2 Announce Type: replace-cross Abstract: Existing financial NLP benchmarks often rely on labels supplied by outside observers, measuring how language is perceived rather than what speakers have committed to in the market. We introduce StakeBench, an evaluation framework for language understanding grounded in market commitment. StakeBench links 560,876 comments from 2,261 resolved markets to verified position, action, and market-odds records across Polymarket and Manifold. Supervision is derived from observable market behavior. Position sides, post-comment trading actions, and market-odds trajectories replace human annotation. Four diagnostic tasks test whether models detect market commitment, identify the revealed side, anticipate future action, and perform collective odds projection. Three commitment-aware metrics measure alignment with revealed preferences rather than perceived sentiment. Validity audits and explicit interpretation boundaries help distinguish observable commitment signals from latent belief and causal market-odds impact. Across 15 LLMs and 18 topics and platform settings, models partially recover position-side signals, with Directed Accuracy from 0.506 to 0.599, but show structural failures on later tasks. Ten of the fifteen models collapse to one or two action labels in future action anticipation, and no model consistently improves on the naive odds-direction baseline in collective odds projection. Model scale is not correlated with performance, finance-domain tuning does not improve revealed-side identification, and platform incentives strongly shape higher-order results. StakeBench is packaged with evaluation code and dataset under CC-BY 4.0.

## Understanding and Tackling Label Errors in Individual-Level Nature Language Understanding

DevFeed: [Understanding and Tackling Label Errors in Individual-Level Nature Language Understanding](<https://devfeed.tech/articles/understanding-and-tackling-label-errors-in-individual-level-nature-language-understanding-128062.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2502.13297>)

Author: Yunpeng Xiao, Kyrie Zhixuan Zhou, Youpeng Zhao, Kai Shu

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Data Quality](<https://devfeed.tech/topics/data-quality.md>)

Tags: [ablation](<https://devfeed.tech/tags/ablation.md>), [annotation](<https://devfeed.tech/tags/annotation.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [detection](<https://devfeed.tech/tags/detection.md>), [errors](<https://devfeed.tech/tags/errors.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [large-language-models](<https://devfeed.tech/tags/large-language-models.md>), [llama3](<https://devfeed.tech/tags/llama3.md>), [natural-language](<https://devfeed.tech/tags/natural-language.md>)

### AI overview

For stance detection and topic-based sentiment analysis, annotating a post without its author's other posts can miss individual subjective perspectives and produce mislabeled data. The proposed guideline considers an individual's other posts before assigning labels. Applying it to expand and re-annotate datasets, the authors report sample error rates as high as 31.7% and 23.3%, respectively. On the re-annotated datasets, GPT-4o and Llama3-70B each exceeded 87% accuracy when individual factors were included; ablation studies also supported their usefulness. The evidence here is limited to the supplied abstract and excerpt.

### Source excerpt

arXiv:2502.13297v2 Announce Type: replace-cross Abstract: Natural language understanding (NLU) is a task that enables machines to understand human language. Some tasks, such as stance detection and sentiment analysis, are closely related to individual subjective perspectives, thus termed individual-level NLU. Previously, these tasks are often simplified to text-level NLU tasks, ignoring individual factors. This not only makes inference difficult and unexplainable but often results in a large number of label errors when creating datasets. To address the above limitations, we propose a new NLU annotation guideline based on individual-level factors. Specifically, we incorporate other posts by the same individual and then annotate individual subjective perspectives after considering all individual posts. We use this guideline to expand and re-annotate the stance detection and topic-based sentiment analysis datasets. We find that error rates in the samples were as high as 31.7\% and 23.3\%. We further use large language models to conduct experiments on the re-annotation datasets and find that the large language models perform well on both datasets after adding individual factors. Both GPT-4o and Llama3-70B can achieve an accuracy greater than 87\% on the re-annotation datasets. We also verify the effectiveness of individual factors through ablation studies. We call on future researchers to add individual factors when creating such datasets. Our re-annotation dataset can be found at https://github.com/24yearsoldstudent/Individual-NLU

## Understanding Developer Information Needs for Software with Embedded Natural Language Prompts

DevFeed: [Understanding Developer Information Needs for Software with Embedded Natural Language Prompts](<https://devfeed.tech/articles/understanding-developer-information-needs-for-software-with-embedded-natural-language-prompts-127177.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2507.17264>)

Author: Jenny T. Liang, Chenyang Yang, Agnia Sergeyuk, Travis D. Breaux, Brad A. Myers

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Tool](<https://devfeed.tech/topics/tool.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-hc](<https://devfeed.tech/tags/cs-hc.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [developer](<https://devfeed.tech/tags/developer.md>), [developer-tools](<https://devfeed.tech/tags/developer-tools.md>), [developers](<https://devfeed.tech/tags/developers.md>), [embedded](<https://devfeed.tech/tags/embedded.md>), [foundation-models](<https://devfeed.tech/tags/foundation-models.md>), [information](<https://devfeed.tech/tags/information.md>), [language](<https://devfeed.tech/tags/language.md>), [natural-language](<https://devfeed.tech/tags/natural-language.md>), [needs](<https://devfeed.tech/tags/needs.md>), [software](<https://devfeed.tech/tags/software.md>)

### AI overview

Developers changing prompts embedded in software need information to plan and justify iterative edits, but existing tools leave notable gaps. In an exploratory mixed-method study, researchers developed and applied a list of 50 information needs, drawing on interviews with 16 developers, observations of 8, a survey of 50, and an evaluation of 72 research and commercial tools. They found that tools did not address 16 needs, including four rated among the most important, and identify these gaps as design opportunities for future tools. This overview is based on the supplied abstract.

### Source excerpt

arXiv:2507.17264v2 Announce Type: replace Abstract: Developers use foundation models (FMs) to create novel forms of intelligent software by writing natural language prompts embedded within a software's implementation---referred to as embedded prompts---that are also passed as a string within an FM's API. Developing software with embedded prompts requires developers to engage in iterative embedded prompt changes to obtain the desired behavior. While the information needs of traditional software have been investigated, the information needs of modern developers for embedded prompt changes are underexplored, despite these needs shaping how developers plan and justify embedded prompt changes. In this work, we perform a mixed-method exploratory study on the information needs of developers making changes to embedded prompts. We create, refine, validate, and apply a list of 50 information needs that developers have while making a change to an embedded prompt by interviewing 16 developers, observing 8 developers make prompt changes, surveying 50 developers, and evaluating 72 research and commercial tools. We find that embedded prompts are not well-supported by existing tools: 16 of the 50 information needs, including four of the most important ones, are unaddressed. We conclude with important design opportunities that should be addressed by future developer tools for embedded prompts.

## Technohybridism: Recursive Semantic Normalization in AI-Mediated Societies

DevFeed: [Technohybridism: Recursive Semantic Normalization in AI-Mediated Societies](<https://devfeed.tech/articles/technohybridism-recursive-semantic-normalization-in-ai-mediated-societies-120278.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.07984>)

Author: Charalampos M. Liapis, Epameinondas Panagopoulos

Published: 2026-11-06T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [Electrical engineering and computer science (EECS)](<https://devfeed.tech/topics/electrical-engineering-and-computer-science-eecs.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-systems](<https://devfeed.tech/tags/ai-systems.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [article](<https://devfeed.tech/tags/article.md>), [change](<https://devfeed.tech/tags/change.md>), [collective](<https://devfeed.tech/tags/collective.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cy](<https://devfeed.tech/tags/cs-cy.md>), [cs-hc](<https://devfeed.tech/tags/cs-hc.md>), [diversity](<https://devfeed.tech/tags/diversity.md>), [field](<https://devfeed.tech/tags/field.md>), [generation](<https://devfeed.tech/tags/generation.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [interaction](<https://devfeed.tech/tags/interaction.md>), [language](<https://devfeed.tech/tags/language.md>), [machine](<https://devfeed.tech/tags/machine.md>), [model](<https://devfeed.tech/tags/model.md>), [persistence](<https://devfeed.tech/tags/persistence.md>), [philosophy](<https://devfeed.tech/tags/philosophy.md>), [practices](<https://devfeed.tech/tags/practices.md>), [process](<https://devfeed.tech/tags/process.md>), [provenance](<https://devfeed.tech/tags/provenance.md>), [semantic](<https://devfeed.tech/tags/semantic.md>)

### AI overview

The article argues that AI-generated language can help reproduce or change shared meanings when people and institutions repeatedly select, reuse, and retain it. It calls this process recursive semantic normalization and treats the relevant unit as a hybrid system of human practice, computational generation, and institutional uptake. Its model distinguishes generation, selection, uptake, and persistence, including machine-attributable changes versus changes that would have happened anyway. The abstract identifies risks including obscured provenance, reduced cognitive diversity, and narrower shared interpretive resources, while emphasizing that AI outputs gain no normative authority. It also proposes an evidential ladder for empirical research; details are not included in the excerpt.

### Source excerpt

arXiv:2611.07984v1 Announce Type: cross Abstract: Language-generating AI systems are no longer only objects of linguistic interaction: their outputs increasingly re-enter the communicative and institutional practices from which their training material was drawn. Drawing on social theory, social epistemology, and the philosophy of language, this article develops a sociotechnical account of that re-entry. We introduce recursive semantic normalization: the process through which machine-generated formulations, once recurrently taken up, evaluated, reused, and institutionally retained, participate in the reproduction and transformation of semantic regularities. Under the heading of technohybridism, we argue that the relevant unit of analysis is the hybrid semantic system, a recursively coupled configuration of human practices, computational generation, and institutional uptake. First, we distinguish the concept from neighbouring accounts, including model collapse, looping effects, machine habitus, machine culture, and the epistemology of AI testimony. Second, we develop an analytical model that separates generation, selection, uptake, and persistence and distinguishes machine-attributable change from change that would have occurred anyway. Third, we analyse uptake as routinized, institutionally diffused, and socially stratified practice and state conditions under which machine-derived regularities become norms. Fourth, we identify social-epistemic consequences: the laundering of provenance that weakens epistemic vigilance, the erosion of the cognitive diversity on which collective inquiry depends, and the narrowing of shared hermeneutical resources. An evidential ladder for empirical research follows. The account attributes no normative authority to machine outputs; what changes is the field within which human and institutional uptake operates.

## Set-Valued Routing for Language-Model Agent Systems: A WildChat Benchmark and Cost-Aware Evaluation

DevFeed: [Set-Valued Routing for Language-Model Agent Systems: A WildChat Benchmark and Cost-Aware Evaluation](<https://devfeed.tech/articles/set-valued-routing-for-language-model-agent-systems-a-wildchat-benchmark-and-cost-aware-evaluation-120059.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2606.28925>)

Author: Ananto Nayan Bala, Faisal Muhammad Shah

Published: 2026-11-06T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ir](<https://devfeed.tech/tags/cs-ir.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-ma](<https://devfeed.tech/tags/cs-ma.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [routing](<https://devfeed.tech/tags/routing.md>), [set](<https://devfeed.tech/tags/set.md>), [systems](<https://devfeed.tech/tags/systems.md>)

### AI overview

The paper treats routing natural-language requests to a fixed catalog of specialized agents as a set-prediction problem, since compositional requests may need multiple capabilities and extra dispatch adds cost. Its benchmark contains 3,000 WildChat prompts and 12 capability-defined agents, with schema-guided reference sets balanced by agent frequency and route size. It compares profile retrieval, multilabel classifiers, a fine-tuned text encoder, and catalog-conditioned zero-shot GPT-4o using set quality, latency, capability coverage, and ordinal-cost utility. The encoder scores 89.59 F1, versus 71.79 for linear SVM and 41.51 for GPT-4o, and reaches 88.02 F1 on a harder semantic-group split. Weighted Agent Routing (WAR) is a deterministic post-scoring rule that adjusts the quality-cost operating point without retraining; Encoder+WAR achieves the best constrained utility. The reported results favor catalog-specific supervision and explicit set construction, with cost-aware decision rules complementing the encoder.

### Source excerpt

arXiv:2606.28925v3 Announce Type: replace-cross Abstract: Language-model agent systems must map natural-language requests to one or more specialized agents before planning or tool execution. This decision is often treated as top-1 selection, although compositional requests can require several capabilities and unnecessary dispatch raises cost. We formulate fixed-catalog routing as set-valued language understanding and construct a benchmark of 3,000 WildChat prompts over 12 capability-defined agents. Reproducible schema-guided reference sets are balanced by agent frequency and route cardinality. We compare profile retrieval, multilabel classifiers, a fine-tuned text encoder, and catalog-conditioned zero-shot GPT-4o. Evaluation combines set quality, latency, capability coverage, and ordinal-cost utility. The encoder attains 89.59 F1, compared with 71.79 for linear SVM and 41.51 for GPT-4o, and retains 88.02 F1 under a harder semantic-group split. We further introduce Weighted Agent Routing (WAR), a deterministic post-scoring rule that changes the quality-cost operating point without retraining; Encoder+WAR obtains the best constrained utility. The results show that catalog-specific supervision and explicit set construction are important for multi-agent dispatch, while cost-aware decision layers can complement strong language encoders.

## ELF: Embedded Language Flows

DevFeed: [ELF: Embedded Language Flows](<https://devfeed.tech/articles/elf-embedded-language-flows-120024.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2605.10938>)

Author: Keya Hu, Linlu Qiu, Yiyang Lu, Hanhong Zhao, Tianhong Li, Yoon Kim, Jacob Andreas, Kaiming He

Published: 2026-11-06T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>)

Tags: [continuous](<https://devfeed.tech/tags/continuous.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [elf](<https://devfeed.tech/tags/elf.md>), [embedding](<https://devfeed.tech/tags/embedding.md>), [flow-matching](<https://devfeed.tech/tags/flow-matching.md>), [guidance](<https://devfeed.tech/tags/guidance.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [modeling](<https://devfeed.tech/tags/modeling.md>), [quality](<https://devfeed.tech/tags/quality.md>), [sampling](<https://devfeed.tech/tags/sampling.md>)

### AI overview

ELF applies continuous-time Flow Matching to language in embedding space, remaining continuous until the final step, when a shared-weight network maps embeddings to discrete tokens. The authors say this requires minimal adaptation to the discrete domain and makes image-domain methods such as classifier-free guidance straightforward to use. In experiments, ELF reportedly outperforms leading discrete and continuous diffusion language models in generation quality while using fewer sampling steps. The supplied text does not provide experimental details or metrics.

### Source excerpt

arXiv:2605.10938v3 Announce Type: replace-cross Abstract: Diffusion and flow-based models have become the de facto approaches for generating continuous data, e.g., in domains such as images and videos. Their success has attracted growing interest in applying them to language modeling. Unlike their image-domain counterparts, today's leading diffusion language models (DLMs) primarily operate over discrete tokens. In this paper, we show that continuous DLMs can be made effective with minimal adaptation to the discrete domain. We propose Embedded Language Flows (ELF), a class of diffusion models in continuous embedding space based on continuous-time Flow Matching. Unlike existing DLMs, ELF predominantly stays within the continuous embedding space until the final time step, where it maps to discrete tokens using a shared-weight network. This formulation makes it straightforward to adapt established techniques from image-domain diffusion models, e.g., classifier-free guidance (CFG). Experiments show that ELF substantially outperforms leading discrete and continuous DLMs, achieving better generation quality with fewer sampling steps. These results suggest that ELF offers a promising path toward effective continuous DLMs.

## Perturb, Repair, Verify: Self-Play Vision-Language Verifiers for Compositional Understanding

DevFeed: [Perturb, Repair, Verify: Self-Play Vision-Language Verifiers for Compositional Understanding](<https://devfeed.tech/articles/perturb-repair-verify-self-play-vision-language-verifiers-for-compositional-understanding-117025.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.05158>)

Author: Zijun Huang, Yingjun Du, Cees G. M. Snoek

Published: 2026-11-04T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [vlm](<https://devfeed.tech/topics/vlm.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [also](<https://devfeed.tech/tags/also.md>), [annotation](<https://devfeed.tech/tags/annotation.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [behaviors](<https://devfeed.tech/tags/behaviors.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [bounded](<https://devfeed.tech/tags/bounded.md>), [changes](<https://devfeed.tech/tags/changes.md>), [counts](<https://devfeed.tech/tags/counts.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [existing](<https://devfeed.tech/tags/existing.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [external](<https://devfeed.tech/tags/external.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [language](<https://devfeed.tech/tags/language.md>), [play](<https://devfeed.tech/tags/play.md>), [repair](<https://devfeed.tech/tags/repair.md>), [verify](<https://devfeed.tech/tags/verify.md>), [vision](<https://devfeed.tech/tags/vision.md>)

### AI overview

PROVE trains vision-language verifiers through a self-play loop using a known-correct image-caption pair. A Mutator changes objects, attributes, counts, or relations; the verifier judges the altered caption and gives textual feedback; then a Repairer attempts a visually faithful correction for re-verification. Rewards reflect informative mismatches and grounded repairs. Because all three roles use the same evolving multimodal backbone, shared updates also improve verification. The abstract reports stronger fine-grained compositional verification across model scales and families, and better visual-detail robustness on the reference backbone. Effects on broader multimodal tasks are limited and benchmark-dependent.

### Source excerpt

arXiv:2611.05158v1 Announce Type: cross Abstract: Vision-language verifiers judge whether an image faithfully matches a textual description, yet existing training pipelines typically rely on static supervision from stronger external judges or curated hard-negative data. Such supervision requires external generation or annotation, is teacher-bounded, and is difficult to adapt as model failures evolve. We introduce PROVE (Perturb-and-Repair Optimization for Vision-Language Evaluation), a self-play framework that trains verifiers through a closed perturb-verify-repair-verify loop. Starting from a known-correct image-caption pair, a Mutator introduces a typed perturbation over objects, attributes, counts, or relations. The verifier then judges the perturbed caption and provides textual feedback, which a Repairer uses to recover a faithful caption for re-verification. These two verification rounds produce structured rewards for informative mismatches and visually grounded repairs. Since perturbation, verification, and repair are role-conditioned behaviors of the same evolving multimodal backbone, updating the shared parameters also improves the verifier itself. Experiments show that PROVE improves fine-grained compositional verification across model scales and families and strengthens visual-detail robustness on the reference backbone, with limited, benchmark-dependent changes on broader multimodal tasks.

## Introducing India-based inference for Kimi K3 on Amazon Bedrock

DevFeed: [Introducing India-based inference for Kimi K3 on Amazon Bedrock](<https://devfeed.tech/articles/introducing-india-based-inference-for-kimi-k3-on-amazon-bedrock-115500.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/introducing-india-based-inference-for-kimi-k3-on-amazon-bedrock/>)

Author: Naomi Teng

Published: 2026-11-03T16:32:54Z

Content type: tutorial

Language: en

Sources: [Artificial Intelligence](<https://devfeed.tech/sources/artificial-intelligence.md>)

Topics: [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [amazon](<https://devfeed.tech/topics/amazon.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [Image](<https://devfeed.tech/topics/image.md>), [Amazon CloudWatch](<https://devfeed.tech/topics/amazon-cloudwatch.md>)

Tags: [amazon](<https://devfeed.tech/tags/amazon.md>), [amazon-bedrock](<https://devfeed.tech/tags/amazon-bedrock.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [based](<https://devfeed.tech/tags/based.md>), [bedrock](<https://devfeed.tech/tags/bedrock.md>), [code](<https://devfeed.tech/tags/code.md>), [country](<https://devfeed.tech/tags/country.md>), [cross](<https://devfeed.tech/tags/cross.md>), [example](<https://devfeed.tech/tags/example.md>), [global](<https://devfeed.tech/tags/global.md>), [india](<https://devfeed.tech/tags/india.md>), [indic](<https://devfeed.tech/tags/indic.md>), [inference](<https://devfeed.tech/tags/inference.md>), [intermediate-200](<https://devfeed.tech/tags/intermediate-200.md>), [k3](<https://devfeed.tech/tags/k3.md>), [kimi](<https://devfeed.tech/tags/kimi.md>), [kimi-k3](<https://devfeed.tech/tags/kimi-k3.md>), [language](<https://devfeed.tech/tags/language.md>), [mumbai](<https://devfeed.tech/tags/mumbai.md>), [offers](<https://devfeed.tech/tags/offers.md>), [options](<https://devfeed.tech/tags/options.md>), [post](<https://devfeed.tech/tags/post.md>)

### AI overview

Amazon Bedrock's `in.moonshotai.kimi-k3` inference profile routes Kimi K3 processing to India Regions, supporting workloads with India data residency requirements. The excerpt describes invoking it through Mumbai and using the model's image input to read Indic-language documents and answer Hinglish questions without a separate OCR or translation service. Example workflows can extract schema-defined JSON, normalize dates and digits, transliterate names, summarize documents, and handle follow-up questions. The article recommends high image detail when small text matters, with increased input-token use, and advises validating extraction results for the intended use. It also covers cross-Region routing, optional prompt caching and request service tiers, and monitoring; invocation logs and metrics are stored in the source Region.

### Source excerpt

Amazon Bedrock now offers in-country inference in India for Kimi K3, keeping processing within the Mumbai and Hyderabad Regions. This post explains the global and geographic cross-Region inference options and shares example code for reading and understanding Indic-language documents with Kimi K3.

## Spatial Transfusion: From Structured Spatial Scene Understanding to Generation and Editing

DevFeed: [Spatial Transfusion: From Structured Spatial Scene Understanding to Generation and Editing](<https://devfeed.tech/articles/spatial-transfusion-from-structured-spatial-scene-understanding-to-generation-and-editing-115295.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.02330>)

Author: Jaeyeon Kim, Sebastian Braun, Hannes Gamper

Published: 2026-11-03T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Machine Perception](<https://devfeed.tech/topics/machine-perception.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [lora](<https://devfeed.tech/topics/lora.md>), [Enhancing AI model performance](<https://devfeed.tech/topics/enhancing-ai-model-performance.md>)

Tags: [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-sd](<https://devfeed.tech/tags/cs-sd.md>), [eess-as](<https://devfeed.tech/tags/eess-as.md>), [flow-matching](<https://devfeed.tech/tags/flow-matching.md>), [generation](<https://devfeed.tech/tags/generation.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [natural-language](<https://devfeed.tech/tags/natural-language.md>), [spatial](<https://devfeed.tech/tags/spatial.md>), [structured](<https://devfeed.tech/tags/structured.md>)

### AI overview

Spatial Transfusion combines spatial audio understanding, generation, and editing in a first-order ambisonics framework. It extends a pretrained understanding model using flow matching over VAE latents, while keeping the backbone frozen and routing modality-specific LoRA adapters. Editing is represented as structured scene transformations drawn from a taxonomy of atomic edits; a curriculum progressively introduces generation, editing, and natural-language interaction. The abstract reports that the approach largely preserves spatial understanding and outperforms full fine-tuning on generation and editing. It supports controllable generation and editing through structured or natural-language interfaces.

### Source excerpt

arXiv:2611.02330v1 Announce Type: cross Abstract: Spatial audio conveys not only what sounds are present, but where they originate and how they interact with the environment. Recent audio-language models increasingly understand and reason about such scenes, yet spatial audio understanding, generation, and editing remain largely separate. We present Spatial Transfusion, a unified first-order ambisonics (FOA) framework that extends a pretrained spatial audio understanding model to generation and editing through flow matching over VAE latents. We formulate editing as a structured scene transformation over a taxonomy of atomic edits. Keeping the pretrained backbone frozen, we introduce modality-based LoRA routing and a curriculum that progressively adds generation, editing, and natural-language interaction. Experiments show that Spatial Transfusion largely preserves spatial understanding while outperforming full fine-tuning in generation and editing. The resulting model supports controllable FOA generation and editing through both structured and natural-language interfaces. Audio samples are available at https://microsoft.github.io/spatial-transfusion/

## Maswao: An automatically collected, prepared, and benchmarked dataset for the South African Sign language

DevFeed: [Maswao: An automatically collected, prepared, and benchmarked dataset for the South African Sign language](<https://devfeed.tech/articles/maswao-an-automatically-collected-prepared-and-benchmarked-dataset-for-the-south-african-sign-language-115398.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.02903>)

Author: Thapelo Sindane, Unarine Netshifhefhe, Vukosi Marivate, Abiodun Modupe, Herkulaas Combrink, Ketro Sithole

Published: 2026-11-03T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [machine-translation](<https://devfeed.tech/topics/machine-translation.md>), [asr](<https://devfeed.tech/topics/asr.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [managing large-scale text data in AI](<https://devfeed.tech/topics/managing-large-scale-text-data-in-ai.md>)

Tags: [asr](<https://devfeed.tech/tags/asr.md>), [audio](<https://devfeed.tech/tags/audio.md>), [automatically](<https://devfeed.tech/tags/automatically.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [bleu](<https://devfeed.tech/tags/bleu.md>), [collect](<https://devfeed.tech/tags/collect.md>), [collection](<https://devfeed.tech/tags/collection.md>), [crop](<https://devfeed.tech/tags/crop.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gradient](<https://devfeed.tech/tags/gradient.md>), [gradient-descent](<https://devfeed.tech/tags/gradient-descent.md>), [labour](<https://devfeed.tech/tags/labour.md>), [language](<https://devfeed.tech/tags/language.md>), [languages](<https://devfeed.tech/tags/languages.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [motion](<https://devfeed.tech/tags/motion.md>), [sign](<https://devfeed.tech/tags/sign.md>)

### AI overview

Maswao is a weakly sentence-aligned South African Sign Language dataset built with minimal human intervention. The framework collects public SASL videos, including news broadcasts, converts accompanying audio to text with available ASR tools, crops the signing area, and uses gradient descent to detect non-motion valleys as sentence boundaries. On sign-to-text translation, Transformer models outperformed the recurrent baselines tested. The best Transformer scored BLEU-1 through BLEU-4 of 13.78, 5.21, 2.52, and 1.54, with ChrF 18.09 and ROUGE-L 10.51. The reported scores suggest the task remains open for further investigation.

### Source excerpt

arXiv:2611.02903v1 Announce Type: cross Abstract: Recent advancements in sign language processing can be attributed to developments in both large-scale dataset collection and algorithmic complexity in areas such as pose estimation techniques, signal processing through Transformer models, and model optimization. However, only a handful of sign languages, such as the American, British, and German, from over 200 globally recognized sign languages, reap the benefits of such advancements, while the rest lag behind. In this study, we aim to address this technological disparity by proposing a framework for the automatic collection and preparation of sign language datasets, with very minimal human-labour intervention, using the South African Sign Language (SASL) as a testbed. Specifically, we automatically collect SASL video content using public resources such as news broadcasts. Additionally, we convert their accompanying audio streams into text using available Automatic Speech Recognition (ASR) tools. Furthermore, we crop the sign language area in the videos and use gradient descent to locate valleys of non-motion to determine end-of-sentence points within long videos to create our weakly sentence-aligned parallel corpora -- the Maswao (Signs) dataset. Finally, we benchmark the weakly aligned dataset on the task of Sign-to-text Translation using Recurrent Neural Networks and Transformer models to compare their individual intricacies. Our results show that Transformer models proposed in the literature outperform recurrent baselines, with the best-performing Transformer achieving BLEU-1 to BLEU-4 scores of 13.78, 5.21, 2.52, and 1.54, respectively, as well as ChrF = 18.09 and ROUGE-L = 10.51. These results indicate that the task remains promising and warrants further investigations.

## OntoDoc: Automatically Generating Ontology Resource Descriptions

DevFeed: [OntoDoc: Automatically Generating Ontology Resource Descriptions](<https://devfeed.tech/articles/ontodoc-automatically-generating-ontology-resource-descriptions-115147.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.01276>)

Author: Juan Guajardo Gutierrez, Mike Tran, Nasim Shirvani Mahdavi, Chengkai Li

Published: 2026-11-03T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [RDF (Resource Description Framework)](<https://devfeed.tech/topics/rdf.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>)

Tags: [agent-framework](<https://devfeed.tech/tags/agent-framework.md>), [all](<https://devfeed.tech/tags/all.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [biomedical](<https://devfeed.tech/tags/biomedical.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ir](<https://devfeed.tech/tags/cs-ir.md>), [database](<https://devfeed.tech/tags/database.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [framework](<https://devfeed.tech/tags/framework.md>), [intermediate](<https://devfeed.tech/tags/intermediate.md>), [knowledge-graph](<https://devfeed.tech/tags/knowledge-graph.md>), [knowledge-graphs](<https://devfeed.tech/tags/knowledge-graphs.md>), [large-language-models-llms](<https://devfeed.tech/tags/large-language-models-llms.md>), [llms](<https://devfeed.tech/tags/llms.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [natural-language](<https://devfeed.tech/tags/natural-language.md>), [ontology](<https://devfeed.tech/tags/ontology.md>), [open-web](<https://devfeed.tech/tags/open-web.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [structured](<https://devfeed.tech/tags/structured.md>), [web](<https://devfeed.tech/tags/web.md>), [wikipedia](<https://devfeed.tech/tags/wikipedia.md>)

### AI overview

OntoDoc generates natural-language descriptions for IRIs in a SPARQL endpoint or RDF file by combining graph evidence with web context. Its four-stage pipeline explores the graph for informative triples, retrieves and re-ranks background material from Wikipedia and the open web, turns both sources into evidence-grounded assumptions, then generates a description from those assumptions. Evaluated on three biomedical, agricultural, and cross-domain knowledge graphs with LLM scoring and 11 human annotators, it matched or exceeded a naive LLM baseline and a template-based verbalizer in accuracy. It improved completeness by an average of 12% over the naive baseline and by up to 2.8 points on a five-point scale over the template-based verbalizer.

### Source excerpt

arXiv:2611.01276v1 Announce Type: new Abstract: Knowledge graphs and ontologies form the backbone of structured, machine-readable knowledge representation, yet ontology terms often lack natural language descriptions, limiting their accessibility and reuse. Existing ontology verbalization approaches tend to rely on rule-based or template-driven techniques that operate on strict, formal schema axioms, resulting in rigid verbalizations that fail to convey richer contextual semantics. We present OntoDoc, a multi-agent framework for automatically generating natural language descriptions for any Internationalized Resource Identifier (IRI) within a SPARQL endpoint or RDF file. OntoDoc operates through a four-stage pipeline: (1) a Database Context agent that adaptively explores the knowledge graph via SPARQL to collect semantically informative triples about the target IRI; (2) a Web Context agent that retrieves and re-ranks encyclopedic background knowledge from Wikipedia and the open web; (3) an Assumption Generation agent that synthesizes both contexts into a structured, evidence-grounded list of intermediate assumptions; and (4) a Description Generation agent that produces the final natural language description from the grounded assumptions. We evaluate OntoDoc on three knowledge graphs spanning biomedical, agricultural, and cross-domain settings, using both automatic scoring via Large Language Models (LLMs) and human evaluation with 11 annotators. OntoDoc consistently matches or exceeds a naive LLM baseline and a template-based ontology verbalizer in accuracy. Furthermore, it vastly outperforms both baselines in the completeness of the generated descriptions, delivering an average 12% improvement in completeness over the naive baseline across all domains, and improving completeness scores by as much as 2.8 points on a 5-point scale relative to the ontology verbalizer.

## Optimization Techniques for Solving the Minimum Red--Blue Separating Set Problem with an Application to Human--AI Text Separation

DevFeed: [Optimization Techniques for Solving the Minimum Red--Blue Separating Set Problem with an Application to Human--AI Text Separation](<https://devfeed.tech/articles/optimization-techniques-for-solving-the-minimum-red-blue-separating-set-problem-with-an-application-to-human-ai-text-separation-112019.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.41081>)

Author: Marko Djukanovi\\'c, Sta\\v{s}a Vuji\\v{c}i\\'c Stankovi\\'c, \\v{Z}iga Zebec, Christian Blum

Published: 2026-11-02T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Combinatorial optimization](<https://devfeed.tech/topics/combinatorial-optimization.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [application](<https://devfeed.tech/tags/application.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [features](<https://devfeed.tech/tags/features.md>), [gap](<https://devfeed.tech/tags/gap.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [interpretability](<https://devfeed.tech/tags/interpretability.md>), [logic](<https://devfeed.tech/tags/logic.md>), [lower-bounds](<https://devfeed.tech/tags/lower-bounds.md>), [mandatory](<https://devfeed.tech/tags/mandatory.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [random-graphs](<https://devfeed.tech/tags/random-graphs.md>), [search](<https://devfeed.tech/tags/search.md>), [set-cover](<https://devfeed.tech/tags/set-cover.md>), [strategies](<https://devfeed.tech/tags/strategies.md>), [text](<https://devfeed.tech/tags/text.md>)

### AI overview

The study compares several approaches to the Minimum Red-Blue Separating Set Problem: a Set-Cover ILP, safe preprocessing, GRASP, Logic-Based Benders Decomposition (LBBD), and Branch-and-Benders-Cut (BBC). Across 250 random graphs of up to 500 vertices, ILP found the best feasible solutions, LBBD gave the strongest lower bounds and smallest mean optimality gap, and BBC certified optimality most often. In a human-AI text separation application using vocabulary-based graph features, BBC had the smallest mean optimality gap and certified 16 optima across 42 instances; GRASP outperformed VNS. Analysis of 272 tests found 113 mandatory and 151 replaceable vocabulary features, indicating that optimal separators varied.

### Source excerpt

arXiv:2610.41081v1 Announce Type: new Abstract: We study the so-called Minimum Red''Blue Separating Set Problem (MRBSSP) which arises from computational geometry. In particular, we propose a compact Set-Cover ILP formulation, safe preprocessing rules, a Greedy Randomized Adaptive Search Procedure (GRASP), and two exact approaches: one is based on Logic-Based Benders Decomposition (LBBD) approach, and the second represents a novel Branch-and-Benders-Cut (BBC) approach. These two exact approaches are compared to ILP--alongside GRASP and the Variable Neighborhood Search (VNS) from literature. Experiments on 250 random graphs with up to 500 vertices reveal complementary strengths: ILP yields the best feasible solutions, LBBD provides the strongest lower bounds and smallest mean optimality gap, and BBC certifies optimality most frequently. We further introduce an NLP application distinguishing human- and AI-generated texts through vocabulary-based graph features. On 42 instances, BBC achieves the smallest mean optimality gap, certifying 16 optima, while GRASP outperforms VNS. Structural analysis of optimal vocabulary separators identifies 113 mandatory and 151 replaceable features across 272 tests, highlighting substantial variability among optimal solutions. Overall, our results demonstrate the effectiveness of complementary optimization strategies and the interpretability of optimal vocabulary-based discrimination.

## "It Would Make a Geordie Feel at Home": Regional Accent Perceptions and Speech Accommodation in Voice AI

DevFeed: ["It Would Make a Geordie Feel at Home": Regional Accent Perceptions and Speech Accommodation in Voice AI](<https://devfeed.tech/articles/it-would-make-a-geordie-feel-at-home-regional-accent-perceptions-and-speech-accommodation-in-voice-ai-108002.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.38005>)

Author: Dana Serditova, Kevin Tang, Jochen Steffens

Published: 2026-10-30T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [User Experience](<https://devfeed.tech/topics/user-experience.md>), [human computer interaction](<https://devfeed.tech/topics/human-computer-interaction.md>), [Conversational AI](<https://devfeed.tech/topics/conversational-ai.md>), [asr](<https://devfeed.tech/topics/asr.md>), [Speech Processing](<https://devfeed.tech/topics/speech-processing.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>)

Tags: [adapting](<https://devfeed.tech/tags/adapting.md>), [ai](<https://devfeed.tech/tags/ai.md>), [asr](<https://devfeed.tech/tags/asr.md>), [bias](<https://devfeed.tech/tags/bias.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-hc](<https://devfeed.tech/tags/cs-hc.md>), [human-ai-interaction](<https://devfeed.tech/tags/human-ai-interaction.md>), [inclusive](<https://devfeed.tech/tags/inclusive.md>), [speech](<https://devfeed.tech/tags/speech.md>), [speech-recognition](<https://devfeed.tech/tags/speech-recognition.md>), [synthetic-voices](<https://devfeed.tech/tags/synthetic-voices.md>), [voice](<https://devfeed.tech/tags/voice.md>), [voice-ai](<https://devfeed.tech/tags/voice-ai.md>)

### AI overview

In interviews following interactions with a simulated voice assistant, Newcastle English speakers described both benefits and tradeoffs in how accents are heard. The Geordie voice conveyed familiarity, warmth, personality and local belonging, while raising concerns about professionalism and authority; the SSBE voice seemed more neutral, clear and credible. When voice technologies failed to recognize them, participants said they often shifted toward a mainstream accent, enunciated more clearly or spoke more slowly. The findings suggest bias can surface both in users' perceived need to accommodate technology and in social stereotypes attached to synthetic voices. The study focuses on Newcastle English speakers and a simulated assistant.

### Source excerpt

arXiv:2610.38005v1 Announce Type: cross Abstract: Automatic speech recognition (ASR) systems often perform less accurately for speakers of regional dialects, but less is known about how users respond to these limitations and how regional voices produced by voice assistants (VAs) are evaluated through a sociolinguistic lens. This study investigates these issues among speakers of Newcastle English (Geordies), examining both linguistic accommodation to voice technologies and perceptions of regional versus mainstream-accented VA voices. Participants interacted with a simulated VA that spoke either Newcastle English or Standard Southern British English (SSBE). They subsequently completed semi-structured interviews about their experiences with the experimental VA and voice technologies, language use, accent attitudes, and place orientation. The findings reveal a two-way process of sociolinguistic bias. The Geordie VA was associated with familiarity, warmth, personality, and local belonging, but also with concerns about professionalism and authority, whereas the SSBE voice was perceived as more neutral, clear, and credible. Participants also reported adapting their speech when voice technologies failed to recognise them, most commonly by shifting to a more mainstream accent, enunciating more clearly, and speaking more slowly. These strategies are similar to accommodation in human-human interaction and reflect users' perceptions of what technologies can recognise, also known as perceived intelligibility pressures. Overall, the findings demonstrate that voice technologies can reproduce existing accent biases both by encouraging users to suppress regional features and by activating established social stereotypes through synthetic voices. The study positions sociolinguistics as a bridge between understanding human-AI interaction and designing inclusive technologies around users' own linguistic priorities.

## Patronizing and Condescending Language Detection in English and Mandarin

DevFeed: [Patronizing and Condescending Language Detection in English and Mandarin](<https://devfeed.tech/articles/patronizing-and-condescending-language-detection-in-english-and-mandarin-99899.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.31808>)

Author: Xinyi Vicky Xiang, Catherine Qingyun Wang, Paula Cort\\'es Lemos

Published: 2026-10-26T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [bert](<https://devfeed.tech/topics/bert.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>)

Tags: [bert](<https://devfeed.tech/tags/bert.md>), [classification](<https://devfeed.tech/tags/classification.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [detection](<https://devfeed.tech/tags/detection.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [english](<https://devfeed.tech/tags/english.md>), [language](<https://devfeed.tech/tags/language.md>), [neural-network](<https://devfeed.tech/tags/neural-network.md>)

### AI overview

The paper proposes automated detection of patronizing and condescending language (PCL), which can belittle vulnerable groups and reinforce stereotypes, yet is difficult to identify even for human judges. It compares a classifier using static word embeddings and a convolutional neural network with one using pre-trained BERT embeddings fine-tuned for PCL. The BERT approach performed considerably better, reaching an F1 score of 0.528 on the development set.

### Source excerpt

arXiv:2610.31808v1 Announce Type: new Abstract: Patronizing and condescending language (PCL) is a form of harmful language that belittles and assumes pity for vulnerable groups, reinforcing negative societal stereotypes. However, unlike overt hate speech, it is often difficult to detect, even for human judges. This paper proposes a system for the automated detection of PCL. Our first approach uses static word embeddings and a convolutional neural network for classification, while our second approach uses pre-trained BERT embeddings fine-tuned for the PCL task. Our second approach was considerably more successful, and achieved an F1 score of 0.528 over the development set.

## CAMGen: A Core Change-Aware Multi-View Framework for Commit Message Generation

DevFeed: [CAMGen: A Core Change-Aware Multi-View Framework for Commit Message Generation](<https://devfeed.tech/articles/camgen-a-core-change-aware-multi-view-framework-for-commit-message-generation-96666.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.31552>)

Author: Bo Xiong, Weisheng Chen, Zongen Ren, Yifeng Zhu, Xutian Li, Runbang Yan, Chong Wang, Peng Liang

Published: 2026-10-23T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [Natural language processing](<https://devfeed.tech/topics/nlp.md>)

Tags: [commit](<https://devfeed.tech/tags/commit.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [dependency](<https://devfeed.tech/tags/dependency.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [framework](<https://devfeed.tech/tags/framework.md>), [functional](<https://devfeed.tech/tags/functional.md>), [generation](<https://devfeed.tech/tags/generation.md>), [impact](<https://devfeed.tech/tags/impact.md>), [languages](<https://devfeed.tech/tags/languages.md>), [large-language-models-llms](<https://devfeed.tech/tags/large-language-models-llms.md>), [repository](<https://devfeed.tech/tags/repository.md>), [view](<https://devfeed.tech/tags/view.md>)

### AI overview

CAMGen aims to generate commit messages that explain how local edits affect functionality implemented across multiple code entities. It identifies core entities representing the change intent, groups repository entities into semantic functional units, and uses units containing the core entities to form a Functional Semantic View. The framework combines that view with structural dependency and historical exemplar views, providing all three alongside the code diff to an LLM. The abstract reports that CAMGen outperformed representative baselines across three programming languages; human evaluation found its messages more complete and clearer about rationale and impact. The excerpt does not provide evaluation details or quantitative results.

### Source excerpt

arXiv:2610.31552v1 Announce Type: new Abstract: Commit messages help developers understand code changes and support software maintenance. Recent approaches use repository context to improve commit message generation (CMG) with large language models (LLMs). However, existing CMG methods provide limited support for understanding how local edits affect functionality jointly implemented by multiple code entities. This functional understanding is essential for generating commit messages that explain the main change and its impact. To address this problem, we propose CAMGen, a core change-aware multi-view commit message generation framework that incorporates functional context into generation. CAMGen identifies core entities that represent the primary change intent. It then forms candidate functional units through repository-wide semantic clustering and uses those containing the core entities to construct a Functional Semantic View. This view supplies semantic context for interpreting the affected functionality. CAMGen also uses context provided by the Structural Dependency View and the Historical Exemplar View. The context from all three views is then provided to the LLM together with the code diff to guide CMG. The experimental results show that CAMGen outperforms representative CMG baselines across three programming languages. Human evaluation further shows that CAMGen generates more complete commit messages with clearer explanations of change rationale and impact. These results demonstrate the value of functional context and complementary repository information for CMG.

## EEG-Captioner: Grounding EEG-to-Language Generation in Fundamental Signal Properties

DevFeed: [EEG-Captioner: Grounding EEG-to-Language Generation in Fundamental Signal Properties](<https://devfeed.tech/articles/eeg-captioner-grounding-eeg-to-language-generation-in-fundamental-signal-properties-97011.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.30995>)

Author: Leyuan Jin, Jiahao Mei, Yixuan Yuan, Shimeng Li, Mengyue Wu, Yue Ding

Published: 2026-10-23T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [also](<https://devfeed.tech/tags/also.md>), [approach](<https://devfeed.tech/tags/approach.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [comparison](<https://devfeed.tech/tags/comparison.md>), [complex](<https://devfeed.tech/tags/complex.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [domain](<https://devfeed.tech/tags/domain.md>), [eeg](<https://devfeed.tech/tags/eeg.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [feature](<https://devfeed.tech/tags/feature.md>), [framework](<https://devfeed.tech/tags/framework.md>), [generation](<https://devfeed.tech/tags/generation.md>), [grounding](<https://devfeed.tech/tags/grounding.md>), [language](<https://devfeed.tech/tags/language.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [lateral](<https://devfeed.tech/tags/lateral.md>), [properties](<https://devfeed.tech/tags/properties.md>), [signal](<https://devfeed.tech/tags/signal.md>)

### AI overview

EEG-Captioner evaluates whether EEG-to-language models capture measurable signal properties, rather than merely generating plausible text. The authors create a dataset of 160,781 EEG segments, each with question-answer pairs covering five property families and a feature-grounded caption. Their Dual-Domain EEG Encoder combines temporal waveforms with spectral-spatial representations; the unified framework trains on both signal-level QA and caption supervision. Among compared encoders, the proposed encoder has the highest overall macro-averaged balanced accuracy. EEG-Captioner reaches 88.81% Macro-BA on signal-level QA and raises caption factual consistency from 82.06% with caption-only language adaptation to 83.46%. The abstract reports these results as evidence that QA and caption supervision provide complementary benefits.

### Source excerpt

arXiv:2610.30995v1 Announce Type: new Abstract: Electroencephalography (EEG)-to-language models offer a promising approach for translating complex neural signals into human-readable descriptions. However, generating plausible language does not necessarily demonstrate that a model has learned to interpret the underlying EEG signal. A direct way to assess such interpretation is to probe objectively measurable properties of the signal itself. We therefore construct an EEG-language dataset of 160,781 segments annotated with signal-level question answering (QA) pairs, where each question targets one of five property families and requires a quantitative comparison involving spectral, spatial, temporal, or lateral signal properties. Each segment is also paired with a feature-grounded EEG caption. We further propose a Dual-Domain EEG Encoder that jointly models temporal waveforms and spectral-spatial representations, and introduce EEG-Captioner, a unified EEG-to-language framework trained with both signal-level QA and EEG caption supervision. Experiments show that the proposed encoder achieves the highest overall macro-averaged balanced accuracy (Macro-BA) among the compared EEG encoders, while EEG-Captioner achieves a Macro-BA of 88.81% on signal-level QA and improves caption factual consistency from 82.06% to 83.46% over caption-only language adaptation, demonstrating the complementary benefits of signal-level QA and caption supervision.

## Stack Trace-Based Crash Deduplication with Transformer Adaptation

DevFeed: [Stack Trace-Based Crash Deduplication with Transformer Adaptation](<https://devfeed.tech/articles/stack-trace-based-crash-deduplication-with-transformer-adaptation-94595.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2508.19449>)

Author: Md Afif Al Mamun, Gias Uddin, Lan Xia, Longyu Zhang

Published: 2026-10-22T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>), [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Transformer](<https://devfeed.tech/topics/transformer.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [Language Engineering](<https://devfeed.tech/topics/language-engineering.md>)

Tags: [crash](<https://devfeed.tech/tags/crash.md>), [crash-reporting](<https://devfeed.tech/tags/crash-reporting.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [deduplication](<https://devfeed.tech/tags/deduplication.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [fully-connected](<https://devfeed.tech/tags/fully-connected.md>), [information-retrieval](<https://devfeed.tech/tags/information-retrieval.md>), [issue-tracking](<https://devfeed.tech/tags/issue-tracking.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [natural-language-processing](<https://devfeed.tech/tags/natural-language-processing.md>), [nlp](<https://devfeed.tech/tags/nlp.md>), [rank](<https://devfeed.tech/tags/rank.md>), [ranking](<https://devfeed.tech/tags/ranking.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>), [stack](<https://devfeed.tech/tags/stack.md>), [string](<https://devfeed.tech/tags/string.md>), [trace](<https://devfeed.tech/tags/trace.md>), [traces](<https://devfeed.tech/tags/traces.md>), [transformer](<https://devfeed.tech/tags/transformer.md>), [triage](<https://devfeed.tech/tags/triage.md>), [unique](<https://devfeed.tech/tags/unique.md>)

### AI overview

dedupT adapts a pretrained language model to stack traces, then trains a fully connected network on its embeddings to rank duplicate crashes. The abstract reports that across four public datasets, it often improves Mean Reciprocal Rank by more than 15% over the best deep learning baseline and by up to 10% over traditional methods, while also achieving higher ROC-AUC for identifying unique crashes. The evidence is limited to the supplied abstract; it does not describe implementation details or dataset-specific results.

### Source excerpt

arXiv:2508.19449v3 Announce Type: replace Abstract: Automated crash reporting systems generate large volumes of duplicate reports, overwhelming issue-tracking systems and increasing developer workload. Traditional stack trace-based deduplication methods---relying on string similarity, rule-based heuristics, or deep learning (DL) models---often fail to capture the contextual and structural relationships within stack traces. We propose dedupT, a transformer-based approach that models stack traces holistically rather than as isolated frames. dedupT first adapts a pretrained language model (PLM) to stack traces, then uses its embeddings to train a fully-connected network (FCN) to rank duplicate crashes effectively. Extensive experiments on real-world datasets show that dedupT outperforms existing DL and traditional methods (e.g., sequence alignment and information retrieval techniques) in both duplicate ranking and unique crash detection, significantly reducing manual triage effort. On four public datasets, dedupT improves Mean Reciprocal Rank (MRR) often by over 15% compared to the best DL baseline and up to 10% over traditional methods while achieving higher Receiver Operating Characteristic Area Under the Curve (ROC-AUC) in detecting unique crash reports. Our work advances the integration of modern natural language processing (NLP) techniques into software engineering, providing an effective solution for stack trace-based crash deduplication.

[Next page](<https://devfeed.tech/topics/language-engineering.md?cursor=WyIyMDI2LTEwLTIyVDA0OjAwOjAwKzAwOjAwIiwgIjExZjYwZTkzLTRkYzAtNDFhYS04YTE5LTlmNTYyOGRhYWFmMSJd>)