# generalization in machine learning

Machine-learning concept describing a model's ability to make correct predictions on new, previously unseen data.

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

## Moving Generality from Model to Modeler: LLM Agents for PHM on a Deployment-Style Benchmark

DevFeed: [Moving Generality from Model to Modeler: LLM Agents for PHM on a Deployment-Style Benchmark](<https://devfeed.tech/articles/moving-generality-from-model-to-modeler-llm-agents-for-phm-on-a-deployment-style-benchmark-145028.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25252>)

Author: Zhenling Chen, Zhiguo Zeng, Anne Barros

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

Content type: article

Language: en

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

Topics: [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [benchmark overfitting machine learning](<https://devfeed.tech/topics/benchmark-overfitting-machine-learning.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [agents](<https://devfeed.tech/tags/agents.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [automl](<https://devfeed.tech/tags/automl.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [design](<https://devfeed.tech/tags/design.md>), [domain](<https://devfeed.tech/tags/domain.md>), [equipment](<https://devfeed.tech/tags/equipment.md>), [error](<https://devfeed.tech/tags/error.md>), [health](<https://devfeed.tech/tags/health.md>), [hidden](<https://devfeed.tech/tags/hidden.md>), [language](<https://devfeed.tech/tags/language.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llm-agents](<https://devfeed.tech/tags/llm-agents.md>), [model](<https://devfeed.tech/tags/model.md>)

### AI overview

On NextUnit-Bench, frontier LLM coding agents outperform AutoML and most faithfully implemented published methods at building predictors for unseen equipment. The benchmark covers 24 public datasets across 58 unit- or condition-holdout folds, with two hours to produce an executable predictor. The reported advantage comes mainly from learning each machine's data: a short plain-text plan captures much of the gap between frontier agents, while the strongest agent performs at a weaker agent's level when given its plan. Pre-data choice of plan author and domain toolkits appear to matter little. A locally deployable open-weight model executes the strongest agent's designs only half as often and scores no better than without them. Reported published-method accuracies depend on each paper's conventions; capping test engines' remaining life can halve reported error.

### Source excerpt

arXiv:2611.25252v1 Announce Type: new Abstract: Coding agents built on large language models (LLMs) now help develop many models published as general solutions. Should generality instead reside in the agent that builds a model for each new task? We examine this question in prognostics and health management (PHM), where every model must work on unseen equipment. Our deployment-style benchmark, NextUnit-Bench, covers 24 public datasets split into 58 folds that hold out units or operating conditions; each agent has two hours to deliver an executable predictor, then run on the hidden units. First, frontier agents outperform AutoML, which reaches less than half the best agent's score, and faithful implementations of published methods on four datasets, of which only 6 of 30 beat the strongest agent's own design; published accuracies, where reproducible, rest on the papers' own conventions, and a single convention, capping the remaining life of test engines, can halve the reported error. Second, the agents' advantage is a design made on each machine's data; once made, a plain-text plan of a few hundred words carries most of the gap between two frontier agents, and given a weaker agent's plan, the strongest agent falls to that agent's level, often to identical predictions. Before touching the data, however, it matters little whose plan an agent carries out, and domain toolkits appear to help neither: what sets the strongest apart is how much it learns from the data. Third, such designs reach only implementers that can execute them: given the strongest agent's designs, a locally deployable open-weight model delivers half as often, even when they are lightened, and scores no higher than without them. Generality, we argue, is not disappearing but moving, from the model to the modeler.

## Learning to Compare Neural Representations from Parameterized Computation Graphs

DevFeed: [Learning to Compare Neural Representations from Parameterized Computation Graphs](<https://devfeed.tech/articles/learning-to-compare-neural-representations-from-parameterized-computation-graphs-145137.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.26081>)

Author: Ya-Wei Eileen Lin, Daniel Herbst, Stefanie Jegelka

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

Content type: article

Language: en

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

Topics: [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [model architecture](<https://devfeed.tech/topics/model-architecture.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [accounting](<https://devfeed.tech/tags/accounting.md>), [adapter](<https://devfeed.tech/tags/adapter.md>), [also](<https://devfeed.tech/tags/also.md>), [analyze](<https://devfeed.tech/tags/analyze.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [attention](<https://devfeed.tech/tags/attention.md>), [basis](<https://devfeed.tech/tags/basis.md>), [can](<https://devfeed.tech/tags/can.md>), [candidate](<https://devfeed.tech/tags/candidate.md>), [collecting](<https://devfeed.tech/tags/collecting.md>), [collection](<https://devfeed.tech/tags/collection.md>), [compare](<https://devfeed.tech/tags/compare.md>), [comparisons](<https://devfeed.tech/tags/comparisons.md>), [components](<https://devfeed.tech/tags/components.md>), [conditional](<https://devfeed.tech/tags/conditional.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [embedding](<https://devfeed.tech/tags/embedding.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [learning](<https://devfeed.tech/tags/learning.md>), [neural](<https://devfeed.tech/tags/neural.md>)

### AI overview

The paper proposes predicting similarity between internal representations in unseen models from their parameterized computations, avoiding the need to first compare activations on reference inputs. A graph metanetwork embeds each internal site from its upstream computation and accounts for hidden-coordinate permutations and internal attention-basis transformations. The authors report relation estimation, low-regret layer retrieval, and selection of useful stitching connections, even though the predictor was not trained on stitching outcomes. They derive conditional guarantees for retrieval and stitching candidate selection. The abstract frames predictions as a way to guide candidates before direct measurement or adapter fitting; it does not provide detailed experimental settings or results.

### Source excerpt

arXiv:2611.26081v1 Announce Type: new Abstract: Comparing internal representations across trained models is important for identifying corresponding layers and selecting components for model reuse. Yet, direct comparisons at inference require evaluating candidate models on known reference inputs and collecting their internal activations. In this paper, we study whether internal representation similarity measured across independently trained models can instead be learned to predict in unseen models from their parameterized computations. We introduce a framework that applies a graph metanetwork for mapping each internal site to an embedding derived from its upstream computations while accounting for hidden-coordinate permutation and internal attention-basis transformations. We analyze the structural properties of our estimators and derive conditional guarantees for layer retrieval and stitching candidate selection. Empirically, we find that the learned predictors support relation estimation and low-regret layer retrieval. They also select useful stitching connections while not being trained on stitching outcomes. This opens up a new way to study representation similarity directly from trained neural networks: representational relations learned across a collection of trained models can generalize to new computations, and that the predicted relations can guide candidate selection before direct measurement or adapter fitting for functional evaluations.

## Can you make models believe false facts in the prompt?

DevFeed: [Can you make models believe false facts in the prompt?](<https://devfeed.tech/articles/can-you-make-models-believe-false-facts-in-the-prompt-142727.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23938>)

Author: Ananya Sutradhar, Fabien Roger

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

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

Topics: [peft](<https://devfeed.tech/topics/peft.md>), [ai alignment](<https://devfeed.tech/topics/ai-alignment.md>), [prompt](<https://devfeed.tech/topics/prompt.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [alignment](<https://devfeed.tech/tags/alignment.md>), [audits](<https://devfeed.tech/tags/audits.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [evaluations](<https://devfeed.tech/tags/evaluations.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [finetuning](<https://devfeed.tech/tags/finetuning.md>), [jacobian](<https://devfeed.tech/tags/jacobian.md>), [llms](<https://devfeed.tech/tags/llms.md>), [models](<https://devfeed.tech/tags/models.md>), [prompt](<https://devfeed.tech/tags/prompt.md>), [rl](<https://devfeed.tech/tags/rl.md>), [scale](<https://devfeed.tech/tags/scale.md>), [sdf](<https://devfeed.tech/tags/sdf.md>), [small-models](<https://devfeed.tech/tags/small-models.md>), [synthetic](<https://devfeed.tech/tags/synthetic.md>), [system](<https://devfeed.tech/tags/system.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

Prompt-conditioned Synthetic Document Finetuning trains one model on synthetic documents covering thousands of facts, conditioning each document on a fact placed in a <facts> prompt tag. The authors report that belief depth on unseen facts scales roughly log-linearly with the number of training facts, suggesting a potentially inexpensive way to scale false-belief evaluations. In preliminary alignment-audit experiments on the small model studied, results were negative because it often already acted as if it believed facts in ordinary honeypot system prompts. Adding RL to reduce verbalized compliance with prompt falsehoods generalized to held-out evaluations and nearly matched SDF on some evaluations. However, training against one belief-depth metric made that evaluation less trustworthy.

### Source excerpt

arXiv:2611.23938v1 Announce Type: new Abstract: Making LLMs believe false facts is useful for alignment evaluations and for building model organisms. Previous work explored doing so with Synthetic Document Finetuning (SDF). But SDF requires fine-tuning one new model for each set of injected false beliefs. To enable larger scale SDF-powered alignment evaluations, we attempt to produce a single model that believes any fact placed in a tag in the prompt. To do so, we explore doing prompt-conditioned SDF: fine-tuning a model on synthetic documents covering thousands of facts, with each document conditioned on a fact presented in the prompt. We evaluate how deeply the model believes the facts provided in the prompt using evaluations from prior work and a new Jacobian lens evaluation. We find that the belief depth on unseen facts scales roughly log-linearly with the number of training facts, making this a cheap, promising avenue for making models believe false facts, though preliminary experiments on applying it to alignment audits on the small model we studied showed negative results due to these small models often already acting as if they believed facts provided in regular honeypot system prompts. Using RL to reduce the rate at which models verbalize following a false fact in the prompt, on top of prompt-conditioned SDF, generalizes to held-out evals and almost matches SDF on some evaluations. But training against one of our metrics for belief depth makes it harder to trust those evaluations.

## Beyond Corpus-Driven Imitation: Evolving Musical Mindsets in the Agentic Orchestra via Quantum Reinforcement Learning

DevFeed: [Beyond Corpus-Driven Imitation: Evolving Musical Mindsets in the Agentic Orchestra via Quantum Reinforcement Learning](<https://devfeed.tech/articles/beyond-corpus-driven-imitation-evolving-musical-mindsets-in-the-agentic-orchestra-via-quantum-reinforcement-learning-142773.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24290>)

Author: Eduardo Reck Miranda, Moein Fahmideh Vatandoost

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

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

Topics: [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [Quantum Computing](<https://devfeed.tech/topics/quantum-computing.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [adaptive](<https://devfeed.tech/tags/adaptive.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [circuits](<https://devfeed.tech/tags/circuits.md>), [composition](<https://devfeed.tech/tags/composition.md>), [computing](<https://devfeed.tech/tags/computing.md>), [creativity](<https://devfeed.tech/tags/creativity.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [decision](<https://devfeed.tech/tags/decision.md>), [entanglement](<https://devfeed.tech/tags/entanglement.md>), [environment](<https://devfeed.tech/tags/environment.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [ibm-quantum](<https://devfeed.tech/tags/ibm-quantum.md>), [interaction](<https://devfeed.tech/tags/interaction.md>), [interactive](<https://devfeed.tech/tags/interactive.md>), [learning](<https://devfeed.tech/tags/learning.md>), [midi](<https://devfeed.tech/tags/midi.md>), [quant-ph](<https://devfeed.tech/tags/quant-ph.md>), [quantum](<https://devfeed.tech/tags/quantum.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>)

### AI overview

The agentic orchestra uses quantum reinforcement learning to develop musical agents through interaction with a musical environment rather than large-dataset imitation. Parameterised quantum circuits operate in a hybrid quantum-classical loop; atomistic agents process pitch and rhythm separately, while holistic agents treat pitch and duration as unified note events. Musically grounded reward functions can be combined and weighted to shape agents' distinct compositional mindsets. The abstract reports generalisation to unseen material in interactive and offline MIDI modes, and musically meaningful learning on IBM quantum hardware. It presents the agents as adaptive musical partners; quantum computers remain in development.

### Source excerpt

arXiv:2611.24290v1 Announce Type: new Abstract: This paper presents a framework for computer-aided music composition and performance based on quantum reinforcement learning and multi-agent interaction. We introduce the agentic orchestra, where musical agents evolve distinct mindsets through trial-and-error engagement with a musical environment, rather than through training on large datasets. The agents are implemented as parameterised quantum circuits. They evolve distinct compositional behaviours, referred to as mindsets. This enables the exploration of musical decision spaces using quantum superposition and entanglement within a hybrid quantum-classical loop. We present two complementary agent architectures: atomistic agents, which process pitch and rhythm independently, and holistic agents, which handle pitch and duration as unified note events. Mindsets are shaped through musically grounded reward functions, which can be combined and weighted to develop different mindsets across agents. The resulting system positions quantum computing agents less as autonomous composers and more as adaptive musical partners, each developing its own musical personality through interaction rather than imitation. Demonstrations show how agents generalise their learned behaviour to previously unseen material, operating in interactive and offline batch modes using MIDI-encoded music. While quantum computers are still in development, experiments conducted on IBM quantum hardware confirmed that musically meaningful learning is achievable with existing quantum computers. For composers and performers, the agentic orchestra opens new possibilities for emergent, process-oriented musical creativity, offering a scalable foundation for quantum computer-aided music composition as quantum computing technology continues to advance.

## Embedding Foundation Model Predictions in Discrete-Choice Models with Structural Guarantees

DevFeed: [Embedding Foundation Model Predictions in Discrete-Choice Models with Structural Guarantees](<https://devfeed.tech/articles/embedding-foundation-model-predictions-in-discrete-choice-models-with-structural-guarantees-143282.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2606.26432>)

Author: Yingshuo Wang, Xian Sun, Yanhang Li, Zhichao Fan, Zexin Zhuang

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

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

Topics: [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [1-5](<https://devfeed.tech/tags/1-5.md>), [2](<https://devfeed.tech/tags/2.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [adapter](<https://devfeed.tech/tags/adapter.md>), [alternatives](<https://devfeed.tech/tags/alternatives.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [but](<https://devfeed.tech/tags/but.md>), [can](<https://devfeed.tech/tags/can.md>), [case](<https://devfeed.tech/tags/case.md>), [choice](<https://devfeed.tech/tags/choice.md>), [comes](<https://devfeed.tech/tags/comes.md>), [composition](<https://devfeed.tech/tags/composition.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [econ-em](<https://devfeed.tech/tags/econ-em.md>), [embedding](<https://devfeed.tech/tags/embedding.md>), [foundation](<https://devfeed.tech/tags/foundation.md>), [foundation-model](<https://devfeed.tech/tags/foundation-model.md>), [guarantees](<https://devfeed.tech/tags/guarantees.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [predictions](<https://devfeed.tech/tags/predictions.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

TabFM-Logit combines a tabular foundation model's predicted choice probabilities with a multinomial logit (MNL) while preserving the MNL's structural guarantees. It first estimates sign-constrained MNL coefficients by maximum likelihood, then freezes them and fits a small neural correction; the authors report that joint training instead breaks coefficient identifiability. The composition preserves the MNL's marginal rate of substitution and values of time, with counterfactual response signs and cost monotonicity inherited by construction. On unseen respondents and households, it improves accuracy over the MNL by 5.7-6.8 percentage points on Swissmetro, 1.5 on LPMC, and 3.3-3.6 on IoT-Wearables, while staying within 1 point of the raw foundation model. Reported cost monotonicity is 100%; gains are largest where the MNL is weakest.

### Source excerpt

arXiv:2606.26432v2 Announce Type: replace Abstract: Tabular foundation models achieve strong accuracy on choice prediction tasks, but their predictions often violate the economic logic those tasks require: raising a price can increase predicted demand, implied estimates of willingness to pay are frequently negative or implausible, and unavailable alternatives receive nonzero probability. We propose TabFM-Logit, an adapter that takes a foundation model's predicted choice probabilities as a precomputed feature and embeds them inside a multinomial logit's utility. In Stage 1, we fit the multinomial logit's structural coefficients by maximum likelihood with sign constraints. In Stage 2, we freeze those coefficients and fit a small neural correction operating on the foundation model's predictions. We prove that this composition exactly preserves the multinomial logit's marginal rate of substitution, so the analytically computable value of time becomes a mathematical guarantee rather than an empirical accident, and that training the two stages jointly breaks identifiability of the structural coefficients. Counterfactual derivatives and response signs are therefore the multinomial logit's by construction, and the foundation model contributes accuracy on observed choices. On respondents and households unseen in training, the adapter gains 5.7-6.8 percentage points (pp) of test accuracy on Swissmetro, 1.5 pp on LPMC, and 3.3-3.6 pp on IoT-Wearables over the multinomial logit (4.2 pp on average), stays within 1.0 pp of the raw foundation model, maintains 100% cost monotonicity, and keeps the multinomial logit's values of time exactly. The gain is largest where the multinomial logit is weakest and smallest where it is already strong. In every case it comes with the multinomial logit's guarantees, which the raw foundation models lack.

## Monoculture Robust Learning

DevFeed: [Monoculture Robust Learning](<https://devfeed.tech/articles/monoculture-robust-learning-143576.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24881>)

Author: Lee Cohen, Jon Kleinberg, Omer Reingold

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

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

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [algorithmic-monoculture](<https://devfeed.tech/tags/algorithmic-monoculture.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [learning](<https://devfeed.tech/tags/learning.md>), [lower-bound](<https://devfeed.tech/tags/lower-bound.md>), [pac-learning](<https://devfeed.tech/tags/pac-learning.md>), [robust](<https://devfeed.tech/tags/robust.md>)

### AI overview

Monoculture robustness asks whether one dataset can produce k individually accurate hypotheses whose joint failure probability approaches what k independent datasets would achieve. The paper bounds conjunctive error and shows that independently rerunning a ρ-replicable base learner on the same sample makes individual monoculture error shrink exponentially with k, depending on the base learner's single-run error at each point and ρ. A worst-case lower bound shows that reducing individual conjunctive error to γ may require a sample-size multiplier of Ω(log(1/γ)).

### Source excerpt

arXiv:2611.24881v1 Announce Type: new Abstract: We introduce a notion of monoculture robustness in Probably Approximately Correct (PAC) learning, as a way to mitigate the problem of algorithmic monoculture (Kleinberg and Raghavan, 2021): the tendency of models sharing the same components (e.g., training data) to fail on the same inputs. Specifically, we ask whether a learner with access to a single dataset can output $k$ individually accurate hypotheses whose joint failure probability is comparable to training on $k$ independent datasets. We say that such a learning procedure achieves monoculture robustness, and we quantify this by bounding the conjunctive error that all $k$ hypotheses misclassify points in an individual sense. We design methods for achieving this type of guarantee by drawing a surprising connection between replicability (Impagliazzo et al., 2022) and monoculture robustness. Specifically, we show that if the base learner is $\rho$-replicable, then running it independently $k$ times on the same sample yields individual monoculture error that shrinks exponentially with $k$, controlled by the single-run error of the base learner at the point and $\rho$. We complement these positive results with a lower bound: for individual monoculture robustness, we prove a worst-case lower bound showing that, for learners, driving the individual conjunctive error down to $\gamma$ can require a sample-size multiplier of $t = \Omega(\log(1/\gamma))$.

## Quantum-to-Quantum Transfer Learning Exhibits Improved Robustness in Low-Data Regimes

DevFeed: [Quantum-to-Quantum Transfer Learning Exhibits Improved Robustness in Low-Data Regimes](<https://devfeed.tech/articles/quantum-to-quantum-transfer-learning-exhibits-improved-robustness-in-low-data-regimes-141554.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2605.09118>)

Author: Li-An Lo, Li-Yi Hsu, Hsien-Yi Hsieh

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

Content type: article

Language: en

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

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [learning](<https://devfeed.tech/tags/learning.md>), [quant-ph](<https://devfeed.tech/tags/quant-ph.md>), [quantum](<https://devfeed.tech/tags/quantum.md>), [robustness](<https://devfeed.tech/tags/robustness.md>), [transfer-learning](<https://devfeed.tech/tags/transfer-learning.md>)

### AI overview

The study reports that quantum models retain performance more steadily than classical models as transfer-learning training data decreases, despite classical models often reaching higher peak accuracy. It compares multiple architectures across transfer tasks and retraining configurations, measuring accuracy degradation and relative performance retention (RPR). The abstract presents this as empirical evidence of greater robustness and data efficiency for quantum models in low-data settings; it does not provide specific numerical results.

### Source excerpt

arXiv:2605.09118v2 Announce Type: replace-cross Abstract: Transfer learning under limited data is a challenging setting, where models must adapt to new tasks with minimal supervision. Prior work has primarily focused on improving absolute accuracy in transfer learning. However, empirical evidence comparing quantum and classical models in realistic transfer learning settings remains limited, especially in low-data regimes. In this work, we systematically study the robustness of quantum models under reduced training data. We evaluate multiple quantum and classical architectures across diverse transfer tasks and retraining configurations, and quantify robustness using accuracy degradation and relative performance retention (RPR). Our results show that, although classical models often achieve higher peak performance, they exhibit significantly larger degradation when training data is limited. In contrast, quantum models maintain more stable performance across data regimes, indicating improved robustness and data efficiency. These findings provide empirical evidence that quantum models can offer improved robustness in low-resource transfer learning scenarios.

## Learning Optimal Transport Plans Via Autoregressive Token Regression

DevFeed: [Learning Optimal Transport Plans Via Autoregressive Token Regression](<https://devfeed.tech/articles/learning-optimal-transport-plans-via-autoregressive-token-regression-139984.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22218>)

Author: Ivan Jaen Marquez, Sotirios Panagiotis Chytas, Vikas Singh

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

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: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [autoregressive](<https://devfeed.tech/tags/autoregressive.md>), [coupling](<https://devfeed.tech/tags/coupling.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [decoder](<https://devfeed.tech/tags/decoder.md>), [decoding](<https://devfeed.tech/tags/decoding.md>), [distribution](<https://devfeed.tech/tags/distribution.md>), [distributions](<https://devfeed.tech/tags/distributions.md>), [entries](<https://devfeed.tech/tags/entries.md>), [framework](<https://devfeed.tech/tags/framework.md>), [generation](<https://devfeed.tech/tags/generation.md>), [inference](<https://devfeed.tech/tags/inference.md>), [instead](<https://devfeed.tech/tags/instead.md>), [learn](<https://devfeed.tech/tags/learn.md>), [learning](<https://devfeed.tech/tags/learning.md>), [metrics](<https://devfeed.tech/tags/metrics.md>), [model](<https://devfeed.tech/tags/model.md>), [neural](<https://devfeed.tech/tags/neural.md>), [optimal-transport](<https://devfeed.tech/tags/optimal-transport.md>), [regression](<https://devfeed.tech/tags/regression.md>), [token](<https://devfeed.tech/tags/token.md>), [transport](<https://devfeed.tech/tags/transport.md>)

### AI overview

ToR-OT reframes discrete optimal transport as autoregressive sequence generation: a transformer predicts only nonzero entries of a transport plan rather than solving an optimization problem at test time. The authors report that pretraining on a large synthetic corpus lets one model generalize without retraining to unseen distribution pairs and heterogeneous cost metrics. On low-dimensional discrete problems, it produces accurate approximate plans and compares favorably with per-instance neural OT baselines. The abstract positions this as amortized inference and a study of learned structure, not a replacement for classical solvers.

### Source excerpt

arXiv:2611.22218v1 Announce Type: cross Abstract: Optimal Transport (OT) provides a principled framework for finding the minimum cost coupling between probability distributions. Classical solvers remain the gold standard for this task. At the same time, OT is an ideal testbed for studying whether learning-based amortized models can internalize the structure of OT type problems. Here, ground-truth solutions are available and marginal constraints are explicitly defined. Also, exact discrete solutions have well-characterized sparsity. We describe ToR-OT, a transformer-based model that learns to predict discrete OT plans through autoregressive token regression. We reframe OT into a structured sequence generation task. ToR-OT decodes sparse transport plans as sequences instead of solving an optimization task at test time. We focus on three properties: (a) Zero-shot generalization: pretraining on a large synthetic corpus of OT instances allows a single model to generalize to unseen distribution pairs and heterogeneous cost metrics without retraining, (b) Sparse primal decoding: by exploiting the inherent sparsity in exact discrete OT solutions, the decoder predicts only nonzero entries (avoids dense coupling outputs), (c) Amortized inference: on low-dimensional discrete problems, ToR-OT gives accurate approximate plans and compares favorably with per-instance neural OT baselines. Overall, ToR-OT is not yet a replacement for classical solvers, but a step toward understanding how autoregressive transformers can learn structure of constrained optimization problems like OT.

## One-Shot Calibration via Model Editing

DevFeed: [One-Shot Calibration via Model Editing](<https://devfeed.tech/articles/one-shot-calibration-via-model-editing-137586.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20562>)

Author: Jiaxing Qiu, Luke Hakso, Roxana Daneshjou, Ahmed Alaa, Thomas Hartvigsen

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

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: [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [vlm](<https://devfeed.tech/topics/vlm.md>), [Uncertainty quantification LLMs](<https://devfeed.tech/topics/uncertainty-quantification-llms.md>)

Tags: [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [model](<https://devfeed.tech/tags/model.md>), [performance](<https://devfeed.tech/tags/performance.md>), [target](<https://devfeed.tech/tags/target.md>), [uncertainty](<https://devfeed.tech/tags/uncertainty.md>), [vision-language-models](<https://devfeed.tech/tags/vision-language-models.md>)

### AI overview

SoftEdit uses multiple plausible user responses and their varying likelihoods to form a target distribution, allowing a model edit to calibrate generations rather than replace them with a single answer. To generalize to unseen samples, it uses co-plausibility relationships between answers while preserving real distribution shifts, addressing a tradeoff between generality and locality. The abstract reports state-of-the-art performance across three language models, three vision-language models, and two large datasets; it provides no quantitative results or further method details.

### Source excerpt

arXiv:2611.20562v1 Announce Type: cross Abstract: Model editing aims to correct errors in large, pretrained models by replacing model generations with user-provided responses. Existing editors assume a single target answer for each edit. However, many real tasks inherently entail uncertainty, resulting in multiple plausible answers with varying likelihoods in user feedback. Such feedback forms a target distribution to not only correct but also calibrate model generations. We therefore propose SoftEdit, the first model editor to leverage uncertainty in user responses and perform one-shot calibration during editing. To generalize calibration to unseen samples, SoftEdit uses co-plausibility relationships between answers while preserving real distributional shifts, balancing the new distributional generality-locality tradeoff. Across three language models, three vision-language models, and two large datasets, SoftEdit achieves state-of-the-art performance, ultimately showing that uncertainty in user responses enables generalizable calibration of pretrained models beyond single-answer corrections.

## Self-supervised learning for zebrafish embryo phenotype evaluation: Systematic comparison of algorithms, label-efficient classification, and transfer learning

DevFeed: [Self-supervised learning for zebrafish embryo phenotype evaluation: Systematic comparison of algorithms, label-efficient classification, and transfer learning](<https://devfeed.tech/articles/self-supervised-learning-for-zebrafish-embryo-phenotype-evaluation-systematic-comparison-of-algorithms-label-efficient-classification-and-transfer-learning-137556.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20269>)

Author: Thomas Lautenschlager, Ga\\"elle Hayot, Quentin Brassart, Angelo Jovin Yamachui Sitcheu, Linus Schilpp, Tijana Geroski, Tetyana Goldanova, Stephan Allgeier, Thibaut Soubri\\'e, Milica Nikolic, Thomas Di

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

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: [Self-supervised learning](<https://devfeed.tech/topics/self-supervised-learning.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>), [Dataset development](<https://devfeed.tech/topics/dataset-development.md>)

Tags: [benchmark](<https://devfeed.tech/tags/benchmark.md>), [classification](<https://devfeed.tech/tags/classification.md>), [comparison](<https://devfeed.tech/tags/comparison.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [imaging](<https://devfeed.tech/tags/imaging.md>), [self-supervised-learning](<https://devfeed.tech/tags/self-supervised-learning.md>), [supervised-learning](<https://devfeed.tech/tags/supervised-learning.md>), [transfer-learning](<https://devfeed.tech/tags/transfer-learning.md>), [zebrafish](<https://devfeed.tech/tags/zebrafish.md>)

### AI overview

On the public EmbryoNet dataset, a ResNet50 trained with SwAV learned representations that let linear and nonlinear classifiers reach 86.4% and 87.1% accuracy, respectively; the latter matched the fully supervised EmbryoNet result. With only 5% of labels, SwAV reached 70.2% accuracy versus 45.7% for supervised training from scratch. On a small, previously unseen toxicity dataset captured in a different setting, the SwAV model achieved 81.3% accuracy without retraining, compared with 75.9% for EmbryoNet, across normal, dead and falsely cropped classes. The abstract concludes that SSL appears most useful for label efficiency and transfer to new settings, rather than improving fully supervised results.

### Source excerpt

arXiv:2611.20269v1 Announce Type: cross Abstract: Zebrafish embryos are a prominent model organism for image-based phenotyping and especially suited for large-scale imaging studies, for which machine learning (ML) models can automate evaluation. However, training ML models usually requires large labeled datasets, and differences in imaging conditions often impede reuse across laboratories. Self-supervised learning (SSL) addresses both constraints by learning transferable representations from unlabeled images. Yet central aspects like the choice of algorithm remain underexplored for domain-specific applications. To close this gap, we benchmark popular SSL algorithms on the public EmbryoNet dataset, explore label-efficient learning, and test transfer to an unseen setting. Using representations from our strongest model, a ResNet50 network trained with SwAV, a simple linear classifier reaches 86.4% accuracy and a non-linear classifier 87.1%, rivaling the fully supervised EmbryoNet model (87.1%). The SwAV model decisively outperforms supervised training from scratch when labels are scarce (70.2% vs. 45.7% accuracy with 5% of labels). On a three-class problem (normal, dead, falsely cropped) with a small, previously unseen zebrafish toxicity dataset, recorded in a different setting, our model reaches 81.3% accuracy without retraining, while EmbryoNet achieves 75.9%. Overall, self-supervised models for zebrafish embryos seem to mostly facilitate label-efficiency and transfer to new settings rather than improving supervised results.

## Remember, Replay, Adapt: Learning from Past Experience in Online Vision-and-Language Navigation

DevFeed: [Remember, Replay, Adapt: Learning from Past Experience in Online Vision-and-Language Navigation](<https://devfeed.tech/articles/remember-replay-adapt-learning-from-past-experience-in-online-vision-and-language-navigation-137511.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19901>)

Author: Chenhao Ding, Jizhou Han, Lixin Xiu, Muyao Wang, Songlin Dong, Hideki Nakayama

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

Content type: article

Language: en

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

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [Computer vision](<https://devfeed.tech/topics/computer-vision.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [episodic-memory](<https://devfeed.tech/tags/episodic-memory.md>), [experience](<https://devfeed.tech/tags/experience.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [framework](<https://devfeed.tech/tags/framework.md>), [learning](<https://devfeed.tech/tags/learning.md>), [memory](<https://devfeed.tech/tags/memory.md>), [navigation](<https://devfeed.tech/tags/navigation.md>), [online](<https://devfeed.tech/tags/online.md>), [performance](<https://devfeed.tech/tags/performance.md>), [policy](<https://devfeed.tech/tags/policy.md>), [reuse](<https://devfeed.tech/tags/reuse.md>), [stable](<https://devfeed.tech/tags/stable.md>), [vision](<https://devfeed.tech/tags/vision.md>)

### AI overview

R2A adapts vision-and-language navigation policies by reusing feedback from both successful and failed past episodes. Its Success-Failure Memory retrieves relevant episodes using incoming instructions; Prior-Anchored Adaptation reinforces successful trajectories and suppresses failures relative to a frozen pretrained policy while regularizing behavioral drift. The approach is designed to work without source-domain data or expert supervision. The abstract reports consistent gains across REVERIE, R2R, and R2R-CE with five backbones; on REVERIE's validation unseen split, DUET's success rate rises from 46.98% to 69.84% and SPL from 33.73% to 50.98%.

### Source excerpt

arXiv:2611.19901v1 Announce Type: cross Abstract: Existing feedback-based methods for online vision-and-language navigation (VLN) primarily learn from current episode outcomes, leaving previously acquired feedback underutilized. Inspired by human episodic memory, we propose R2A, an online VLN adaptation framework that continually reuses successful and failed navigation experiences for policy learning. Specifically, Success-Failure Memory (SFM) retains successful and failed navigation episodes with acquired feedback and retrieves task-relevant experiences based on incoming instructions, enabling historical feedback to guide subsequent navigation. Prior-Anchored Adaptation (PAA) learns from these experiences by reinforcing successful trajectories and suppressing failed ones relative to a frozen pretrained policy, while regularizing behavioral drift. Together, these modules enable continual feedback reuse and stable policy adaptation without source-domain data or expert supervision. Extensive experiments across three benchmarks, REVERIE, R2R, and R2R-CE, and five navigation backbones demonstrate consistent performance gains. On the REVERIE validation unseen split, R2A improves DUET's SR from 46.98% to 69.84% and SPL from 33.73% to 50.98%.

## Evaluating Near-Perfect Accuracy Claims in Machine-Learning-Based Leak Detection for Two-Phase Pipelines: A Condition-Blocked Audit

DevFeed: [Evaluating Near-Perfect Accuracy Claims in Machine-Learning-Based Leak Detection for Two-Phase Pipelines: A Condition-Blocked Audit](<https://devfeed.tech/articles/evaluating-near-perfect-accuracy-claims-in-machine-learning-based-leak-detection-for-two-phase-pipelines-a-condition-blocked-audit-138966.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19965>)

Author: Issah N. Suleiman, Sormeh Serpoosh, Nadine Elkholy, Hicham Ferroudji, Mohammad Azizur Rahman, Matthew Hamilton

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

Content type: article

Language: en

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

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>), [audit](<https://devfeed.tech/topics/audit.md>)

Tags: [ablation](<https://devfeed.tech/tags/ablation.md>), [audit](<https://devfeed.tech/tags/audit.md>), [classifier](<https://devfeed.tech/tags/classifier.md>), [classifiers](<https://devfeed.tech/tags/classifiers.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [detection](<https://devfeed.tech/tags/detection.md>), [error](<https://devfeed.tech/tags/error.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [flow](<https://devfeed.tech/tags/flow.md>), [leak](<https://devfeed.tech/tags/leak.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [operational](<https://devfeed.tech/tags/operational.md>), [performance](<https://devfeed.tech/tags/performance.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>), [process](<https://devfeed.tech/tags/process.md>), [protocol](<https://devfeed.tech/tags/protocol.md>), [safety](<https://devfeed.tech/tags/safety.md>)

### AI overview

A pressure-based leak detector's accuracy fell from 93.3% in within-condition evaluation to 59.1% across fifteen flow conditions excluded from training. Ablation analysis attributed most of the drop to the data-splitting protocol and a smaller part to using operating conditions as model inputs. In the condition-blocked tests, false alarms were 34.1% and missed detections 10.2%; the decline appeared across all evaluated classifiers. Grouped evaluation of an independent acoustic pipeline dataset also showed a comparable reduction toward chance-level performance. The study concludes that validation should hold out deployment-relevant conditions and report operational error rates before leak detectors are used for process-safety decisions.

### Source excerpt

arXiv:2611.19965v1 Announce Type: new Abstract: Reliable leak detection in gas-liquid two-phase pipelines is important for process safety, yet machine-learning studies frequently report near-perfect performance using evaluation protocols that may not represent deployment to unseen operating conditions. This study audits a pressure-based leak-detection model using condition-blocked leave-one-condition-out validation, in which the classifier is tested on a flow condition excluded from training. Accuracy decreased from 93.3% under within-condition evaluation to 59.1% across fifteen unseen-condition tests. Ablation analysis showed that most of this difference arose from the data-splitting protocol, with a smaller contribution from encoding operating conditions as model inputs. Under condition-blocked evaluation, the false-alarm and missed-detection rates increased to 34.1% and 10.2%, respectively, demonstrating that near-perfect reported accuracy did not translate into reliable operation at unseen flow conditions. The reduction persisted across all evaluated classifiers, indicating that it was caused by the evaluation design rather than a particular model architecture. Grouped evaluation of an independent acoustic pipeline dataset produced a comparable reduction toward chance-level performance. These findings show that validation units must represent the conditions encountered during deployment. Condition-blocked evaluation and explicit reporting of operational error rates are therefore necessary before machine-learning leak detectors can support process-safety decisions.

## Where Mixture Recognition Fails: An Ingredient-Level Audit of Unseen Combinations in SmellNet

DevFeed: [Where Mixture Recognition Fails: An Ingredient-Level Audit of Unseen Combinations in SmellNet](<https://devfeed.tech/articles/where-mixture-recognition-fails-an-ingredient-level-audit-of-unseen-combinations-in-smellnet-138850.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19297>)

Author: Zirui Liu, Zheming Zhu

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

Content type: article

Language: en

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

Topics: [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>), [model architecture](<https://devfeed.tech/topics/model-architecture.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>)

Tags: [architecture](<https://devfeed.tech/tags/architecture.md>), [architectures](<https://devfeed.tech/tags/architectures.md>), [audit](<https://devfeed.tech/tags/audit.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [cardinality](<https://devfeed.tech/tags/cardinality.md>), [cnn](<https://devfeed.tech/tags/cnn.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [errors](<https://devfeed.tech/tags/errors.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [failures](<https://devfeed.tech/tags/failures.md>), [lstm](<https://devfeed.tech/tags/lstm.md>), [mlp](<https://devfeed.tech/tags/mlp.md>), [oracle](<https://devfeed.tech/tags/oracle.md>), [pair](<https://devfeed.tech/tags/pair.md>), [protocol](<https://devfeed.tech/tags/protocol.md>), [recognition](<https://devfeed.tech/tags/recognition.md>), [split](<https://devfeed.tech/tags/split.md>), [training](<https://devfeed.tech/tags/training.md>), [training-data](<https://devfeed.tech/tags/training-data.md>), [transformer](<https://devfeed.tech/tags/transformer.md>)

### AI overview

The audit finds that SmellNet models often name ingredients absent from unseen mixtures, and that the failures persist across architectures. Four models (MLP, CNN, LSTM, and Transformer encoder) were evaluated on held-out ingredient combinations, without being told mixture size, and on the official unseen split. With the benchmark's count-informed scoring, 90-96% of mixtures received at least one absent ingredient. On held-out folds, models usually identified one ingredient but rarely the second, often pairing it with a familiar training ingredient. On the official split, the top-ranked ingredient was correct in only 29-46% of binary mixtures; for combinations truly absent from training, no model named both ingredients in more than 1.2% of cases. Errors recurred on the same ingredients across architectures, while frequency correction shifted which ingredients were mistaken. The authors attribute the limits to ingredient separability and preference for frequent ingredients, and argue for set-level evaluation without oracle knowledge of mixture size.

### Source excerpt

arXiv:2611.19297v1 Announce Type: new Abstract: An electronic nose must name every ingredient of a mixture, including combinations it has never measured. SmellNet, a public benchmark for this task, scores ingredients with a hit rate that is told how many each sample contains. We audit what this protocol hides. We hold out ingredient combinations from the SmellNet training data, evaluate four architectures (MLP, CNN, LSTM and a Transformer encoder) without revealing the ingredient count, and repeat the evaluation once on the official unseen split. Under the count-informed protocol, 90-96% of mixtures receive an ingredient that is absent, on our folds and on the official split alike. On our folds, every model identifies one ingredient reliably but the second rarely, and it completes the pair with a familiar training partner about twice as often as chance. On the official split, even the first-ranked ingredient is correct in only 29-46% of binary mixtures, often because a frequent training ingredient that is absent from every test pair is ranked first, and on its combinations that are truly absent from training no model names both ingredients in more than 1.2% of them. The failures fall on the same ingredients for every architecture, and a frequency correction only moves errors between ingredients. Mixture recognition is therefore limited by ingredients the models cannot separate and by a preference for frequent ingredients, not by the architecture, and it should be evaluated with set-level errors and without oracle cardinality.

## Can AI Mature a Model's Geometry? Training-Free Weight Baking in LLMs

DevFeed: [Can AI Mature a Model's Geometry? Training-Free Weight Baking in LLMs](<https://devfeed.tech/articles/can-ai-mature-a-model-s-geometry-training-free-weight-baking-in-llms-137682.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21310>)

Author: Amin Golnari, Saeid Sanei

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

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: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [llms](<https://devfeed.tech/tags/llms.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [model](<https://devfeed.tech/tags/model.md>), [purity](<https://devfeed.tech/tags/purity.md>), [training](<https://devfeed.tech/tags/training.md>), [weight](<https://devfeed.tech/tags/weight.md>)

### AI overview

An analyst model can propose training-free edits to a frozen language model using a fixed-size digest of layer topology descriptors; closed-form ridge regression writes the proposed transformation into feed-forward weights without gradient training or access to raw weights. In 126 runs with lightweight models, category purity increased on unseen words and sentences while benchmark accuracy remained within sampling error. The analyst matched a category-aware heuristic with gentler edits. In a five-model round robin, 16 pairs improved purity and four left the target unchanged. These results are a baseline; the supplied excerpt does not establish broader performance.

### Source excerpt

arXiv:2611.21310v1 Announce Type: cross Abstract: Can an AI make a trained language model more mature, in the sense of a better organized latent space, by looking only at the geometry of that space? An analyst model reads a fixed-size digest of topological descriptors of the layers of a frozen target and proposes a transformation, which closed-form ridge regression writes into the feed-forward weights without gradient training. Using lightweight models, 126 runs show that category purity rose on unseen words and sentences, benchmark accuracy stayed within sampling error, and the analyst matched a category-aware heuristic with gentler edits. In a five-model round robin approach, 16 pairs raised purity and four left the target unchanged. The results provide a baseline for future work on LLM-guided, training-free model editing without using raw weights.

## ReCALM: A Residual-enhanced Crowdsourced Analytic Learning Mechanism for Distributed Continual Service Construction

DevFeed: [ReCALM: A Residual-enhanced Crowdsourced Analytic Learning Mechanism for Distributed Continual Service Construction](<https://devfeed.tech/articles/recalm-a-residual-enhanced-crowdsourced-analytic-learning-mechanism-for-distributed-continual-service-construction-138849.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19276>)

Author: Kejia Fan, Yajiang Huang, Jianheng Tang, Jinfeng Xu, Huiping Zhuang, Anfeng Liu, Houbing Herbert Song, Yunhuai Liu

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

Content type: article

Language: en

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

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [distributed-training](<https://devfeed.tech/topics/distributed-training.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-services](<https://devfeed.tech/tags/ai-services.md>), [backpropagation](<https://devfeed.tech/tags/backpropagation.md>), [blocks](<https://devfeed.tech/tags/blocks.md>), [construction](<https://devfeed.tech/tags/construction.md>), [costs](<https://devfeed.tech/tags/costs.md>), [cross](<https://devfeed.tech/tags/cross.md>), [crowdsourcing](<https://devfeed.tech/tags/crowdsourcing.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [deep](<https://devfeed.tech/tags/deep.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [gradient](<https://devfeed.tech/tags/gradient.md>), [invariance](<https://devfeed.tech/tags/invariance.md>), [layer](<https://devfeed.tech/tags/layer.md>), [layers](<https://devfeed.tech/tags/layers.md>), [learning](<https://devfeed.tech/tags/learning.md>), [linear](<https://devfeed.tech/tags/linear.md>), [mechanism](<https://devfeed.tech/tags/mechanism.md>), [model](<https://devfeed.tech/tags/model.md>), [service](<https://devfeed.tech/tags/service.md>)

### AI overview

ReCALM targets catastrophic forgetting in continual service construction across heterogeneous worker data. It extends gradient-free analytic learning beyond single-layer linear models with residual blocks that have closed-form solutions, and uses a cross-recursive least-squares strategy for distributed continual training. The abstract reports that a 20-layer version improves accuracy by 8.85%-9.83% over a single-layer analytic baseline and reduces computational costs by over 94% compared with gradient-based baselines. These are reported results; the excerpt provides no further experimental details.

### Source excerpt

arXiv:2611.19276v1 Announce Type: new Abstract: Crowdsourcing has emerged as a promising paradigm for constructing AI services by harnessing the collective capabilities of distributed workers. As consumers' demands continually evolve, service models must acquire new knowledge while retaining previously learned information. However, due to the inherited reliance on gradients for backpropagation, existing approaches typically suffer from catastrophic forgetting during continual model updates on heterogeneous data. Recent studies have introduced analytic learning, a gradient-free paradigm that achieves invariance through recursive closed-form solutions. Nevertheless, existing analytic methods are typically constrained by single-layer linear networks, which tend to suffer from underfitting due to their inability to perform deep representation learning. To address this issue, we propose ReCALM, a Residual-enhanced Crowdsourced Analytic Learning Mechanism that achieves deep nonlinear invariance for distributed continual service construction. Inspired by the success of ResNet, we develop gradient-free residual blocks for ReCALM with closed-form solutions. To implement distributed continual training for ReCALM, we further devise a cross-recursive strategy via least squares. Both theoretical and experimental analyses verify ReCALM's dual superiority in heterogeneity invariance and deep representation. With 20 residual layers, ReCALM improves accuracy by 8.85%-9.83% over the single-layer analytic baseline, while reducing computational costs by over 94% compared with gradient-based baselines.

## Differentiable Exact Learning

DevFeed: [Differentiable Exact Learning](<https://devfeed.tech/articles/differentiable-exact-learning-137707.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21486>)

Author: Hristo Papazov, Francesco D'Angelo, Nicolas Flammarion

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

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: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Finite-state machine](<https://devfeed.tech/topics/finite-state-machine.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>), [Training AI Models](<https://devfeed.tech/topics/training-ai-models.md>)

Tags: [algorithms](<https://devfeed.tech/tags/algorithms.md>), [arithmetic](<https://devfeed.tech/tags/arithmetic.md>), [controller](<https://devfeed.tech/tags/controller.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-fl](<https://devfeed.tech/tags/cs-fl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [environment](<https://devfeed.tech/tags/environment.md>), [finite-state](<https://devfeed.tech/tags/finite-state.md>), [learning](<https://devfeed.tech/tags/learning.md>), [machine](<https://devfeed.tech/tags/machine.md>), [protocol](<https://devfeed.tech/tags/protocol.md>), [replicate](<https://devfeed.tech/tags/replicate.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [stat-ml](<https://devfeed.tech/tags/stat-ml.md>), [state](<https://devfeed.tech/tags/state.md>), [target](<https://devfeed.tech/tags/target.md>), [traces](<https://devfeed.tech/tags/traces.md>), [tracking](<https://devfeed.tech/tags/tracking.md>), [train](<https://devfeed.tech/tags/train.md>), [training](<https://devfeed.tech/tags/training.md>), [universality](<https://devfeed.tech/tags/universality.md>)

### AI overview

The paper's approach combines a state-tracking controller, an unbounded external environment, and policy-trajectory observations (expert traces showing observations and actions but hiding internal states) to learn controllers that generalize to much longer tasks. Its Differentiable Finite-State Transducers (DFSTs) use multilinear updates, allow log-parallel training, and can emulate arbitrary finite-state controllers. Compact DFSTs trained on tiny datasets show extreme length generalization on arithmetic and universal simulation; the authors report similar results with RNNs and GRUs. They also extract interpretable discrete controllers from trained DFSTs and prove them equivalent to the target algorithms. For Softmax DFSTs, taking the argmax of transition probabilities yields controllers that provably perform binary arithmetic and universal Turing-machine simulation exactly. Exact-learning proofs for neural controllers require assumptions, which the authors say are empirically supported.

### Source excerpt

arXiv:2611.21486v1 Announce Type: cross Abstract: We introduce a framework for differentiable exact learning that couples an unbounded external environment to a state-tracking neural controller, which we train on policy-trajectory observations (PTOs): expert behavior traces that expose observations and actions while hiding internal states. In addition to the learning framework, we introduce Differentiable Finite-State Transducers (DFSTs) which use multilinear updates, support log-parallel training, and emulate arbitrary finite-state controllers, yielding computational universality through environmental interaction. From tiny training datasets, compact DFSTs achieve extreme length generalization on arithmetic and universal-simulation tasks. We replicate the success of DFSTs on RNNs and GRUs, thereby supporting a broader learning recipe: State-Tracking Controller + Unbounded External Environment + PTOs = Extreme Length Generalization. Beyond length generalization, our metric-extraction protocol recovers interpretable discrete controllers from trained DFSTs, which we prove equivalent to the target arithmetic and universal-simulation algorithms. Furthermore, for our Softmax DFST models, simply taking $\mathrm{argmax}$ over the transition probabilities yields discrete controllers which provably exactly perform universal Turing-machine simulation and binary arithmetic. Proving exact learning for the neural controllers requires assumptions, which we empirically support. We make our code, training recipes, model weights, and proof checkers publicly available at https://github.com/fanagor/DEL.

## How Instacart Uses LLMs to Make People Buy More

DevFeed: [How Instacart Uses LLMs to Make People Buy More](<https://devfeed.tech/articles/how-instacart-uses-llms-to-make-people-buy-more-136317.md>)

Original publisher: [Read original article](<https://blog.bytebytego.com/p/how-instacart-uses-llms-to-make-people>)

Author: ByteByteGo

Published: 2026-11-16T16:30:56Z

Content type: article

Language: en

Sources: [ByteByteGo](<https://devfeed.tech/sources/bytebytego.md>)

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [AI-driven information retrieval](<https://devfeed.tech/topics/ai-driven-information-retrieval.md>), [Retrieval Augmented Generation (RAG)](<https://devfeed.tech/topics/retrieval-augmented-generation-rag.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>)

Tags: [a-b-testing](<https://devfeed.tech/tags/a-b-testing.md>), [a100](<https://devfeed.tech/tags/a100.md>), [adapter](<https://devfeed.tech/tags/adapter.md>), [advertising](<https://devfeed.tech/tags/advertising.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-software](<https://devfeed.tech/tags/agentic-software.md>), [agents](<https://devfeed.tech/tags/agents.md>), [api](<https://devfeed.tech/tags/api.md>), [authorization](<https://devfeed.tech/tags/authorization.md>), [autoscaling](<https://devfeed.tech/tags/autoscaling.md>), [behaviour](<https://devfeed.tech/tags/behaviour.md>), [brand](<https://devfeed.tech/tags/brand.md>), [business-context](<https://devfeed.tech/tags/business-context.md>), [cache](<https://devfeed.tech/tags/cache.md>), [instacart](<https://devfeed.tech/tags/instacart.md>), [llms](<https://devfeed.tech/tags/llms.md>), [product-search](<https://devfeed.tech/tags/product-search.md>), [search](<https://devfeed.tech/tags/search.md>), [shopping](<https://devfeed.tech/tags/shopping.md>)

### AI overview

Instacart combines purchase-history candidates, LLM interpretation, and embedding-based relevance checks to map shopping queries to catalog categories. For rewrites, it separates synonyms, broader terms, and substitutes; the article reports coverage above 95% and precision above 90%. For semantic tags, an offline LLM teacher creates validated examples and cached results, while a LoRA-fine-tuned Llama-3 8B handles uncached live queries. Merging the adapter and moving from A100 to H100 reduced reported serving latency from nearly 700 ms to a 300 ms target. FP8 offered another 10% latency reduction but slightly lowered recall, so Instacart deployed the unquantized model. Caching meant only 2% of queries needed live inference. The article reports A/B test results for tail queries: 6% lower average scroll depth, 50% fewer related complaints, and marginal latency increase.

### Source excerpt

In this article, we'll look at how Instacart used LLMs to build a product search that could accurately understand a user's shopping intention.

## Local LLMs Handle Textbook RAG but Fail on New Research Papers

DevFeed: [Local LLMs Handle Textbook RAG but Fail on New Research Papers](<https://devfeed.tech/articles/local-llms-handle-textbook-rag-but-fail-on-new-research-papers-135976.md>)

Original publisher: [Read original article](<https://towardsdatascience.com/local-llms-handle-textbook-rag-but-fail-on-new-research-papers/>)

Author: Waleed Esmail

Published: 2026-11-16T12:00:00Z

Content type: article

Language: en

Sources: [Towards Data Science](<https://devfeed.tech/sources/towards-data-science.md>)

Topics: [Local AI](<https://devfeed.tech/topics/local-ai.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>), [Hallucination detection](<https://devfeed.tech/topics/hallucination-detection.md>), [gpt-oss](<https://devfeed.tech/topics/gpt-oss.md>)

Tags: [large-language-model](<https://devfeed.tech/tags/large-language-model.md>), [large-language-models](<https://devfeed.tech/tags/large-language-models.md>), [llms](<https://devfeed.tech/tags/llms.md>), [local](<https://devfeed.tech/tags/local.md>), [local-llm-inference](<https://devfeed.tech/tags/local-llm-inference.md>), [local-llms](<https://devfeed.tech/tags/local-llms.md>), [papers](<https://devfeed.tech/tags/papers.md>), [rag](<https://devfeed.tech/tags/rag.md>), [rag-retrieval-augmented-generation](<https://devfeed.tech/tags/rag-retrieval-augmented-generation.md>)

### AI overview

In this evaluation, five setups scored 12-20 of 20 on textbook questions but only 0-9 of 15 on a paper published that year, despite identical retrieval and prompts. The author concludes that small local models handle familiar material better than new research, where they must rely on supplied passages. Grading answers alongside retrieved passages showed many errors came from missing evidence: figure descriptions raised figure-question results from 1/10 to 9/10, chapter summary trees raised overview results from 1/8 to 8/8, and neighboring chunks improved complete retrieval from 16/20 to 19/20. Quantization did not explain the gap. Qwen3.8-27B was close to Opus in the reported comparison at lower cost, but took about three minutes per answer. The author recommends writing gold questions first, separating familiar from new material, inspecting passages when grading, improving retrieval before changing models, and routing figures, web reading, and derivations to stronger models. Results are limited by a 35-question evaluation and grader drift.

### Source excerpt

Five setups, one blind grading pack: textbook scores ran 12 to 20 of 20, but a paper published this year dropped them to 0 to 9 of 15. The post Local LLMs Handle Textbook RAG but Fail on New Research Papers appeared first on Towards Data Science.

## Uncertainty-Aware Continual Learning for 2D XRD Interpretation Using Simulated and Experimental Data

DevFeed: [Uncertainty-Aware Continual Learning for 2D XRD Interpretation Using Simulated and Experimental Data](<https://devfeed.tech/articles/uncertainty-aware-continual-learning-for-2d-xrd-interpretation-using-simulated-and-experimental-data-135794.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.17547>)

Author: Zihan Wang, Hyun Sang Park, Ali Rachidi, David Elbert, Todd C. Hufnagel, Wei Chen

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

Content type: article

Language: en

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

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [2d](<https://devfeed.tech/tags/2d.md>), [alignment](<https://devfeed.tech/tags/alignment.md>), [classification](<https://devfeed.tech/tags/classification.md>), [cnn](<https://devfeed.tech/tags/cnn.md>), [continual-learning](<https://devfeed.tech/tags/continual-learning.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [experimental](<https://devfeed.tech/tags/experimental.md>), [knowledge-distillation](<https://devfeed.tech/tags/knowledge-distillation.md>), [planning](<https://devfeed.tech/tags/planning.md>), [proof-of-concept](<https://devfeed.tech/tags/proof-of-concept.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [stat-ml](<https://devfeed.tech/tags/stat-ml.md>), [uncertainty](<https://devfeed.tech/tags/uncertainty.md>), [uncertainty-estimation](<https://devfeed.tech/tags/uncertainty-estimation.md>)

### AI overview

The proof-of-concept combines maximum mean discrepancy-based domain adaptation to reduce simulation-to-experiment mismatch, a CNN-SNGP for predictive uncertainty, and replay with knowledge distillation for sequential updates to 2D XRD classifiers. Tested on simulated and previously collected experimental patterns for single-phase body-centered cubic, face-centered cubic, and hexagonal close-packed structures arranged in predefined sequential blocks, it showed accurate classification across domains, improved domain alignment, and less forgetting than naïve fine-tuning. Uncertainty was evaluated on unfamiliar and noisy inputs. Experimental planning, instrument control, and closed-loop deployment remain future work.

### Source excerpt

arXiv:2611.17547v1 Announce Type: cross Abstract: Autonomous experimentation systems require characterization models that can interpret heterogeneous measurements and maintain predictive performance as data accumulate. Developing such models involves addressing domain mismatch between data sources, overconfident predictions on unfamiliar inputs, and catastrophic forgetting during sequential updates. In this work, we present a proof-of-concept framework for uncertainty-aware continual interpretation of two-dimensional X-ray diffraction (XRD) data. The framework integrates established components: maximum mean discrepancy-based domain adaptation to reduce simulation-to-experiment discrepancies, a convolutional neural network with a spectral-normalized neural Gaussian process (CNN--SNGP) for predictive uncertainty estimation, and replay with knowledge distillation for sequential model updating. We evaluate the framework using simulated and previously-collected experimental data for crystal structure classification of single-phase body-centered cubic, face-centered cubic, and hexagonal close-packed patterns. Data are organized into predefined sequential blocks to assess adaptation and retention of previously learned knowledge. Within this setting, the results show accurate classification across simulated and experimental data, improved alignment between the two domains, and reduced forgetting compared with na\"ive fine-tuning. Predictive uncertainty is also evaluated on unfamiliar and noisy inputs. The contribution is the integration and evaluation of these capabilities for sequential XRD interpretation. The framework provides a basis for future integration into autonomous experimentation systems, with experimental planning, instrument control, and closed-loop deployment remaining areas for future development.

## LECAT: Measuring LLM Category Learning from Pretraining Data

DevFeed: [LECAT: Measuring LLM Category Learning from Pretraining Data](<https://devfeed.tech/articles/lecat-measuring-llm-category-learning-from-pretraining-data-135197.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18419>)

Author: Jimin Mun, Maarten Sap, Yanai Elazar

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

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>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>), [Open Source Models & Datasets](<https://devfeed.tech/topics/open-source-models-datasets.md>)

Tags: [acquisition](<https://devfeed.tech/tags/acquisition.md>), [context](<https://devfeed.tech/tags/context.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [exceptions](<https://devfeed.tech/tags/exceptions.md>), [expression](<https://devfeed.tech/tags/expression.md>), [framework](<https://devfeed.tech/tags/framework.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [learning](<https://devfeed.tech/tags/learning.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llms](<https://devfeed.tech/tags/llms.md>), [measuring](<https://devfeed.tech/tags/measuring.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [pairs](<https://devfeed.tech/tags/pairs.md>), [pmi](<https://devfeed.tech/tags/pmi.md>), [pretraining](<https://devfeed.tech/tags/pretraining.md>), [robust](<https://devfeed.tech/tags/robust.md>), [text](<https://devfeed.tech/tags/text.md>)

### AI overview

LeCat evaluates category knowledge as memorization, generalization to unseen instances, and flexible use in context, then traces how training text expresses that knowledge. In tests with open-source pretrained models and 199 curated category-feature pairs, stronger context and category-feature co-occurrence (PMI) were associated with better learning in larger models, while verbatim rule-like statements diminished flexible use. A controlled 68M model trained from scratch confirmed that verbatim repetition reduced flexible learning; mixing diverse forms of knowledge expression produced more robust category learning. Contextual flexibility remained difficult. These findings are based on the described evaluations and training setup.

### Source excerpt

arXiv:2611.18419v1 Announce Type: new Abstract: Humans know that airplanes are fast, that this extends to red airplanes, but that it does not hold for broken airplanes. How do LLMs acquire category knowledge that humans learn and apply effortlessly? We introduce LeCat, a two-part framework to analyze LLM's Learning of Category knowledge. First, we use LeCat to evaluate models on their ability to learn category knowledge, distinguishing between three cases: memorization of in-domain category knowledge, generalization to unseen instances, and ability to flexibly use knowledge in appropriate contexts. Then, we use LeCat to trace how this knowledge can be learned in pretraining from text that may encode category knowledge in various ways, such as verbatim repetition or exceptions, which we refer to as forms of "knowledge expression". We first assess open-source pretrained models using 199 curated category-feature pairs and analyze their training data. Our findings suggest that strengthening context and category-feature co-occurrence likelihood (PMI) improve learning for larger models and verbatim rule-like statements diminish flexible category learning. To systematically isolate these effects, we then train a 68M model from scratch with precise control over the categorical information each sees in their training data, confirming that verbatim repetition reduces flexible learning but mixture of diverse knowledge expressions lead to more robust category learning. Together, our findings highlight that while contextual flexibility is persistently difficult for models, how category knowledge is expressed in pretraining greatly impacts models' acquisition of robust category knowledge.

## Federated Physics-Informed Thermal Field Prediction Across Unseen Geometries in Distributed Additive Manufacturing

DevFeed: [Federated Physics-Informed Thermal Field Prediction Across Unseen Geometries in Distributed Additive Manufacturing](<https://devfeed.tech/articles/federated-physics-informed-thermal-field-prediction-across-unseen-geometries-in-distributed-additive-manufacturing-135832.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.17794>)

Author: Amirul Islam Saimon (James), Zhenyu (James), Kong

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

Content type: article

Language: en

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

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [distributed-training](<https://devfeed.tech/topics/distributed-training.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [2d](<https://devfeed.tech/tags/2d.md>), [additive-manufacturing](<https://devfeed.tech/tags/additive-manufacturing.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [federated](<https://devfeed.tech/tags/federated.md>), [federated-learning](<https://devfeed.tech/tags/federated-learning.md>), [field](<https://devfeed.tech/tags/field.md>), [inverse-design](<https://devfeed.tech/tags/inverse-design.md>), [manufacturing](<https://devfeed.tech/tags/manufacturing.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [physics-informed](<https://devfeed.tech/tags/physics-informed.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [thermal](<https://devfeed.tech/tags/thermal.md>), [uncertainty-quantification](<https://devfeed.tech/tags/uncertainty-quantification.md>)

### AI overview

Fed-PIGANO trains physics-informed geometry-aware neural operator models at separate additive manufacturing sites and aggregates their encoders and prediction networks without sharing proprietary part geometries. On a 2D steady-state metal AM benchmark, geometries from five polygonal families were split across three sites in four scenarios, from matching to strongly different family mixes. Across seven replications, mean relative L2 error was 1.83%-2.04% in all 12 scenario-site cases, versus as high as 20.46% for local-only training; PDE residuals were also consistently lower. Accuracy was comparable to a centralized model trained on pooled geometries. The evidence is limited to this benchmark and supports potential use for predicting thermal fields on new designs, including shape optimization, inverse design, and uncertainty quantification.

### Source excerpt

arXiv:2611.17794v1 Announce Type: new Abstract: In additive manufacturing (AM), each change in part geometry requires a new thermal simulation, which makes design evaluation over many candidates costly. Geometry-aware models reduce this cost by predicting thermal fields on unseen parts with a single learned model, but they assume that diverse geometries are available centrally. This assumption is restrictive when each manufacturing site holds only a limited set of proprietary part designs that it cannot share. We propose Fed-PIGANO, a federated framework that trains physics-informed geometry-aware neural operator (PI-GANO) models at individual sites and aggregates their geometry encoders and prediction networks without sharing part geometries. We evaluate the framework on a 2D steady-state thermal benchmark for metal AM, with geometries from five polygonal families distributed across three sites under four scenarios that range from identical to strongly different family compositions. Over seven replications, Fed-PIGANO achieves mean relative L2 errors of 1.83% to 2.04% across all 12 scenario-site cases, compared with up to 20.46% for local-only training, and consistently yields lower PDE residuals. Its prediction accuracy is comparable to that of a centralized model trained on the pooled geometries of all sites. These results demonstrate that federated learning lets manufacturing sites benefit from each other's geometries without sharing them, enabling rapid thermal prediction on new part designs for tasks such as shape optimization, inverse design, and uncertainty quantification.

## PATTERN Can Be Solved from Memory: Testing Graph Neural Networks on Unseen Patterns

DevFeed: [PATTERN Can Be Solved from Memory: Testing Graph Neural Networks on Unseen Patterns](<https://devfeed.tech/articles/pattern-can-be-solved-from-memory-testing-graph-neural-networks-on-unseen-patterns-135729.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.15020>)

Author: Amin Najafgholizadeh, Mohsen Afsharchi

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

Content type: article

Language: en

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

Topics: [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Graph Mining](<https://devfeed.tech/topics/graph-mining.md>)

Tags: [benchmark](<https://devfeed.tech/tags/benchmark.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [generator](<https://devfeed.tech/tags/generator.md>), [graph](<https://devfeed.tech/tags/graph.md>), [graph-neural-networks](<https://devfeed.tech/tags/graph-neural-networks.md>), [inference](<https://devfeed.tech/tags/inference.md>), [logistic-regression](<https://devfeed.tech/tags/logistic-regression.md>), [lookup](<https://devfeed.tech/tags/lookup.md>), [pattern](<https://devfeed.tech/tags/pattern.md>), [patterns](<https://devfeed.tech/tags/patterns.md>), [roc-auc](<https://devfeed.tech/tags/roc-auc.md>), [split](<https://devfeed.tech/tags/split.md>), [statistics](<https://devfeed.tech/tags/statistics.md>), [training](<https://devfeed.tech/tags/training.md>), [validation](<https://devfeed.tech/tags/validation.md>)

### AI overview

PATTERN's official split reuses 100 planted patterns across training and test graphs, allowing a lookup from training labels to reach 96.6% class-weighted accuracy without fitting a model; the same lookup scores 50.6% on unseen patterns, while a matcher using the generator reaches 99.2% on the official split. Of five retrained graph networks, GRASS and PPGT fall from the top two to the bottom two on unseen patterns, each losing 8.1 points even with pattern sizes matched. Logistic regression on two-hop node statistics exceeds all five networks' unseen-pattern runs. CKGConv's advantage follows the training pattern library and disappears when each training and validation graph gets a new pattern. The authors say the official split measures finding familiar patterns in new graphs and recommend reporting the same checkpoint's unseen-pattern score as well. They also report generator-based inference results on CLUSTER and Minesweeper, with the latter based on four of eight planned chains and none converging.

### Source excerpt

arXiv:2610.15020v2 Announce Type: replace Abstract: PATTERN asks graph neural networks to find a planted subgraph in a new host graph. The hosts change, but the benchmark draws just 100 patterns and reuses them across training and test splits. A lookup that extracts these patterns from the training labels achieves 96.6% class-weighted accuracy on the official test split, above every published result, without fitting a model or consulting the generator. On unseen patterns, the same lookup scores 50.6%. A matcher that also uses the generator reaches 99.2% on the official split. Among five graph networks we retrained, GRASS and PPGT go from the top two on the official split to the bottom two on unseen patterns. Each loses 8.1 percentage points even when familiar and unseen patterns are matched in size. Logistic regression on two-hop node statistics scores above every run of the five networks on unseen patterns. For CKGConv, the advantage follows the training patterns: it reappears with each of three independently drawn pattern libraries and disappears when every training and validation graph gets a new pattern. Generator-based inference also scores 82.3% on CLUSTER, compared with a published best of 81.6%. On Minesweeper's released board, four of eight planned chains produced samples within the time budget; none converged, but their pooled predictions reached 99.86% ROC AUC, above the published best of 99.55%. PATTERN's official split tests its intended task of finding familiar patterns in new graphs. We recommend reporting the same checkpoint's score on unseen patterns alongside it.

## Model Annealing: Data-Efficient Learning through Review and Consolidation

DevFeed: [Model Annealing: Data-Efficient Learning through Review and Consolidation](<https://devfeed.tech/articles/model-annealing-data-efficient-learning-through-review-and-consolidation-135075.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.17524>)

Author: Ali Mohammadi Ruzbahani, Rezvan Abouzadeh

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: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [machine learning overfitting](<https://devfeed.tech/topics/machine-learning-overfitting.md>), [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>), [model architecture](<https://devfeed.tech/topics/model-architecture.md>), [Inference](<https://devfeed.tech/topics/inference.md>)

Tags: [1-3](<https://devfeed.tech/tags/1-3.md>), [adversarial](<https://devfeed.tech/tags/adversarial.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [budget](<https://devfeed.tech/tags/budget.md>), [confidence](<https://devfeed.tech/tags/confidence.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [distillation](<https://devfeed.tech/tags/distillation.md>), [dreams](<https://devfeed.tech/tags/dreams.md>), [inference](<https://devfeed.tech/tags/inference.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [learning](<https://devfeed.tech/tags/learning.md>), [memory](<https://devfeed.tech/tags/memory.md>), [mixer](<https://devfeed.tech/tags/mixer.md>), [mlp](<https://devfeed.tech/tags/mlp.md>), [model](<https://devfeed.tech/tags/model.md>), [network](<https://devfeed.tech/tags/network.md>), [predictions](<https://devfeed.tech/tags/predictions.md>), [raises](<https://devfeed.tech/tags/raises.md>), [review](<https://devfeed.tech/tags/review.md>)

### AI overview

Model annealing uses the remaining training budget for repeated review: after each learning-rate reheating and cooldown, it stores that model's predictions, then trains the next student on labels, softened memory predictions, and adversarial examples found where student and memory disagree. The final student is deployed, so the prediction memory adds no inference cost. Across the tested CIFAR-100, Tiny-ImageNet, and Imagewoof settings, AdamW-trained ConvNeXt, ViT, and MLP-Mixer models beat cross-entropy top-1 accuracy in 28 of 33 configurations, averaging 1.3 points, and matched snapshot distillation in accuracy while improving likelihood, calibration, and robustness to corruptions and small adversarial perturbations. Results differ for the SGD-trained ResNet-18: accuracy fell, though small-perturbation robustness improved. Ablations attribute gains to relearning and losses to reheating; memory and adversarial dreams also curb the overconfidence that snapshot-only relearning compounds with little data.

### Source excerpt

arXiv:2611.17524v1 Announce Type: new Abstract: A small labelled dataset is memorized long before the training budget runs out, and the one-hot labels that remain to be fitted then mostly sharpen confidence. Model annealing spends the rest of the budget on review and consolidation, relearning from the network's own knowledge. Training is divided into episodes whose learning rate is raised again and cooled, every cooled state joins a memory of past selves, and each new episode learns from the labels, from the tempered predictions of this memory, and from adversarial dreams sought where student and memory disagree most. The memory averages predictions rather than weights, and the deployed network is the last student, so inference costs nothing extra. Across CIFAR-100, Tiny-ImageNet and Imagewoof, four architecture families and 5\% to 100\% of the labels, model annealing raises the top-1 accuracy of the AdamW-trained ConvNeXt, ViT and MLP-Mixer over cross-entropy training in 28 of 33 configurations, by 1.3 points on average, and is on par in accuracy with snapshot distillation, the closest competitor. On these families it is better than both in likelihood, calibration and robustness to corruptions and small adversarial perturbations. For an SGD-trained ResNet-18 it lowers accuracy but still improves robustness to small adversarial perturbations. Separating the reheated schedule from relearning shows that relearning helps every family and that the losses come from reheats, which undo most of this network's fit to its training set. How early the accuracy of a cross-entropy run saturates largely separates the AdamW families from ResNet-18 but does not predict the gain within them. With little data, relearning from the latest snapshot without dreams, as snapshot distillation does, compounds overconfidence in the AdamW families from episode to episode. The memory and the dreams each curb most of this drift, and together they remove it.

## Where Tree Ensembles Decide: Structural Local Explanation with Decision Predicate Graphs

DevFeed: [Where Tree Ensembles Decide: Structural Local Explanation with Decision Predicate Graphs](<https://devfeed.tech/articles/where-tree-ensembles-decide-structural-local-explanation-with-decision-predicate-graphs-135893.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18382>)

Author: Sylvio Barbon Junior, Luca Longo

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

Content type: article

Language: en

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

Topics: [generalization in machine learning](<https://devfeed.tech/topics/generalization-in-machine-learning.md>)

Tags: [class](<https://devfeed.tech/tags/class.md>), [critical](<https://devfeed.tech/tags/critical.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [decision-rules](<https://devfeed.tech/tags/decision-rules.md>), [explain](<https://devfeed.tech/tags/explain.md>), [graph](<https://devfeed.tech/tags/graph.md>), [local](<https://devfeed.tech/tags/local.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [model](<https://devfeed.tech/tags/model.md>), [predicate](<https://devfeed.tech/tags/predicate.md>), [tabular-data](<https://devfeed.tech/tags/tabular-data.md>), [trees](<https://devfeed.tech/tags/trees.md>)

### AI overview

The local Decision Predicate Graph (DPG) records the executed tree rules behind a sample's prediction, with class support that reproduces it. It identifies a critical predicate: an executed threshold whose crossing is estimated to move the model toward a chosen class, along with the trees sharing that threshold. Across 16 datasets and four ensemble families, this approach separated differently predicted neighbours better than random or nearest-threshold predicates. Crossing one selected threshold recovered the true class in 46-69% of wrong predictions, compared with 29-50% when applying the same rule to the feature ranked first by TreeSHAP; planted-fault thresholds were found in 46-71% of resulting errors, versus 6-21% at random. Training-free edits did not reduce error on unseen data, so the authors present the method as complementary to attributions and counterfactuals.

### Source excerpt

arXiv:2611.18382v1 Announce Type: new Abstract: Tree ensembles are among the most accurate models for tabular data, yet the computation behind a single prediction is spread over hundreds of decision rules. Feature attributions and counterfactuals explain such predictions through the influence of features or a change of the input, leaving open which of the thresholds applied by the trees a prediction responds to, while structural explanations compress these rules and lose the trees behind them. To address this gap, we propose a local Decision Predicate Graph (local DPG), which gathers the rules executed for a sample into a verifiable record of the computation, with a class support that reproduces every explained prediction. From these rules we select the critical predicate, the executed predicate whose crossing is estimated to move the model most strongly toward a chosen class, which offers the user a concrete threshold and the trees that share it, without an external explainer. Across 16 datasets and four ensemble families, it separates differently predicted neighbours better than random and nearest-threshold predicates. A single crossing recovers the true class in 46-69% of wrong predictions, against 29-50% when the same rule is applied to the feature ranked first by TreeSHAP, a measure close to the selection criterion, and on planted faults it finds the faulty threshold in 46-71% of the resulting errors, against 6-21% at random. Although training-free edits of these predicates did not reduce the error on unseen data, the critical predicate complements attributions and counterfactuals by naming where a prediction can be changed.

[Next page](<https://devfeed.tech/topics/generalization-in-machine-learning.md?cursor=WyIyMDI2LTExLTE2VDA1OjAwOjAwKzAwOjAwIiwgIjE0MGU1NjAwLTE4ZTYtNGZhYS05NDIyLTVmNWU1YmE3OWJiNCJd>)