# Advanced AI workflows

Advanced AI workflows are end-to-end processes for building AI solutions, including data preparation, model training, and deployment or inference.

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

## LoongForge-Embodied: A Unified Framework for Training Heterogeneous Embodied Models

DevFeed: [LoongForge-Embodied: A Unified Framework for Training Heterogeneous Embodied Models](<https://devfeed.tech/articles/loongforge-embodied-a-unified-framework-for-training-heterogeneous-embodied-models-145515.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25460>)

Author: LoongForge Team

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

Content type: article

Language: en

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

Topics: [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Machine Perception](<https://devfeed.tech/topics/machine-perception.md>), [Distributed Systems & Parallel Computing](<https://devfeed.tech/topics/distributed-systems-parallel-computing.md>)

Tags: [acceleration](<https://devfeed.tech/tags/acceleration.md>), [architectures](<https://devfeed.tech/tags/architectures.md>), [concurrency](<https://devfeed.tech/tags/concurrency.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [data](<https://devfeed.tech/tags/data.md>), [data-infrastructure](<https://devfeed.tech/tags/data-infrastructure.md>), [ddp](<https://devfeed.tech/tags/ddp.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [execution](<https://devfeed.tech/tags/execution.md>), [fp8](<https://devfeed.tech/tags/fp8.md>), [framework](<https://devfeed.tech/tags/framework.md>), [fsdp](<https://devfeed.tech/tags/fsdp.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [i-o](<https://devfeed.tech/tags/i-o.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [layer](<https://devfeed.tech/tags/layer.md>), [memory](<https://devfeed.tech/tags/memory.md>), [native](<https://devfeed.tech/tags/native.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

LoongForge-Embodied uses a shared abstraction layer to make training infrastructure reusable across heterogeneous VLA and WAM architectures without constraining model-specific execution needs. The native-PyTorch framework combines shared and model- or hardware-specific optimizations for computation, memory, data I/O, execution, and distributed communication, including FP8 communication for DDP and FSDP. It also includes infrastructure for synthesizing robot data from egocentric data and concurrent simulator evaluation. Across representative models, the paper reports 1.47x-4.38x higher end-to-end training throughput than corresponding open-source baselines and 4.1x-23.2x faster evaluation under the same protocols.

### Source excerpt

arXiv:2611.25460v1 Announce Type: new Abstract: Training Vision-Language-Action (VLA) models and World-Action Models (WAMs) is computationally expensive, while their heterogeneous architectures and fragmented training stacks make system optimizations difficult to reuse across models. To address these challenges, we present LoongForge-Embodied, a native-PyTorch framework that unifies and accelerates the training of heterogeneous embodied models. At its core, LoongForge-Embodied introduces a common abstraction layer that decouples model-specific implementations from shared training infrastructure, enabling cross-model reuse without constraining individual architectures or execution requirements. Buildingon this unified infrastructure, the framework combines reusable acceleration mechanisms with model-specific and hardware-aware optimization strategies across computation, execution, memory, data I/O, and distributed communication, including FP8-based communication optimizations for DDP and FSDP. Beyond training acceleration, LoongForge-Embodied integrates a unified data infrastructure with a pipeline for synthesizing robot data from egocentric data, and an efficient evaluation framework that supports concurrency and optimized computation in simulator. Experiments across representative VLA and WAM models demonstrate 1.47x-4.38x end-to-end training throughput improvements over their corresponding open-source baselines, as well as 4.1x-23.2x faster evaluation under the same protocols.

## Sage-Evolve-Preview:Adaptive Observation for Post-Training Self-Improvement

DevFeed: [Sage-Evolve-Preview:Adaptive Observation for Post-Training Self-Improvement](<https://devfeed.tech/articles/sage-evolve-preview-adaptive-observation-for-post-training-self-improvement-145635.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.26132>)

Author: Yinhao Tang, Youqing Fang, Yanan Sun, Xintian Shen, Hanyu Sun, Wenran Liu, Yue Wu, Weiming Zhang, Bin Liu, Wenwei Zhang

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

Content type: article

Language: en

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

Topics: [post-training](<https://devfeed.tech/topics/post-training.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>)

Tags: [adaptive](<https://devfeed.tech/tags/adaptive.md>), [agent](<https://devfeed.tech/tags/agent.md>), [anchor](<https://devfeed.tech/tags/anchor.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [budget](<https://devfeed.tech/tags/budget.md>), [controller](<https://devfeed.tech/tags/controller.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [diagnostic](<https://devfeed.tech/tags/diagnostic.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [evolution](<https://devfeed.tech/tags/evolution.md>), [harness](<https://devfeed.tech/tags/harness.md>), [independent](<https://devfeed.tech/tags/independent.md>), [long-context](<https://devfeed.tech/tags/long-context.md>), [post-training](<https://devfeed.tech/tags/post-training.md>), [progress](<https://devfeed.tech/tags/progress.md>), [rsi](<https://devfeed.tech/tags/rsi.md>), [score](<https://devfeed.tech/tags/score.md>), [self-improvement](<https://devfeed.tech/tags/self-improvement.md>), [tasks](<https://devfeed.tech/tags/tasks.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

AdaObs addresses plateaus in post-training recursive self-improvement by adapting what the training loop observes while keeping its anchor objective fixed. At a plateau, its controller sharpens diagnostics, raises task difficulty, and adds related tasks; validated gains feed into training, and checkpoints remain scored against the anchor. Across eight anchors in long-context reading comprehension and instruction following, mean scores rose from 66.4 to 73.7 under a matched data budget, outperforming the FixedObs anchor-only baseline. The evolved harness also improved the seven-task average on independent PostTrainBench across two agent backends. The evidence is limited to the reported benchmarks and settings.

### Source excerpt

arXiv:2611.26132v1 Announce Type: new Abstract: Post-training recursive self-improvement (RSI) iteratively improves a target capability, operationalized by a fixed anchor benchmark, through repeated intervention, training, and evaluation. Yet most loops use the anchor's aggregate score as their sole observation, so early gains give way to stalled progress and repetitive exploration. When this anchor-only feedback ceases to reveal actionable gaps, a plateau may reflect an exhausted observation rather than an exhausted capability. We introduce adaptive observation (AdaObs), an observation-side mechanism of harness evolution that keeps the anchor objective fixed while adapting the active observation. At a plateau, a controller refines diagnostic granularity, escalates difficulty, and adds related tasks; validated increments are fused back into training, while every checkpoint remains evaluated against the anchor. Across eight independently optimized anchors in long-context reading comprehension and instruction following, AdaObs raises the mean score from $66.4$ to $73.7$ under a matched data budget, outperforming FixedObs, the anchor-only baseline. The evolved harness further improves the seven-task average on the independent PostTrainBench across two agent backends. These results identify observation adaptation as a missing mechanism of harness evolution for sustaining post-training RSI.

## From AI Surrogate to Engineering Tool: Toward Production Use of Fiber-Orientation Prediction in Crash CAE

DevFeed: [From AI Surrogate to Engineering Tool: Toward Production Use of Fiber-Orientation Prediction in Crash CAE](<https://devfeed.tech/articles/from-ai-surrogate-to-engineering-tool-toward-production-use-of-fiber-orientation-prediction-in-crash-cae-145619.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25991>)

Author: Osamu Ito, Yoshikazu Nakagawa, Tomoaki Tatsukawa

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

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>), [implementation](<https://devfeed.tech/topics/implementation.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Tool](<https://devfeed.tech/topics/tool.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [cae](<https://devfeed.tech/tags/cae.md>), [crash](<https://devfeed.tech/tags/crash.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [fiber](<https://devfeed.tech/tags/fiber.md>), [orientation](<https://devfeed.tech/tags/orientation.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [production](<https://devfeed.tech/tags/production.md>), [tool](<https://devfeed.tech/tags/tool.md>)

### AI overview

The integrated fiber-orientation workflow cuts preparation time for crash CAE from about 25 hours to 0.1 hours per part and can be run by crash engineers without molding-specialist support. Its GUI combines gate-setting support for unknown gates, voxelization, surrogate prediction, mapping to the crash model, and report generation. Gate support achieved 90% gate-count accuracy and 11% mean position error relative to flow length. In component crash analysis, predicted orientation reduced absolute peak-force error from 19.0% with an isotropic baseline to 11.4%, and predicted fracture stayed within the region observed in testing. Following staged validation and expert approval in March 2025, the system was transferred to Honda's crash-safety group and is in initial, limited use for directed fracture-prediction scenarios. The excerpt emphasizes that workflow integration, data conversion, and reporting helped make the surrogate usable in design.

### Source excerpt

arXiv:2611.25991v1 Announce Type: new Abstract: Crash simulations of glass-fiber-reinforced plastic parts are highly sensitive to the fiber-orientation field supplied to the material model. In conventional workflows, that field comes from resin-flow analysis, which requires molding-specialist support and takes about 25 hours of work per part. We previously proposed a surrogate model that addresses this bottleneck by predicting the orientation field in about 3 minutes. This paper reports the implementation, validation, and operational design that turned the model into an engineering tool for Honda design workflows. We present staged validation from 2D plates to production parts, evaluation against a predefined engineering angular-error criterion, and the expert review that approved design use in March 2025; the system has since been transferred to the crash-safety group and is in initial, limited use in directed fracture-prediction scenarios. The developed system combines injection-gate setting support for cases in which gates are unknown (90% gate-count accuracy, mean position error 11% of flow length), voxelization, surrogate inference, automatic mapping to the crash CAE model, and report generation in a single GUI. The complete orientation-preparation workflow takes about 0.1 hours, a 99.6% reduction from the conventional 25-hour flow-analysis route, and can be run directly by crash CAE engineers without molding-specialist support. As evidence of practical impact, component-level crash analysis using the predicted orientation field reduced the absolute peak-force error from 19.0% with an isotropic baseline to 11.4%, while confining predicted fracture to the test-observed region. The deployment lesson is that a surrogate model alone is not enough for practical adoption: gate-setting support, automatic data conversion, and shareable reporting are what make the model usable as a design tool.

## Explainable AI for Solar PV Power Forecasting: From Model Selection to Trustworthy, Reproducible Deployment

DevFeed: [Explainable AI for Solar PV Power Forecasting: From Model Selection to Trustworthy, Reproducible Deployment](<https://devfeed.tech/articles/explainable-ai-for-solar-pv-power-forecasting-from-model-selection-to-trustworthy-reproducible-deployment-145603.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25894>)

Author: Nuzhat Khan, Muhammad Paend Bakht, Ab Al-Hadi Ab Rahman, Fabiano Pallonetto, Amy Fahy, Mohamed Khalafalla Hassan, Paul Stynes, Muhammad Anique Aslam, Terence O Donnell, Mohammad Riadh

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

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>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Hyperparameter optimization](<https://devfeed.tech/topics/hyperparameter-optimization.md>), [Feature Engineering](<https://devfeed.tech/topics/feature-engineering.md>), [tinyml](<https://devfeed.tech/topics/tinyml.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [climate](<https://devfeed.tech/tags/climate.md>), [cross-validation](<https://devfeed.tech/tags/cross-validation.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [deep](<https://devfeed.tech/tags/deep.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [edge](<https://devfeed.tech/tags/edge.md>), [explainable-ai](<https://devfeed.tech/tags/explainable-ai.md>), [feature](<https://devfeed.tech/tags/feature.md>), [features](<https://devfeed.tech/tags/features.md>), [forecasting](<https://devfeed.tech/tags/forecasting.md>), [foundation](<https://devfeed.tech/tags/foundation.md>), [foundation-models](<https://devfeed.tech/tags/foundation-models.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [model-selection](<https://devfeed.tech/tags/model-selection.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [pv](<https://devfeed.tech/tags/pv.md>), [shap](<https://devfeed.tech/tags/shap.md>), [solar](<https://devfeed.tech/tags/solar.md>)

### AI overview

On a one-year test of an openly available 16.32 kWp solar plant in Porto, Portugal, forecasting one day ahead from a 4 km weather grid, the best models achieved 11.6% nRMSE--about half the error of smart persistence. Input-feature choice affected accuracy more than the choice among strong algorithms, and shuffled cross-validation was mildly optimistic. Conformal calibration raised coverage of nominal 80% prediction intervals from 64.5% to 81.7%. SHAP rankings were stable across retraining periods but shifted when correlated inputs were removed, so they describe model behavior rather than the plant itself. The paper presents a reproducible workflow combining automated model search and explainable AI; the supplied text is an abstract and does not provide implementation details.

### Source excerpt

arXiv:2611.25894v1 Announce Type: new Abstract: Machine learning (ML) is now a standard tool for forecasting the output of solar photovoltaic (PV) plants, yet published accuracy figures are often hard to compare and harder to trust. Many studies validate with random splits, report no reference forecast, and present feature attributions without saying what they can and cannot show. This paper reviews the main families of ML methods for PV power forecasting and resource assessment, from tree ensembles to deep and foundation models, and sets out a reproducible workflow that combines automated model search with explainable AI. The workflow is then tested on an openly available dataset: a 16.32 kWp plant in Porto, Portugal, forecast one day ahead from a 4 km numerical weather prediction grid. Over a full test year, the best models reach a normalised root mean square error (nRMSE) of 11.6% of installed capacity, about half the error of smart persistence. The choice of input features changed accuracy far more than the choice among strong algorithms, and shuffled cross-validation was mildly optimistic. After conformal calibration, the nominal 80% prediction intervals covered 81.7% of observations, against 64.5% before. SHAP rankings stayed stable when the model was retrained on different periods but shifted when correlated inputs were removed, so attributions should be read as properties of a model rather than of the plant. The paper closes with practical lessons and open problems, including deployment on edge hardware and the effect of a changing climate.

## ShapleyPipe: A Game-Theoretic Approach to Data Preparation Pipeline Construction

DevFeed: [ShapleyPipe: A Game-Theoretic Approach to Data Preparation Pipeline Construction](<https://devfeed.tech/articles/shapleypipe-a-game-theoretic-approach-to-data-preparation-pipeline-construction-145728.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2510.27168>)

Author: Jing Chang, Shuyuan Zheng, Jinbin Huang, Chang Liu, Jianbin Qin

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

Content type: article

Language: en

Sources: [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Feature Engineering](<https://devfeed.tech/topics/feature-engineering.md>)

Tags: [automated](<https://devfeed.tech/tags/automated.md>), [classification](<https://devfeed.tech/tags/classification.md>), [coherence](<https://devfeed.tech/tags/coherence.md>), [construction](<https://devfeed.tech/tags/construction.md>), [contribution](<https://devfeed.tech/tags/contribution.md>), [contributions](<https://devfeed.tech/tags/contributions.md>), [cs-db](<https://devfeed.tech/tags/cs-db.md>), [data-preparation](<https://devfeed.tech/tags/data-preparation.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [framework](<https://devfeed.tech/tags/framework.md>), [frontier](<https://devfeed.tech/tags/frontier.md>), [game](<https://devfeed.tech/tags/game.md>), [machine](<https://devfeed.tech/tags/machine.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [practical](<https://devfeed.tech/tags/practical.md>), [rank](<https://devfeed.tech/tags/rank.md>), [records](<https://devfeed.tech/tags/records.md>), [reduce](<https://devfeed.tech/tags/reduce.md>), [scaling](<https://devfeed.tech/tags/scaling.md>), [transformation](<https://devfeed.tech/tags/transformation.md>)

### AI overview

ShapleyPipe builds data preparation pipelines by estimating each operator's position-dependent contribution to downstream model utility. Its Pipeline Shapley value distinguishes repeated uses of the same operator type, and the framework selects and orders operators using those contributions. Hierarchical ShapleyPipe (H-SP) makes the approach more practical through representative selection based on a Category Coherence Assumption and adaptive early termination. Across 27 classification datasets, H-SP reports the highest average accuracy (0.847) and best average rank among evaluated practical methods, on the empirical accuracy-runtime Pareto frontier. On three datasets with up to 581K records, searching 10K subsamples with 16 workers averaged 891 seconds and accuracy of 0.705.

### Source excerpt

arXiv:2510.27168v2 Announce Type: replace Abstract: Data preparation is a critical yet highly time-consuming component of machine learning (ML) workflows and has a substantial impact on downstream model performance. Automated data preparation pipeline construction requires selecting appropriate operators, such as missing-value imputation, feature scaling, and feature transformation, and placing them at suitable pipeline positions. However, existing methods lack a principled and interpretable mechanism for understanding the marginal contribution of each operator at a specific position. To address this limitation, we study data preparation pipeline construction from a cooperative game-theoretic perspective. We formulate the task as a pipeline construction game, in which operators act as players that cooperatively improve downstream model utility. Based on this formulation, we propose the Pipeline Shapley value (PipeShapley), a contribution-attribution method that explicitly captures position-dependent contributions and distinguishes repeated occurrences of the same operator type. Building on PipeShapley, we introduce ShapleyPipe, an interpretable pipeline construction framework that selects and orders operators according to their position-aware contributions. To make PipeShapley practical, we develop Hierarchical ShapleyPipe (H-SP) based on a Category Coherence Assumption to exploit coherent contribution patterns, combining MAB-based representative selection with adaptive early termination. Experiments on 27 classification datasets show that H-SP achieves the highest average accuracy of 0.847 and the best average rank among the evaluated practical methods, while lying on the empirical accuracy--runtime Pareto frontier. On three datasets with up to 581K records, 10K-subsample search and 16-worker parallel execution reduce the average wall-clock time to 891 seconds while retaining the highest average accuracy of 0.705.

## SlotFlow-GB: A Neural Framework for Trans-Dimensional Inference of Overlapping Galactic Binaries

DevFeed: [SlotFlow-GB: A Neural Framework for Trans-Dimensional Inference of Overlapping Galactic Binaries](<https://devfeed.tech/articles/slotflow-gb-a-neural-framework-for-trans-dimensional-inference-of-overlapping-galactic-binaries-145133.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.26057>)

Author: Niklas Houba, Giovanni Giarda, Davide Piras

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

Content type: article

Language: en

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

Topics: [Deep neural networks](<https://devfeed.tech/topics/deep-neural-networks.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>)

Tags: [astro-ph-im](<https://devfeed.tech/tags/astro-ph-im.md>), [binaries](<https://devfeed.tech/tags/binaries.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gr-qc](<https://devfeed.tech/tags/gr-qc.md>), [inference](<https://devfeed.tech/tags/inference.md>), [neural](<https://devfeed.tech/tags/neural.md>), [overlap](<https://devfeed.tech/tags/overlap.md>), [physics-data-an](<https://devfeed.tech/tags/physics-data-an.md>), [physics-ins-det](<https://devfeed.tech/tags/physics-ins-det.md>)

### AI overview

SlotFlow-GB estimates both how many Galactic binaries are present and each binary's parameters in one evaluation, without requiring a source count in advance. In year-long LISA simulations with up to ten binaries per 2 μHz band, the inferred count was within one of the true count in 94% of cases, and the model identified binaries despite strong frequency overlap. A complete evaluation with parameter sampling took 41 ms per band on an NVIDIA GH200 GPU or 530 ms on eight laptop CPU threads. Its output distributions can also initialize later likelihood-based inference. The reported demonstration focuses on high-SNR cases where overlap, rather than individual-source detectability, is the main challenge.

### Source excerpt

arXiv:2611.26057v1 Announce Type: cross Abstract: The Laser Interferometer Space Antenna (LISA) will observe the millihertz gravitational-wave sky, including millions of long-lived Galactic binaries (GBs), tens of thousands of which are expected to be individually resolvable throughout the mission's lifetime. Extracting these sources requires jointly determining their number and physical parameters from strongly overlapping signals, posing a computationally demanding trans-dimensional inference problem. We introduce SlotFlow-GB, a neural framework that constructs a learned approximation to the trans-dimensional posterior, producing a probability distribution over source multiplicity and parameter densities for individual binaries in a single evaluation, without requiring their number to be specified in advance. We demonstrate this framework using year-long LISA simulations containing up to ten binaries within a 2 $\mu$Hz frequency interval, focusing on a high-SNR regime in which source overlap, rather than individual-source detectability, is the primary challenge. The inferred source count differs from the true count by at most one in 94% of cases, while the model identifies individual binaries even when their signals overlap strongly in frequency. A complete evaluation, including parameter sampling, takes just 41 ms per band on one NVIDIA GH200 GPU, or 530 ms on eight laptop CPU threads. We further demonstrate that these distributions can be used to initialize subsequent likelihood-based inference, connecting fast amortized predictions with joint parameter estimation.

## What Is BIO Tagging and How to Use It for Document AI Training Data

DevFeed: [What Is BIO Tagging and How to Use It for Document AI Training Data](<https://devfeed.tech/articles/what-is-bio-tagging-and-how-to-use-it-for-document-ai-training-data-146784.md>)

Original publisher: [Read original article](<https://www.sitepoint.com/what-is-bio-tagging-and-how-to-use-it-for-document-ai-training-data/>)

Author: Bilal Ahmad

Published: 2026-11-19T12:24:33Z

Content type: tutorial

Language: en

Sources: [SitePoint](<https://devfeed.tech/sources/sitepoint.md>)

Topics: [document ai](<https://devfeed.tech/topics/document-ai.md>), [AI-Ready Data](<https://devfeed.tech/topics/ai-ready-data.md>), [Data Quality](<https://devfeed.tech/topics/data-quality.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Transformers](<https://devfeed.tech/topics/transformers.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-training](<https://devfeed.tech/tags/ai-training.md>), [ai-training-data](<https://devfeed.tech/tags/ai-training-data.md>), [bio](<https://devfeed.tech/tags/bio.md>), [data](<https://devfeed.tech/tags/data.md>), [document-ai](<https://devfeed.tech/tags/document-ai.md>)

### AI overview

BIO tagging labels each token as the beginning (B), continuation (I), or outside (O) of an entity. For document AI, the article pairs token labels with normalized bounding boxes so models such as LayoutLMv3 can learn both text and page position. It recommends defining a consistent, fine-grained label taxonomy before annotation, extracting words and coordinates from PDFs or OCR, assigning BIO labels, and formatting parallel words, boxes, and labels for training. Regex pre-labeling can cover some predictable fields, while active learning and auto-labeling platforms shift effort toward review; names, addresses, and line descriptions still need human judgment. Validate entity boundaries, BIO sequences, label balance, and bounding boxes before training. The article suggests starting with 50-100 labeled documents, fine-tuning a pretrained checkpoint, then checking errors before scaling; its stated accuracy ranges depend on document diversity and assume pretrained fine-tuning.

### Source excerpt

null Continue reading What Is BIO Tagging and How to Use It for Document AI Training Data on SitePoint.

## ReliFlow: Reliability-Calibrated Residual Transport for Combinatorial Single-Cell Perturbation Prediction

DevFeed: [ReliFlow: Reliability-Calibrated Residual Transport for Combinatorial Single-Cell Perturbation Prediction](<https://devfeed.tech/articles/reliflow-reliability-calibrated-residual-transport-for-combinatorial-single-cell-perturbation-prediction-142735.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23999>)

Author: Jinrui David, Yun Peng

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

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: [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Tensorflow](<https://devfeed.tech/topics/tensorflow.md>)

Tags: [combined](<https://devfeed.tech/tags/combined.md>), [component](<https://devfeed.tech/tags/component.md>), [conditional](<https://devfeed.tech/tags/conditional.md>), [correlation](<https://devfeed.tech/tags/correlation.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cross](<https://devfeed.tech/tags/cross.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [distribution](<https://devfeed.tech/tags/distribution.md>), [effect](<https://devfeed.tech/tags/effect.md>), [error](<https://devfeed.tech/tags/error.md>), [field](<https://devfeed.tech/tags/field.md>), [flow](<https://devfeed.tech/tags/flow.md>), [flow-matching](<https://devfeed.tech/tags/flow-matching.md>), [matching](<https://devfeed.tech/tags/matching.md>), [model](<https://devfeed.tech/tags/model.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [reference](<https://devfeed.tech/tags/reference.md>), [reliability](<https://devfeed.tech/tags/reliability.md>), [residual](<https://devfeed.tech/tags/residual.md>), [single-cell](<https://devfeed.tech/tags/single-cell.md>), [stat-ml](<https://devfeed.tech/tags/stat-ml.md>), [transport](<https://devfeed.tech/tags/transport.md>)

### AI overview

ReliFlow separates errors in estimated single-perturbation effects from genuine interactions by estimating effect reliability through training-only cross-fitting and using it to calibrate an additive source distribution. Its conditional flow-matching velocity combines a known correction to that reference with a learned residual transport field. In the reported evaluations, MAE fell from 0.142 to 0.131 for unseen combinations of observed Norman genetic perturbations, from 0.198 to 0.178 when constituent singles were withheld, and from 0.191 to 0.176 on held-out ComboSciPlex drug combinations. Non-additive residual correlation rose from 0.574 to 0.638; ablations indicate reliability weighting contributes to performance. The authors report a moderate increase in training cost.

### Source excerpt

arXiv:2611.23999v1 Announce Type: cross Abstract: Predicting the transcriptional response to combined perturbations requires distinguishing interactions between interventions from errors in their individual effect estimates. The latter become more pronounced when single-perturbation evidence is sparse or withheld. We introduce a conditional flow-matching model that estimates the reliability of single-gene effects from training-only cross-fitting and uses those estimates to calibrate the additive source distribution. The velocity is decomposed into a known correction of the additive reference and a learned residual transport field. On Norman genetic perturbations, reduces mean absolute error (MAE) from 0.142 to 0.131 for unseen combinations of observed singles and from 0.198 to 0.178 when constituent singles are withheld. On held-out ComboSciPlex drug combinations, MAE decreases from 0.191 to 0.176. Non-additive residual correlation improves from 0.574 to 0.638, and component ablations identify the contribution of reliability weighting. The results indicate that accounting for uncertain constituent effects improves generalization to unseen genetic and chemical combinations, with a moderate increase in training cost.

## HardwareDx: A Foundational Graph Model System for Datacenter Machine Repairs

DevFeed: [HardwareDx: A Foundational Graph Model System for Datacenter Machine Repairs](<https://devfeed.tech/articles/hardwaredx-a-foundational-graph-model-system-for-datacenter-machine-repairs-142652.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23398>)

Author: Yuanlin Wen, Elan Markowitz, Zubo Gu, Sami Abu-El-Haija, Bryan Perozzi, Mikhail Galkin, Ali Parviz, Aayush Gupta, Venkata Chivukula, Griffin Hadfield, Aravind Akella, Daniel J. Frame, Vahab Mirrokni,

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

Content type: article

Language: en

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

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [systems](<https://devfeed.tech/topics/systems.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [cost-efficiency](<https://devfeed.tech/tags/cost-efficiency.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ar](<https://devfeed.tech/tags/cs-ar.md>), [datacenter](<https://devfeed.tech/tags/datacenter.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [diagnose](<https://devfeed.tech/tags/diagnose.md>), [diagnosis](<https://devfeed.tech/tags/diagnosis.md>), [downtime](<https://devfeed.tech/tags/downtime.md>), [foundation-model](<https://devfeed.tech/tags/foundation-model.md>), [google](<https://devfeed.tech/tags/google.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [graph](<https://devfeed.tech/tags/graph.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [maintenance](<https://devfeed.tech/tags/maintenance.md>), [model](<https://devfeed.tech/tags/model.md>), [onboarding](<https://devfeed.tech/tags/onboarding.md>), [quality-assurance-qa](<https://devfeed.tech/tags/quality-assurance-qa.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [reliability](<https://devfeed.tech/tags/reliability.md>), [tpus](<https://devfeed.tech/tags/tpus.md>)

### AI overview

HardwareDx applies a graph foundation model to diagnose hardware across multiple datacenter platforms with one model, aiming to reduce repair downtime and speed onboarding of new products. The abstract reports production deployment at Google, with downtime reductions of 18.3% for GPUs and 16.1% for TPUs, and QA time reductions of 28.3% and 25.1%, respectively. These results are presented as real-world evaluations; the supplied text does not describe the model's technical details or evaluation methods.

### Source excerpt

arXiv:2611.23398v1 Announce Type: new Abstract: The rapid scale-up of AI infrastructure has made datacenter reliability a cornerstone of the cloud business economy. Since hardware failures in hyper-scale AI clusters can stall multi-million dollar training workloads, the downtime of individual machines can cause outsized economic losses. Minimizing machine downtime, therefore, is more important than ever. To address this, we present Hardware Diagnoser (HardwareDx), an AI-driven hardware diagnosis system that applies a graph foundation model approach to automated datacenter maintenance. It allows a single model to diagnose many platforms and accelerate onboarding of new hardware products. Deployed at production scale at Google, HardwareDx yields substantial efficiency gains across diverse compute architectures. Real-world evaluations demonstrate that it reduces production downtime by 18.3% on GPU platforms and 16.1% on TPUs. Furthermore, it significantly accelerates hardware deployment, cutting quality assurance (QA) time by 28.3% for GPUs and 25.1% for TPUs. These results establish HardwareDx as a field-tested solution for maximizing the reliability and cost efficiency of next generation AI fleets.

## Piccolo: Co-Evolving Decision Models and Data through Recursive Self-Improvement

DevFeed: [Piccolo: Co-Evolving Decision Models and Data through Recursive Self-Improvement](<https://devfeed.tech/articles/piccolo-co-evolving-decision-models-and-data-through-recursive-self-improvement-142820.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24651>)

Author: Zhenqing Ling, Qirui Jiao, Daoyuan Chen, Cheng Chen, Xinji Mai, Haowen Gao, Zhuo Zhang, Xuchen Pan, Mengsu Ding, Jiaji Deng, Yuexiang Xie, Xiaoming Qin, Weirui Kuang, Zhaoyang Liu, Yaliang Li, Bolin D

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

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>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-harnesses](<https://devfeed.tech/tags/agent-harnesses.md>), [agentic-data-engineering](<https://devfeed.tech/tags/agentic-data-engineering.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [continuous](<https://devfeed.tech/tags/continuous.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [curation](<https://devfeed.tech/tags/curation.md>), [data](<https://devfeed.tech/tags/data.md>), [data-driven](<https://devfeed.tech/tags/data-driven.md>), [data-workflows](<https://devfeed.tech/tags/data-workflows.md>), [decision-models](<https://devfeed.tech/tags/decision-models.md>), [family](<https://devfeed.tech/tags/family.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [first](<https://devfeed.tech/tags/first.md>), [games](<https://devfeed.tech/tags/games.md>), [leaderboard](<https://devfeed.tech/tags/leaderboard.md>), [memory](<https://devfeed.tech/tags/memory.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [recursive-self-improvement](<https://devfeed.tech/tags/recursive-self-improvement.md>), [research](<https://devfeed.tech/tags/research.md>), [rsi](<https://devfeed.tech/tags/rsi.md>)

### AI overview

Piccolo is a 0.8B-27B multimodal decision-model family developed through recursive co-improvement of models and training data. Research agents analyze behavior, build data, run experiments, and feed results into later proposals, linked by semantic slices and shared research memory; the training recipes use hundreds of thousands of examples. Piccolo-27B scores 75.59 on Decision Index Vision Full, with 9B and 4B variants scoring 68.35 and 67.46; all three rank first on their size categories' Private benchmark. The authors report that earlier backbones can outperform newer ones on visual tasks under the same recipe, targeted curation can let 4B-9B models rival 27B baselines on specialized workflows, and lower training loss may not improve--and can harm--discrete decision accuracy. The evidence is the supplied abstract, not full paper text.

### Source excerpt

arXiv:2611.24651v1 Announce Type: cross Abstract: We present Piccolo, a family of multimodal decision models (0.8B to 27B) that coevolves model capabilities and training data via Agentic Data Engineering and recursive self-improvement (RSI). The models predict probabilities over candidate answers for general-purpose text and visual judgments, agent harnesses, data workflows, and interactive games. Piccolo-27B tops the Decision Index Vision Full leaderboard with a score of 75.59, while Piccolo-9B and Piccolo-4B achieve 68.35 and 67.46, respectively. All three rank first on the Private benchmark in their respective model-size categories. Research agents analyze model behavior, construct training data, conduct experiments, and use the results to refine subsequent proposals. Semantic slices and a shared research memory connect these actions across the model family's development. With training recipes comprising hundreds of thousands of examples, data selection and feedback-driven refinement are central to the self-improvement process. Our large-scale RSI exploration yields three counter-intuitive insights for decision-model development: (1) Backbone-recipe co-adaptation outweighs raw model recency: earlier backbones can surpass newer generations on visual tasks under identical recipes; (2) Targeted data curation enables 4B-9B models to rival 27B baselines on specialized workflows; and (3) Continuous loss reduction can decouple from and even degrade discrete decision accuracy. Together with the released weights and tools from the Data-Juicer and AgentScope ecosystems, Piccolo provides an actionable blueprint for building frontier decision models and scaling data-driven RSI.

## Building Predictive AI Models When Synthetic Data Is All You Have

DevFeed: [Building Predictive AI Models When Synthetic Data Is All You Have](<https://devfeed.tech/articles/building-predictive-ai-models-when-synthetic-data-is-all-you-have-144066.md>)

Original publisher: [Read original article](<https://engineering.salesforce.com/building-predictive-ai-models-when-synthetic-data-is-all-you-have/>)

Author: Taniya Ghosh

Published: 2026-11-19T00:59:15Z

Content type: article

Language: en

Sources: [Salesforce](<https://devfeed.tech/sources/salesforce-engineering-blog.md>)

Topics: [Predictive AI](<https://devfeed.tech/topics/predictive-ai.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [AI Models](<https://devfeed.tech/topics/ai-models.md>), [Feature Engineering](<https://devfeed.tech/topics/feature-engineering.md>), [Open Source Models & Datasets](<https://devfeed.tech/topics/open-source-models-datasets.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [customer-data](<https://devfeed.tech/tags/customer-data.md>), [data](<https://devfeed.tech/tags/data.md>), [data-360](<https://devfeed.tech/tags/data-360.md>), [predictive-ai](<https://devfeed.tech/tags/predictive-ai.md>), [synthetic-data](<https://devfeed.tech/tags/synthetic-data.md>), [training-data](<https://devfeed.tech/tags/training-data.md>)

### AI overview

Salesforce's team built reusable IT service management prediction templates before real customer incidents were available for training. Product-manager scenarios, varied within guardrails and expanded with open-source LLMs, supplied synthetic data for testing. To avoid being misled by aggregate accuracy on rare SLA breaches, the team checked whether scores changed as expected when one risk factor changed and whether the top three predictors made sense to service managers; these scenario tests did not establish production accuracy. Data 360 transforms customer incident histories into training data, Einstein on Data 360 trains an XGBoost model, and a scoring transform writes predictions for display on incident records. Templates carry the data contract and model structure, while each customer's history supplies its own learned parameters. The excerpt says training requires about 400 completed incidents and scoring runs in scheduled batches about hourly; thin or poor-quality histories may reduce reliability, and immediate scoring remained a need. At interview time, the team had received no reports of the anticipated challenges becoming production problems and was awaiting more user反馈.

### Source excerpt

In our Engineering Energizers Q&A series, we highlight the engineering minds driving innovation across Salesforce. Today, we spotlight Taniya Ghosh, Senior Manager of Software Engineering, whose team faced a brutal engineering contradiction: build predictive AI models that warn of escalating incidents before they had real customer incidents to learn from. Synthetic training data gave them [...] The post Building Predictive AI Models When Synthetic Data Is All You Have appeared first on Salesforce Engineering Blog.

## How Salesforce is advancing global-scale model serving with cross-Region inference routing in Amazon SageMaker AI

DevFeed: [How Salesforce is advancing global-scale model serving with cross-Region inference routing in Amazon SageMaker AI](<https://devfeed.tech/articles/how-salesforce-is-advancing-global-scale-model-serving-with-cross-region-inference-routing-in-amazon-sagemaker-ai-141212.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/how-salesforce-is-advancing-global-scale-model-serving-with-cross-region-inference-routing-in-amazon-sagemaker-ai/>)

Author: Apoorva Chandra

Published: 2026-11-18T20:13:18Z

Content type: tutorial

Language: en

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

Topics: [Amazon SageMaker](<https://devfeed.tech/topics/amazon-sagemaker.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [AI Integration Strategies](<https://devfeed.tech/topics/ai-integration-strategies.md>), [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [amazon](<https://devfeed.tech/tags/amazon.md>), [amazon-sagemaker](<https://devfeed.tech/tags/amazon-sagemaker.md>), [amazon-sagemaker-ai](<https://devfeed.tech/tags/amazon-sagemaker-ai.md>), [call](<https://devfeed.tech/tags/call.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [cross](<https://devfeed.tech/tags/cross.md>), [customer-solutions](<https://devfeed.tech/tags/customer-solutions.md>), [data](<https://devfeed.tech/tags/data.md>), [data-residency](<https://devfeed.tech/tags/data-residency.md>), [global](<https://devfeed.tech/tags/global.md>), [inference](<https://devfeed.tech/tags/inference.md>), [intermediate-200](<https://devfeed.tech/tags/intermediate-200.md>), [model](<https://devfeed.tech/tags/model.md>), [model-serving](<https://devfeed.tech/tags/model-serving.md>), [region](<https://devfeed.tech/tags/region.md>), [regions](<https://devfeed.tech/tags/regions.md>), [routing](<https://devfeed.tech/tags/routing.md>), [salesforce](<https://devfeed.tech/tags/salesforce.md>), [scale](<https://devfeed.tech/tags/scale.md>), [serving](<https://devfeed.tech/tags/serving.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

Salesforce replaced its Internal Gateway's basic health checks, retries, and Region switching with an Amazon SageMaker AI inference router that sends calls across its existing endpoints and accounts. The router prefers local capacity, shifts traffic when request errors indicate an unhealthy or unavailable destination, and gradually returns traffic after recovery; high latency alone does not trigger failover. Residency-tagged requests use a target containing only the local endpoint. Endpoint owners authorize cross-account access with resource policies, and response fields identify the serving resource and routing reason for logs and CloudWatch metrics. In a saturation test, traffic stayed local during a latency rise, then shifted after container timeouts; about 78% reached the remote Region, with p50 latency returning to 0.2 seconds and under 100 ms of cross-Region overhead at p50. The router does not deploy models or add capacity, so registered Regions need matching model artifacts and sufficient capacity; a receiving endpoint can cap remote traffic. Salesforce plans further tests of sustained saturation, latency, and cascading failures.

### Source excerpt

Learn how Salesforce uses the cross-Region inference router in Amazon SageMaker AI to turn fragmented Regional capacity into reliable model serving. Salesforce adds capacity securely across Regions and accounts while keeping its data residency commitments, and sees per request where each call was served and why.

## From store outlets to methane leaks, every scrap of power is now fair game for AI compute

DevFeed: [From store outlets to methane leaks, every scrap of power is now fair game for AI compute](<https://devfeed.tech/articles/from-store-outlets-to-methane-leaks-every-scrap-of-power-is-now-fair-game-for-ai-compute-140453.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/18/distributed-inference-turns-stranded-power-ai-compute-dellatsc26/>)

Author: Jonathan Anthony

Published: 2026-11-18T11:32:01Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [clean energy data centers](<https://devfeed.tech/topics/clean-energy-data-centers.md>)

Tags: [accelerated-computing](<https://devfeed.tech/tags/accelerated-computing.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-compute](<https://devfeed.tech/tags/ai-compute.md>), [ai-data-centers](<https://devfeed.tech/tags/ai-data-centers.md>), [ai-factories](<https://devfeed.tech/tags/ai-factories.md>), [ai-inference](<https://devfeed.tech/tags/ai-inference.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [ai-supercomputers](<https://devfeed.tech/tags/ai-supercomputers.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [compute](<https://devfeed.tech/tags/compute.md>), [cube-event-coverage](<https://devfeed.tech/tags/cube-event-coverage.md>), [data-center](<https://devfeed.tech/tags/data-center.md>), [dell](<https://devfeed.tech/tags/dell.md>), [dell-technologies](<https://devfeed.tech/tags/dell-technologies.md>), [dell-technologies-inc](<https://devfeed.tech/tags/dell-technologies-inc.md>), [dellatsc26](<https://devfeed.tech/tags/dellatsc26.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [digital-twins](<https://devfeed.tech/tags/digital-twins.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [distributed-inference](<https://devfeed.tech/tags/distributed-inference.md>), [energy-efficiency](<https://devfeed.tech/tags/energy-efficiency.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [high-performance-computing](<https://devfeed.tech/tags/high-performance-computing.md>), [high-speed-networking](<https://devfeed.tech/tags/high-speed-networking.md>), [hpc](<https://devfeed.tech/tags/hpc.md>), [hpc-infrastructure](<https://devfeed.tech/tags/hpc-infrastructure.md>), [hpc-unites](<https://devfeed.tech/tags/hpc-unites.md>), [ihab-tarazi](<https://devfeed.tech/tags/ihab-tarazi.md>), [inference](<https://devfeed.tech/tags/inference.md>), [infra](<https://devfeed.tech/tags/infra.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [john-furrier](<https://devfeed.tech/tags/john-furrier.md>), [memory-pooling](<https://devfeed.tech/tags/memory-pooling.md>), [models](<https://devfeed.tech/tags/models.md>), [news](<https://devfeed.tech/tags/news.md>), [nyse-dell](<https://devfeed.tech/tags/nyse-dell.md>), [power](<https://devfeed.tech/tags/power.md>), [rack](<https://devfeed.tech/tags/rack.md>), [rack-scale](<https://devfeed.tech/tags/rack-scale.md>), [rebecca-knight](<https://devfeed.tech/tags/rebecca-knight.md>), [sc26](<https://devfeed.tech/tags/sc26.md>), [sc26-2026](<https://devfeed.tech/tags/sc26-2026.md>), [sc26eventpage](<https://devfeed.tech/tags/sc26eventpage.md>), [scale](<https://devfeed.tech/tags/scale.md>), [scientific-simulation](<https://devfeed.tech/tags/scientific-simulation.md>), [scinet](<https://devfeed.tech/tags/scinet.md>), [store](<https://devfeed.tech/tags/store.md>), [supercomputing](<https://devfeed.tech/tags/supercomputing.md>), [systems](<https://devfeed.tech/tags/systems.md>), [thecube](<https://devfeed.tech/tags/thecube.md>)

### AI overview

Dell says it is extending liquid cooling across AI compute, networking and storage as AI clusters grow to tens or hundreds of thousands of GPUs on a single fabric. Its 100% liquid-cooled networking system arrives as a factory-pretested rack with switches, optics and cabling; Dell says installation and startup can take hours instead of assembling components on-site and spending weeks debugging. Shared cooling and power specifications could let operators build one facility for GPUs and networking rather than separate facilities. For distributed inference, Dell is developing a control plane across different GPUs and CPUs to connect up to 1,000 small sites with spare power. Tarazi says potential sources include store power and methane. These are Dell's claims in a sponsored interview; the excerpt does not establish deployment results.

### Source excerpt

AI factories are outgrowing the traditional data center as models get bigger and distributed inference moves from concept to deployment. The infrastructure behind those models now has to span racks, sites and even scattered pockets of power. Dell Technologies Inc. is betting that the answer lies in building complete systems rather than components, from rack-scale [...] The post From store outlets to methane leaks, every scrap of power is now fair game for AI compute appeared first on SiliconANGLE.

## Deep Reinforcement Learning for Heat Pump and Underfloor Heating Control: An End-to-End Case Study

DevFeed: [Deep Reinforcement Learning for Heat Pump and Underfloor Heating Control: An End-to-End Case Study](<https://devfeed.tech/articles/deep-reinforcement-learning-for-heat-pump-and-underfloor-heating-control-an-end-to-end-case-study-140112.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23063>)

Author: Antonio Manjavacas, David Su\\'arez-Caro, Alejandro Campoy-Nieves, Guillermo Rey, Gonzalo Satr\\'ustegui, Miguel Molina-Solana

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

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>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Simulation](<https://devfeed.tech/topics/simulation.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [algorithms](<https://devfeed.tech/tags/algorithms.md>), [case-study](<https://devfeed.tech/tags/case-study.md>), [control](<https://devfeed.tech/tags/control.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-sy](<https://devfeed.tech/tags/cs-sy.md>), [deep-reinforcement-learning](<https://devfeed.tech/tags/deep-reinforcement-learning.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [digital-twin](<https://devfeed.tech/tags/digital-twin.md>), [eess-sy](<https://devfeed.tech/tags/eess-sy.md>), [energy-consumption](<https://devfeed.tech/tags/energy-consumption.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [heat-pump](<https://devfeed.tech/tags/heat-pump.md>), [robustness](<https://devfeed.tech/tags/robustness.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [strategies](<https://devfeed.tech/tags/strategies.md>)

### AI overview

The case study reports that a deep reinforcement learning controller for an integrated heat pump and underfloor heating system maintained indoor temperatures within prescribed comfort limits in an occupied residential building. Researchers first compared algorithms and control strategies in simulation, then trained agents using a digital twin of the building. They deployed the best-performing agent without additional online fine-tuning; it adjusted heat pump supply temperature and underfloor heating valve positions. The abstract presents this as evidence that deployment is feasible, while noting practical challenges in transferring policies from digital environments to real heating systems.

### Source excerpt

arXiv:2611.23063v1 Announce Type: new Abstract: Efficient control of underfloor heating (UFH) systems coupled with heat pumps (HPs) is challenging due to the high thermal inertia of buildings and the slow dynamic response of UFH systems. Deep Reinforcement Learning (DRL) offers a data-driven control approach capable of simultaneously optimizing energy consumption and thermal comfort. However, the application of DRL-based controllers in occupied buildings under real operating conditions remains limited. This work presents and end-to-end case study (development, evaluation, and real-world deployment) of a DRL-based controller for an integrated UFH-HP system. First, a simulation-based pilot study was conducted to assess different DRL algorithms, control strategies, and configurations. Subsequently, a digital twin of a real residential building was developed and used to train DRL agents under simulated operating conditions. The best-performing agent was then deployed in the physical building without additional online fine-tuning. The results show that the controller successfully maintained indoor temperatures within the prescribed comfort limits while dynamically adjusting the HP supply temperature and UFH valve positions. These findings demonstrate the feasibility of deploying DRL controllers in occupied buildings and highlight their robustness and the practical challenges associated with transferring control policies from digital environments to real heating systems.

## COTTI: Contrastive Training and Time-Aware Inference for Robust Industrial Recommendation under Promotional Drift

DevFeed: [COTTI: Contrastive Training and Time-Aware Inference for Robust Industrial Recommendation under Promotional Drift](<https://devfeed.tech/articles/cotti-contrastive-training-and-time-aware-inference-for-robust-industrial-recommendation-under-promotional-drift-140055.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22597>)

Author: Maolei Huang, Jing Tian, Dexuan Duan, Fangfang Lin, Huawei Cao, Shuhan Song

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: [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>)

Tags: [2-2](<https://devfeed.tech/tags/2-2.md>), [3-14](<https://devfeed.tech/tags/3-14.md>), [aba](<https://devfeed.tech/tags/aba.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [adaptive](<https://devfeed.tech/tags/adaptive.md>), [adjustment](<https://devfeed.tech/tags/adjustment.md>), [aggregate](<https://devfeed.tech/tags/aggregate.md>), [alignment](<https://devfeed.tech/tags/alignment.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [analyze](<https://devfeed.tech/tags/analyze.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [drift](<https://devfeed.tech/tags/drift.md>), [industrial](<https://devfeed.tech/tags/industrial.md>), [inference](<https://devfeed.tech/tags/inference.md>), [recommendation](<https://devfeed.tech/tags/recommendation.md>), [robust](<https://devfeed.tech/tags/robust.md>), [time](<https://devfeed.tech/tags/time.md>), [training](<https://devfeed.tech/tags/training.md>), [under](<https://devfeed.tech/tags/under.md>)

### AI overview

COTTI separates historical recommendation training from event-time score adjustment to address promotional drift. It uses Cross-Period Contrastive Alignment for paired user histories, Active Re-weighting based on historical-interest distance, and Adaptive Boundary Alignment to adjust logits while keeping predictor parameters fixed. On Fliggy data, it reached offline AUC 0.8127, versus 0.8101 for TEN and 0.8123 for TEN with ABA. Cross-category analysis found improved discrimination but lower paid-item recall for novel categories. Serving measured 38.9 ms p99 latency and 14,700 requests/s; ABA added 2.2 ms to p99. In a ten-day online experiment against the serving HDR model, COTTI produced relative gains of 3.14% in conversion rate and 2.68% in revenue per mille.

### Source excerpt

arXiv:2611.22597v1 Announce Type: new Abstract: Promotional events shift traffic composition, purchase behavior, and the scale of conversion probabilities on different time horizons. We present \textbf{COTTI} (\textbf{CO}ntrastive \textbf{T}raining and \textbf{T}ime-aware \textbf{I}nference), an industrial recommendation framework that separates historical representation learning from event-time score adjustment. Cross-Period Contrastive Alignment (CPC) regularizes paired histories of the same user with adaptive consistency weights. Active Re-weighting (RW) controls the influence of promotional examples through historical-interest distance. Adaptive Boundary Alignment (ABA) combines stochastic prediction variability with a temporal event kernel to adjust logits while keeping predictor parameters fixed. We analyze the weighted objective and bound ABA-induced probability and ranking changes. On Fliggy data, COTTI attains an offline AUC of 0.8127, compared with 0.8101 for TEN and 0.8123 for TEN with ABA. Cross-category analysis reveals improved discrimination alongside declines in paid-item recall for novel categories. End-to-end measurements give a p99 latency of 38.9\,ms and throughput of 14,700 requests/s; enabling ABA adds 2.2\,ms to p99. A ten-day online experiment against the serving HDR model yields relative gains of 3.14\% in conversion rate and 2.68\% in revenue per mille. Together, these results characterize the trade-off between aggregate prediction, novel-category discovery, and serving cost in promotional recommendation.

## CNN Architecture Comparison for Malware Classification with Limited Training Data

DevFeed: [CNN Architecture Comparison for Malware Classification with Limited Training Data](<https://devfeed.tech/articles/cnn-architecture-comparison-for-malware-classification-with-limited-training-data-141475.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21936>)

Author: Noor Abo Hasan

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

Content type: comparison

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>)

Tags: [architecture](<https://devfeed.tech/tags/architecture.md>), [classification](<https://devfeed.tech/tags/classification.md>), [cnn](<https://devfeed.tech/tags/cnn.md>), [comparison](<https://devfeed.tech/tags/comparison.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [feature](<https://devfeed.tech/tags/feature.md>), [fusion](<https://devfeed.tech/tags/fusion.md>), [image](<https://devfeed.tech/tags/image.md>), [malware](<https://devfeed.tech/tags/malware.md>), [robustness](<https://devfeed.tech/tags/robustness.md>), [size](<https://devfeed.tech/tags/size.md>), [training](<https://devfeed.tech/tags/training.md>), [training-data](<https://devfeed.tech/tags/training-data.md>), [trojan](<https://devfeed.tech/tags/trojan.md>), [worm](<https://devfeed.tech/tags/worm.md>)

### AI overview

The study compares pre-trained VGG16 alone with a dual-path model combining VGG16 and a custom CNN through intermediate feature fusion, for image-based classification of C2Lop.P and Dialplatform.B malware in the Malimg dataset. Both models achieved perfect classification across training proportions of 20%, 40%, 60%, and 80%. The comparison also found differences in training efficiency, noise robustness, and architectural complexity; the abstract reports that the simpler model could be more efficient and robust without sacrificing classification performance. The evidence is limited to this abstract and does not specify the secondary metric results in detail.

### Source excerpt

arXiv:2611.21936v1 Announce Type: new Abstract: Traditional malware classification methods struggle to keep pace with rapidly evolving threats. Image-based malware classification offers a promising alternative by converting malware binaries into grayscale images for analysis with deep learning. While effective, organizations often face practical constraints in acquiring sufficient training data, especially for newer threats. This thesis investigates dual-path CNN architectures with intermediate feature fusion as a potential enhancement for malware classification, with particular focus on how varying training data size affects their comparative performance against traditional single-path approaches. This research implements and compares two models: a baseline single-path using the pre-trained VGG16 architecture and a dual-path model that combines VGG16 with a custom CNN through intermediate feature fusion. Both architectures are evaluated on a binary classification task distinguishing between two malware families (C2Lop.P trojan and Dialplatform.B worm) from the Malimg dataset. Each model is trained across four different training data proportions (20%, 40%, 60%, and 80%), with three training runs per size to ensure statistical robustness, while being evaluated on complementary test sets of varying sizes. Though both architectures achieved perfect classification performance across all training percentages, the study revealed significant differences in secondary metrics including training efficiency, robustness to noise perturbation, and architectural complexity. The findings challenge common assumptions about complex architectures, demonstrating that simpler models can offer superior efficiency and robustness without sacrificing classification performance in image-based malware analysis.

## VRLLM: Turning LLMs into Powerful Tabular Classifiers

DevFeed: [VRLLM: Turning LLMs into Powerful Tabular Classifiers](<https://devfeed.tech/articles/vrllm-turning-llms-into-powerful-tabular-classifiers-141764.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22308>)

Author: Yingtong Dou, Shubham Jain, Junpeng Wang, Hadi Abdullah, Vineeth Rakesh Mohan, Yuzhong Chen, Liang Wang, Yiwei Cai

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

Content type: article

Language: en

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

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [Model Development](<https://devfeed.tech/topics/model-development.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [model architecture](<https://devfeed.tech/topics/model-architecture.md>), [Frontier AI](<https://devfeed.tech/topics/frontier-ai.md>), [llama](<https://devfeed.tech/topics/llama.md>)

Tags: [ablation](<https://devfeed.tech/tags/ablation.md>), [adapting](<https://devfeed.tech/tags/adapting.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [case-study](<https://devfeed.tech/tags/case-study.md>), [classification](<https://devfeed.tech/tags/classification.md>), [classifiers](<https://devfeed.tech/tags/classifiers.md>), [component](<https://devfeed.tech/tags/component.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [data-preparation](<https://devfeed.tech/tags/data-preparation.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [feature](<https://devfeed.tech/tags/feature.md>), [ffn](<https://devfeed.tech/tags/ffn.md>), [frontier](<https://devfeed.tech/tags/frontier.md>), [industrial](<https://devfeed.tech/tags/industrial.md>), [interpretability](<https://devfeed.tech/tags/interpretability.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [llms](<https://devfeed.tech/tags/llms.md>)

### AI overview

Visa Research reports that fine-tuning a pretrained Llama 3.2 1B on billions of labeled tabular rows produced VRLLM, which outperformed a range of tabular classifiers and the group's production deep learning model on the studied task. The paper describes its data preparation, training recipe, design ablations, and deployment validation, and benchmarks frontier LLMs on the same task. Interpretability experiments suggest feature-specific predictive associations; ablations indicate the earliest two feed-forward network layers contribute to linearly decodable task information. The abstract presents this as an industrial case study, so the claims are specific to the task and evidence summarized here.

### Source excerpt

arXiv:2611.22308v1 Announce Type: new Abstract: Large Language Models (LLMs) have shown strong performance on text, but whether pretrained LLMs are effective for large-scale tabular classification remains unclear, as many previous works demonstrate that LLMs' tabular data classification performance falls behind tree-based models or neural architectures designed for tabular data. At Visa Research, we managed to fine-tune a pretrained Llama 3.2 1B LLM on billions of rows of labeled tabular data and make it outperform a wide array of tabular classifiers and even our competitive production deep learning model. For ease of reference, we refer to the model as the Visa Research LLM (VRLLM). In this paper, we share how we achieved this exciting result, covering data preparation, training recipe, ablation studies of design choices and deployment validation. We also benchmark frontier LLMs on the same task and analyze their prediction behavior across model families and generations. Our interpretability experiments suggest that VRLLM learns feature-specific predictive associations, while component ablations indicate that the earliest two FFN layers contribute to the emergence of linearly decodable task information. Our paper provides an engineering recipe and a focused empirical case study on adapting compact LLMs to industrial tabular prediction tasks.

## PRUE: A Practical Recipe for Field Boundary Segmentation at Scale

DevFeed: [PRUE: A Practical Recipe for Field Boundary Segmentation at Scale](<https://devfeed.tech/articles/prue-a-practical-recipe-for-field-boundary-segmentation-at-scale-141546.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2603.27101>)

Author: Gedeon Muhawenayo, Caleb Robinson, Subash Khanal, Zhanpei Fang, Isaac Corley, Alexander Wollam, Tianyi Gao, Leonard Strnad, Ryan Avery, Lyndon Estes, Ana M. T\\'arano, Nathan Jacobs, Hannah Kerner

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

Content type: article

Language: en

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

Topics: [Image processing](<https://devfeed.tech/topics/image-processing.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [geospatial](<https://devfeed.tech/topics/geospatial.md>), [Dataset development](<https://devfeed.tech/topics/dataset-development.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [all](<https://devfeed.tech/tags/all.md>), [alternatives](<https://devfeed.tech/tags/alternatives.md>), [approach](<https://devfeed.tech/tags/approach.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [based](<https://devfeed.tech/tags/based.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [boundary](<https://devfeed.tech/tags/boundary.md>), [changes](<https://devfeed.tech/tags/changes.md>), [combines](<https://devfeed.tech/tags/combines.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [deep](<https://devfeed.tech/tags/deep.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [design](<https://devfeed.tech/tags/design.md>), [field](<https://devfeed.tech/tags/field.md>), [practical](<https://devfeed.tech/tags/practical.md>), [recipe](<https://devfeed.tech/tags/recipe.md>), [scale](<https://devfeed.tech/tags/scale.md>)

### AI overview

On the Fields of The World benchmark, a U-Net semantic segmentation model outperformed the evaluated instance-based and geospatial foundation models across performance and deployment metrics. The authors' approach combines a U-Net backbone with composite loss functions and targeted data augmentations to improve robustness to real-world variation. It reports 76% IoU and 47% object-F1, gains of 6 and 9 percentage points over the previous baseline, respectively. The abstract says the work releases all models and model-derived field-boundary datasets for five countries.

### Source excerpt

arXiv:2603.27101v3 Announce Type: replace-cross Abstract: Large-scale maps of field boundaries are essential for agricultural monitoring tasks. Existing deep learning approaches for satellite-based field mapping are sensitive to illumination, spatial scale, and changes in geographic location. We conduct the first systematic evaluation of segmentation and geospatial foundation models (GFMs) for global field boundary delineation using the Fields of The World (FTW) benchmark. We evaluate 18 models under unified experimental settings, showing that a U-Net semantic segmentation model outperforms instance-based and GFM alternatives on a suite of performance and deployment metrics. We propose a new segmentation approach that combines a U-Net backbone, composite loss functions, and targeted data augmentations to enhance performance and robustness under real-world conditions. Our model achieves a 76% IoU and 47% object-F1 on FTW, an increase of 6% and 9% over the previous baseline. Our approach provides a practical framework for reliable, scalable, and reproducible field boundary delineation across model design, training, and inference. We release all models and model-derived field boundary datasets for five countries.

## Test-Time Coverage: Test-Conditioned Data Curation for Deployment-Aware Learning

DevFeed: [Test-Time Coverage: Test-Conditioned Data Curation for Deployment-Aware Learning](<https://devfeed.tech/articles/test-time-coverage-test-conditioned-data-curation-for-deployment-aware-learning-139794.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2607.22697>)

Author: Nadine Chang, Maying Shen, Shizhe Diao, Jialiang Wang, Jingde Chen, Thomas Breuel, Pavlo Molchanov, Rafid Mahmood, Jose M. Alvarez

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: [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [AI agents in autonomous vehicles](<https://devfeed.tech/topics/ai-agents-in-autonomous-vehicles.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [autonomous-driving](<https://devfeed.tech/tags/autonomous-driving.md>), [coverage](<https://devfeed.tech/tags/coverage.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>), [curation](<https://devfeed.tech/tags/curation.md>), [data](<https://devfeed.tech/tags/data.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [distribution](<https://devfeed.tech/tags/distribution.md>), [inference](<https://devfeed.tech/tags/inference.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [learning](<https://devfeed.tech/tags/learning.md>), [test](<https://devfeed.tech/tags/test.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

TTCov uses unlabeled deployment samples to guide training-data selection before training, without adapting model weights at inference. It builds a task Atlas of deployment-relevant atomic propositions, expands it with concepts found in deployment samples, and uses their observed frequencies to form a Knowledge Atlas target. It then selects a budgeted training set to approximate that target. In autonomous driving, the authors report greater deployment-relevant coverage, closer distribution matching, and stronger end-to-end driving performance than data-curation baselines, with city-to-city expansion for novel domains. The supplied abstract does not provide quantitative results or methodological details.

### Source excerpt

arXiv:2607.22697v2 Announce Type: replace Abstract: Deployed AI systems are often trained from broad candidate data pools, necessitating data curation towards the deployment test distribution. However, standard data curation methods score training-side criteria rather than directly optimizing deployment match. We introduce TTCov (Test-Time Coverage), a data-level test-conditioned curation method that uses test-side information before training instead of updating model weights at inference. TTCov decomposes deployment-conditioned curation into coverage and distribution. To represent coverage, it builds a task Atlas, a collection of LLM-based atomic propositions (APs) describing deployment-relevant concepts, seeded from open task knowledge and expanded with unmatched APs extracted from unlabeled deployment samples. To represent distribution, it instantiates the matched deployment APs with their frequencies, yielding a Knowledge Atlas (K-Atlas) that operationalizes the deployment distribution as a curation target. TTCov then selects a budgeted training set whose deployment APs distribution approximates this target. We apply TTCov towards autonomous driving (AD), keeping adaptation off the inference path while selecting data with greater deployment-relevant coverage, closer K-Atlas matching, and stronger downstream end-to-end driving performance than data-curation baselines, including seamless adaptability to novel domains via city-to-city expansion. Code available at https://github.com/NVlabs/TTCov.

## Proximal Targets: Training Looped Refiners on Problems with Many Correct Answers

DevFeed: [Proximal Targets: Training Looped Refiners on Problems with Many Correct Answers](<https://devfeed.tech/articles/proximal-targets-training-looped-refiners-on-problems-with-many-correct-answers-140072.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22737>)

Author: Syed Affan Daimi

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: [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Deep neural networks](<https://devfeed.tech/topics/deep-neural-networks.md>)

Tags: [capacity](<https://devfeed.tech/tags/capacity.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deep](<https://devfeed.tech/tags/deep.md>), [draft](<https://devfeed.tech/tags/draft.md>), [errors](<https://devfeed.tech/tags/errors.md>), [fixed](<https://devfeed.tech/tags/fixed.md>), [found](<https://devfeed.tech/tags/found.md>), [gram](<https://devfeed.tech/tags/gram.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [grid](<https://devfeed.tech/tags/grid.md>), [hrm](<https://devfeed.tech/tags/hrm.md>), [input](<https://devfeed.tech/tags/input.md>), [list](<https://devfeed.tech/tags/list.md>), [model](<https://devfeed.tech/tags/model.md>), [more](<https://devfeed.tech/tags/more.md>), [network](<https://devfeed.tech/tags/network.md>), [node](<https://devfeed.tech/tags/node.md>), [nodes](<https://devfeed.tech/tags/nodes.md>), [parameter](<https://devfeed.tech/tags/parameter.md>), [problems](<https://devfeed.tech/tags/problems.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

The authors argue that training every refinement pass toward one stored answer creates an irreducible loss floor when puzzles have many valid answers, and can make the model blend them into an invalid answer. Their proximal target instead trains each pass toward a nearby correct answer, found from an answer list, by starting a search solver at the draft, or within a quality band controlled by a knob. They report that, with the same 1.7M-parameter network and five seeds, this approach solves 98.8% of Sudoku puzzles with 2-64 answers versus 2.5% for the standard target, and 93.8% of sparse Sudoku puzzles with over 10,000 answers without an answer list versus 0.0%. It also reaches 90.9% on 200-node 3-colourings after training on graphs up to 50 nodes, versus 36.9%. Against GRAM, it has higher single-run correctness and finds more answers across twenty runs, though GRAM performs slightly better on its own 10-node colouring benchmark. The excerpt says the knob serves as a calibrated strength dial in Scrabble.

### Source excerpt

arXiv:2611.22737v1 Announce Type: new Abstract: Looped refiners such as HRM and TRM apply one small network repeatedly to a draft answer, and deep supervision trains every pass toward one stored answer per puzzle. We prove that when a puzzle has many correct answers this target sets a loss floor that no model capacity removes, and that the loss-minimising pass blends the answers into an invalid grid. We propose the proximal target: train each pass toward the correct answer nearest to its own input, found by scanning a list of answers, by a search solver started at the draft, or within a band of quality set by a knob; correct answers become fixed points, and a pass only has to make fewer errors than its input has for the loop to converge, which its training loss bounds. With the same 1.7M-parameter network and 5 seeds, the proximal target solves 98.8% of Sudoku puzzles with 2-64 answers (standard target: 2.5%), 93.8% of sparse puzzles with more than 10,000 answers without any list of answers (0.0%), and 90.9% of 200-node 3-colourings after training on graphs of at most 50 nodes (36.9%), and different random starts end at different answers. Against GRAM, a stochastic-latent reasoner, a single run is correct far more often (98.8% vs. 57.7%) and twenty runs find more of the answers (53.6% vs. 37.4%), while GRAM is slightly ahead on its own 10-node colouring benchmark; in Scrabble the knob becomes a calibrated strength dial.

## Building new research tools for Maya epigraphy on Modal

DevFeed: [Building new research tools for Maya epigraphy on Modal](<https://devfeed.tech/articles/building-new-research-tools-for-maya-epigraphy-on-modal-142067.md>)

Original publisher: [Read original article](<https://modal.com/blog/scaling-intelligence-asya>)

Author: Peri Langlois

Published: 2026-11-18T00:00:00Z

Content type: article

Language: en

Sources: [Modal](<https://devfeed.tech/sources/modal.md>)

Topics: [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [3D](<https://devfeed.tech/topics/3d.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>)

Tags: [3d](<https://devfeed.tech/tags/3d.md>), [3d-model](<https://devfeed.tech/tags/3d-model.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-compute](<https://devfeed.tech/tags/cloud-compute.md>), [collaboration](<https://devfeed.tech/tags/collaboration.md>), [collective](<https://devfeed.tech/tags/collective.md>), [culture](<https://devfeed.tech/tags/culture.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [drawing](<https://devfeed.tech/tags/drawing.md>), [maya](<https://devfeed.tech/tags/maya.md>), [modal](<https://devfeed.tech/tags/modal.md>), [research](<https://devfeed.tech/tags/research.md>), [stanford](<https://devfeed.tech/tags/stanford.md>), [tools](<https://devfeed.tech/tags/tools.md>), [writing](<https://devfeed.tech/tags/writing.md>)

### AI overview

Miguel Novelo combines family and scholarly knowledge with multiple AI models to investigate Maya glyphs, checking model-generated connections against sources because general-purpose models alone lack the cultural and linguistic understanding he needs. Using Modal, he built a pipeline that turns 2D glyph images into 3D models from roughly 100 syllabic signs; this also lets epigraphers generate, rotate, and relight carving models from photographs, reducing work that once took about a week to a fraction of that time. His installation invites visitors to draw in sand, transforming their marks into glyphs in a growing projected codex. He plans to refine the process with collected drawings and ultimately digitize carvings in 3D to study whether carving depth and style reveal patterns that flat images miss. The excerpt also notes concerns about applying AI to Indigenous writing systems; Novelo frames his work around transmission and collaboration while acknowledging the need for respect and repair.

### Source excerpt

Miguel Novelo grew up surrounded by Maya glyphs carved into stone across Mexico's Yucatán peninsula, printed in books, and painted on ceramics. For centuries, Maya scribes used this writing system to record histories, trace the movements of the heavens, and keep calendars. For most of his life, he saw them only as drawings. "I never thought I could understand them," he says. Novelo is an artist and lecturer at Stanford, where he teaches a course called World Building: Video, Sound, and Space.

## Announcing AI data preparation capabilities in Amazon SageMaker: Build high-quality datasets for model development workflows

DevFeed: [Announcing AI data preparation capabilities in Amazon SageMaker: Build high-quality datasets for model development workflows](<https://devfeed.tech/articles/announcing-ai-data-preparation-capabilities-in-amazon-sagemaker-build-high-quality-datasets-for-model-development-workflows-138418.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/announcing-ai-data-preparation-capabilities-in-amazon-sagemaker-build-high-quality-datasets-for-model-development-workflows/>)

Author: Varun Rajan

Published: 2026-11-17T19:40:01Z

Content type: release

Language: en

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

Topics: [Amazon SageMaker](<https://devfeed.tech/topics/amazon-sagemaker.md>), [Dataset development](<https://devfeed.tech/topics/dataset-development.md>), [Data Quality](<https://devfeed.tech/topics/data-quality.md>), [AI-Ready Data](<https://devfeed.tech/topics/ai-ready-data.md>), [human feedback](<https://devfeed.tech/topics/human-feedback.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Amazon S3](<https://devfeed.tech/topics/amazon-s3.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-skill](<https://devfeed.tech/tags/agent-skill.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-data](<https://devfeed.tech/tags/ai-data.md>), [amazon](<https://devfeed.tech/tags/amazon.md>), [amazon-sagemaker](<https://devfeed.tech/tags/amazon-sagemaker.md>), [amazon-sagemaker-ai](<https://devfeed.tech/tags/amazon-sagemaker-ai.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [apis](<https://devfeed.tech/tags/apis.md>), [auto-label](<https://devfeed.tech/tags/auto-label.md>), [build](<https://devfeed.tech/tags/build.md>), [data](<https://devfeed.tech/tags/data.md>), [data-preparation](<https://devfeed.tech/tags/data-preparation.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [development](<https://devfeed.tech/tags/development.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [examples](<https://devfeed.tech/tags/examples.md>), [foundational-100](<https://devfeed.tech/tags/foundational-100.md>), [model](<https://devfeed.tech/tags/model.md>), [model-development](<https://devfeed.tech/tags/model-development.md>), [quality](<https://devfeed.tech/tags/quality.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

Amazon SageMaker AI adds modular data preparation capabilities that let teams generate synthetic examples, label datasets, review uncertain labels, assess quality, and transform data before training or evaluation. Auto-labeling supports built-in and custom tasks; for tasks with fixed answers, multiple models label each row and their agreement supplies a confidence score, with low-confidence rows routed to human review in SageMaker Studio. Synthetic data can be grounded in seed data and documents, and generated pipelines can be inspected and modified. Teams can chain these steps, for example generating task-specific data, reshaping it for SFT, DPO, or RL, and checking for duplicates, leakage, and rule violations. Work is driven through Python SDK APIs or an AI-ML agent skill, with artifacts open to inspection and editing. Data stays in the team's S3 buckets in its chosen Region under an IAM execution role it controls. The capabilities are generally available in US East (N. Virginia) and US West (Oregon).

### Source excerpt

Amazon SageMaker AI now offers AI data preparation capabilities for building high-quality training and evaluation data. Generate synthetic examples, assess and improve datasets, auto-label at scale, and bring in expert review through Python SDK APIs and an agent skill, all from your own notebook or IDE.

## LanceDB raises $57.5M to accelerate AI model training with multimodal data platform

DevFeed: [LanceDB raises $57.5M to accelerate AI model training with multimodal data platform](<https://devfeed.tech/articles/lancedb-raises-57-5m-to-accelerate-ai-model-training-with-multimodal-data-platform-138062.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/17/lancedb-raises-57-5m-to-accelerate-ai-model-training-with-multimodal-data-platform/>)

Author: Kyt Dotson

Published: 2026-11-17T16:30:34Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [LanceDB vector database](<https://devfeed.tech/topics/lancedb-vector-database.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [AI-Ready Data](<https://devfeed.tech/topics/ai-ready-data.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-databases](<https://devfeed.tech/tags/ai-databases.md>), [ai-model-training](<https://devfeed.tech/tags/ai-model-training.md>), [ai-startup-funding](<https://devfeed.tech/tags/ai-startup-funding.md>), [ai-training-data](<https://devfeed.tech/tags/ai-training-data.md>), [artificial-iintelligence](<https://devfeed.tech/tags/artificial-iintelligence.md>), [chang-she](<https://devfeed.tech/tags/chang-she.md>), [data-platform](<https://devfeed.tech/tags/data-platform.md>), [kamil-sindi](<https://devfeed.tech/tags/kamil-sindi.md>), [lancedb](<https://devfeed.tech/tags/lancedb.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [nasdaq-msft](<https://devfeed.tech/tags/nasdaq-msft.md>), [nasdaq-nflx](<https://devfeed.tech/tags/nasdaq-nflx.md>), [news](<https://devfeed.tech/tags/news.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>), [top-story-5](<https://devfeed.tech/tags/top-story-5.md>)

### AI overview

LanceDB raised $57.5 million in a Series B led by Bessemer Venture Partners and Innovation Endeavors, bringing total funding to $98.5 million. Its multimodal lakehouse is built on the open-source Lance format, which supports structured data and large files such as images and video, with rapid retrieval and modification of individual records. The platform connects to existing cloud storage so data stays in place while workloads run across processing infrastructure; LanceDB says this gives teams one access point to search, curate, and prepare datasets that may otherwise span separate systems. The article reports training workflows can involve 10-20 petabytes in a table, sometimes approaching 80-100 petabytes, and says dataset preparation can take weeks or months. Runway, Netflix, and Microsoft are cited as customers. LanceDB plans to use the funding to improve curation, feature engineering, and data loading, and to support video generation, world models, and robotics.

### Source excerpt

LanceDB Inc., the developer of database technology designed to speed up artificial intelligence models, today announced it has raised $57.5 million in early-stage funding to bring a multimodal lakehouse to researchers, model builders and platform teams. Bessemer Venture Partners and Innovation Endeavors led the Series B round. Other investors in the round included Theory Ventures, [...] The post LanceDB raises $57.5M to accelerate AI model training with multimodal data platform appeared first on SiliconANGLE.

## Bridging the Terminology Gap: A Data-Centric Alignment Framework for Multilingual Translation in Cross-Border Corporation

DevFeed: [Bridging the Terminology Gap: A Data-Centric Alignment Framework for Multilingual Translation in Cross-Border Corporation](<https://devfeed.tech/articles/bridging-the-terminology-gap-a-data-centric-alignment-framework-for-multilingual-translation-in-cross-border-corporation-137502.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19858>)

Author: Jinsheng Wei, Song Cheng, Bo Xiao, Ke Zeng

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

Content type: article

Language: en

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

Topics: [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [post-training](<https://devfeed.tech/topics/post-training.md>), [Advanced AI workflows](<https://devfeed.tech/topics/advanced-ai-workflows.md>), [AI-Ready Data](<https://devfeed.tech/topics/ai-ready-data.md>)

Tags: [alignment](<https://devfeed.tech/tags/alignment.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [cross-border](<https://devfeed.tech/tags/cross-border.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [data](<https://devfeed.tech/tags/data.md>), [domain](<https://devfeed.tech/tags/domain.md>), [evaluations](<https://devfeed.tech/tags/evaluations.md>), [framework](<https://devfeed.tech/tags/framework.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [latency](<https://devfeed.tech/tags/latency.md>), [llms](<https://devfeed.tech/tags/llms.md>), [online](<https://devfeed.tech/tags/online.md>), [parallel](<https://devfeed.tech/tags/parallel.md>), [platform](<https://devfeed.tech/tags/platform.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [supervised-fine-tuning](<https://devfeed.tech/tags/supervised-fine-tuning.md>), [training](<https://devfeed.tech/tags/training.md>), [translation](<https://devfeed.tech/tags/translation.md>)

### AI overview

BTG targets terminology gaps, rigid translations, weak instruction following, and high latency in long-instruction translation for cross-border collaboration. It mines cross-lingual term alignments from domain documents, uses LLMs to create terminology-rich parallel corpora, then filters the data with heuristics and XCOMET scoring. Training first uses supervised fine-tuning to inject terminology, then reinforcement learning to refine output preferences. The abstract reports improved terminology accuracy and instruction following in offline evaluations including XCOMET, GEMBA, and Acc@T, as well as online human assessments; it gives no quantitative results or specific latency findings. BTG is reported as deployed on the authors' collaboration platform.

### Source excerpt

arXiv:2611.19858v1 Announce Type: cross Abstract: As globalization accelerates the demand for seamless cross-border enterprise collaboration, multilingual translation remains bottlenecked by insufficient domain-specific terminology, rigid terminology translation, weak instruction-following capabilities, and high inference latency in long-instruction scenarios. To overcome these barriers, we propose Bridging the Terminology Gap (BTG), a data-centric framework. First, BTG mines cross-lingual term alignments from domain documents and leverages LLMs to synthesize terminology-rich parallel corpora. Second, a hybrid filtering strategy combining heuristics and XCOMET scoring ensures high data fidelity. Third, a two-stage training paradigm employs supervised fine-tuning (SFT) for targeted terminology injection, followed by reinforcement learning (RL) to refine output preferences, effectively balancing domain knowledge integration with model reliability. Extensive offline evaluations (e.g., XCOMET, GEMBA, Acc@T) and online human assessments demonstrate significant improvements in terminology accuracy and instruction-following capabilities, validating the effectiveness of our framework, which is now deployed on our cross-border collaboration platform.

[Next page](<https://devfeed.tech/topics/advanced-ai-workflows.md?cursor=WyIyMDI2LTExLTE3VDA1OjAwOjAwKzAwOjAwIiwgIjhkMDVhNmZlLTBhODQtNDM3YS1iYjkzLTdkODk0NzlmOTFiYiJd>)