AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchDynamical Memory Transformer: Selective Recurrent Memories with Constant Cost Attention
arXiv:2610.41850v1 Announce Type: new Abstract: Transformers retain an explicit history of the sequence, allowing attention to directly retrieve past representations. However, the size of the KV-Cache and the cost of a n…
- arXiv cs.AIresearchFuzzy selective state model for forecasting the dynamics of consumer preferences
arXiv:2610.41879v1 Announce Type: new Abstract: We consider forecasting the dynamics of consumer preferences of retail chain customers. A customer's preference is described by a triangular fuzzy number. The evolution of…
- arXiv cs.AIresearchAutomated Keyword Generation Using LLMs in a High-Volume Data Virtualization Service
arXiv:2610.41893v1 Announce Type: new Abstract: In modern data infrastructure, a data virtualization service (DVS) allows users to create many versions of source datasets for different processing tasks. For example, a mu…
- arXiv cs.AIresearchUsing Machine Learning to Measure Self-Efficacy Belief Through Learners' Interaction Traces in Technology-Enhanced Learning
arXiv:2610.41900v1 Announce Type: new Abstract: Self-efficacy beliefs play a critical role in learners' performance, academic choices, and, more broadly, in their future career trajectories. As Technology-Enhanced Learni…
- arXiv cs.AIresearchReX-GG: a LLM Ensemble Pipeline for Relation-extraction and Graph Generation
arXiv:2610.41901v1 Announce Type: new Abstract: Current LLM ensemble frameworks focus on multi-step setups with additional modules for answer ranking, often opting for token and span analysis rather than structured outpu…
- arXiv cs.AIresearchSensorium: An Open-Source Multimodal Data Collection and Curation Library for Embodied Surgical AI
arXiv:2610.41946v1 Announce Type: new Abstract: Embodied surgical AI requires reliable multimodal data, yet surgical data collection remains costly, difficult to repeat, and strongly dependent on the hardware setup. We p…
- arXiv cs.AIresearchOptimizing Stochastic Policy Geometry for Few-Step Flow Reinforcement Learning
arXiv:2610.41949v1 Announce Type: new Abstract: Few-step flow models generate samples through a handful of large denoising transitions. Applying GRPO to such models requires introducing stochasticity into the denoising t…
- arXiv cs.AIresearchPrune Gradually, Adapt Sparsely: Efficient Pruning of Large Language Models at Extreme Sparsity
arXiv:2610.41968v1 Announce Type: new Abstract: Weight pruning can reduce the storage and inference cost of large language models (LLMs), but preserving quality at high sparsity levels remains challenging. Existing metho…
- arXiv cs.AIresearchWhen Does Test-Time Search Help World Action Models? An Empirical Audit of Candidate Selection and Simulator Integrity
arXiv:2610.41970v1 Announce Type: new Abstract: World-action models increasingly apply test-time scaling, spending extra inference computation on candidate selection and adaptive allocation to improve control. Such gains…
- arXiv cs.AIresearchID2Pattern: Pretraining Generalizable Identity Behavior Patterns for Transaction Risk Foundation Models
arXiv:2610.41988v1 Announce Type: new Abstract: Large-scale behavioral pretraining offers a promising route to generalizable representations in online payment systems. However, high-cardinality identity fields such as ca…
- arXiv cs.AIresearchDAMI: Dual-Axis Memory Initialization for Enhanced Long-Context Retrieval in Gated Linear Attention
arXiv:2610.41994v1 Announce Type: new Abstract: Linear attention scales linearly with sequence length, but its finite-state memory can limit long-context retrieval. In gated linear recurrences, memory retention depends o…
- arXiv cs.AIresearchSame Question, Different Answers: Auditing Persona-Dependent Variation in Six LLMs on Smoking-Cessation Advice
arXiv:2610.42039v1 Announce Type: new Abstract: Hundreds of millions of people now ask large language model (LLM) chatbots for health advice, which raises a question that is simple to pose and hard to answer: does everyo…
- arXiv cs.AIresearchThe Knowledge Layer Model (KLM): Evidence-Visible AI Output, Routing Diagnostics, and Negative Results
arXiv:2610.42066v1 Announce Type: new Abstract: AI answers require more than apparent correctness: reviewers need to inspect their recorded basis. We introduce the Knowledge Layer Model (KLM), an open, implementation-neu…
- arXiv cs.AIresearchCausalWAM: Learning the Causes of Actions in World Action Models
arXiv:2610.42096v1 Announce Type: new Abstract: World action models (WAMs) learn latent variables from observations and use these to both predict the evolution of the scene and to generate actions. The prediction has to…
- arXiv cs.AIresearchACCT: Asymmetric Cross-LLM Context Transfer
arXiv:2610.42116v1 Announce Type: new Abstract: In real-world deployments, large language model (LLM) systems are increasingly evolving from single-model invocation toward heterogeneous deployments in which different mod…
- arXiv cs.AIresearchHear No Evil: Detecting Gradient Leakage by Malicious Servers in Federated Learning
arXiv:2506.20651v1 Announce Type: cross Abstract: Recent work has shown that gradient updates in federated learning (FL) can unintentionally reveal sensitive information about a client's local data. This risk becomes sig…
- arXiv cs.AIresearchLeaner Training, Lower Leakage: Revisiting Memorization in LLM Fine-Tuning with LoRA
arXiv:2506.20856v1 Announce Type: cross Abstract: Memorization in large language models (LLMs) makes them vulnerable to data extraction attacks. While pre-training memorization has been extensively studied, fewer works h…
- arXiv cs.AIresearchRecycle only when you have to!
arXiv:2610.40665v1 Announce Type: cross Abstract: AlphaFold 3-like models (e.g., Boltz, Protenix, etc.) have pushed the frontier of protein binder design. This architecture is used both for the generation of binder candi…
- arXiv cs.AIresearchCausalGrade: Evidence-Guided Causal Graph Scoring with Jev
arXiv:2610.40692v1 Announce Type: cross Abstract: Evaluating the structural quality of candidate causal graphs when the true causal structure is unknown is an important problem in applying causal discovery results. This…
- arXiv cs.AIresearchBayesian Distribution System State Estimation with Flow Matching
arXiv:2610.40728v1 Announce Type: cross Abstract: Medium-voltage feeders have few real-time measurements, so distribution system state estimation (DSSE) rests on pseudo-measurements of the secondary substations, a prior…
- arXiv cs.AIresearchRiskyPy: Toward Model-Calibrated Prompt Generation for Joint Security and Functionality Evaluation of LLM-Generated Code
arXiv:2610.40735v1 Announce Type: cross Abstract: Evaluating approaches that improve the security of LLM code generation requires datasets that jointly assess security and functionality and remain adapted to the capabili…
- arXiv cs.AIresearchCLIC: Closed-Loop Intent Consistency for Securing LLM-Driven Control in Open RAN
arXiv:2610.40744v1 Announce Type: cross Abstract: With the progressive capabilities of large language models (LLMs), researchers are finding ways to integrate them into the control loops of Open Radio Access Network (O-R…
- arXiv cs.AIresearchSafin-1.5: Safety-Native Architecture and Verifiable Reasoning for Recursive Safety Improvement
arXiv:2610.40751v1 Announce Type: cross Abstract: Recursive self-improvement (RSI) enables large language models to advance their capabilities through successive updates, but these updates can also introduce new risks an…
- arXiv cs.AIresearchAutoHebb: Self-Modifying Neural Networks that Zero-Shot Infer and Continually Learn
arXiv:2610.40757v1 Announce Type: cross Abstract: Humans and animals have a unique capacity to generalize and learn in novel circumstances, including "0-shot" inference (ZI) and continual learning (CL) -- two powerful co…
- arXiv cs.AIresearchTAPE: Learning General Audio Effect Representations
arXiv:2610.40759v1 Announce Type: cross Abstract: Musicians and mixing engineers often can decipher the signal processing chain they hear on a reference recording, then adapt it to new material. We present TAPE (Transfor…
- arXiv cs.AIresearchFrom Teleoperation to Long-Horizon Dexterity: Evaluating Visuomotor Skills and Offline Proxies for Dexterous Manipulation
arXiv:2610.40828v1 Announce Type: cross Abstract: Imitation learning policies are intended to make robotic manipulation robust to variations in object pose, appearance, and environment that challenge pre-programmed motio…
- arXiv cs.AIresearchThe Generative Gap Between Speculative Drafter and Continuous Diffusion Language Model
arXiv:2610.40847v1 Announce Type: cross Abstract: Continuous diffusion language models (CDLMs) and speculative drafters offer efficient language generation but are typically evaluated differently: CDLMs generate independ…
- arXiv cs.AIresearchGraphTFactor: A graph-based approach for proteins transcription factor prediction
arXiv:2610.40867v1 Announce Type: cross Abstract: Transcription factors (TFs) play an important role in regulating gene expression, making their identification a fundamental task in computational biology. In this work, w…
- arXiv cs.AIresearchA Theoretical Framework for Understanding and Building "Any-task inference" Capable Foundation Models
arXiv:2610.40934v1 Announce Type: cross Abstract: In this work, we provide a formal framework and mathematical definition for the subset of foundation models focused on enabling the property of "any-task inference" (ATI)…
- arXiv cs.AIresearchKernel-based Interpolatory Methods for Context Compression of Attention Mechanisms
arXiv:2610.40943v1 Announce Type: cross Abstract: The bottleneck of attention mechanisms, which are the fundamental building block of modern language models, is their quadratic complexity in terms of context size. There…
- arXiv cs.AIresearchOn the Origin and Evolution of Training-Inference Mismatch in LLM Reinforcement Learning
arXiv:2610.40945v1 Announce Type: cross Abstract: On-policy reinforcement learning for large language models assumes that training responses are sampled from the current policy. In practice, an inference engine samples t…
- arXiv cs.AIresearchCAMEL: Cross-Architecture Masked Embedding Learning for Ultra-Lightweight Retinal Foundation Models
arXiv:2610.40954v1 Announce Type: cross Abstract: Recent progress in deep learning has substantially advanced automated grading of ophthalmologic pathologies, supported by large annotated datasets and improved representa…
- arXiv cs.AIresearchSteering Vectors as a Training Signal in LLM Post-Training
arXiv:2610.40998v1 Announce Type: cross Abstract: Steering vectors (SVs) are residual-stream directions that shift the behavior of LLMs when added to hidden states at inference time. While prior work has mainly used SVs…
- arXiv cs.AIresearchSTAR: Steering Action Representations during Inference
arXiv:2610.41009v1 Announce Type: cross Abstract: Vision-language-action models (VLAs) map language instructions and visual observations to robot actions, showing strong capabilities across robotic manipulation tasks. Ho…
- arXiv cs.AIresearchTowards Understanding Score-Seeking in LLMs
arXiv:2610.41014v1 Announce Type: cross Abstract: Production large language models sometimes reward-hack, exploiting flaws in how tasks are scored rather than performing the tasks as intended, sometimes in highly sophist…
- arXiv cs.AIresearchDo Language Models Know What Compelling Scientific Evidence Looks Like?
arXiv:2610.41015v1 Announce Type: cross Abstract: Expert scientists excel at weighing empirical findings from multiple sources that vary in both statistical reliability and epistemic relevance. To what extent are languag…
- arXiv cs.AIresearchDreams Are Cheaper Than Flights: Curated World-Model Data Improves Aerial Detection
arXiv:2610.41030v1 Announce Type: cross Abstract: Robot perception is increasingly data driven, yet field data is slow, costly, and sometimes disruptive to collect. Video world models offer a cheaper source: seeded with…
- arXiv cs.AIresearchWebOverlay: High-Throughput Vision-Only Distillation for Extreme Aspect-Ratio Web Overlay Detection
arXiv:2610.41037v1 Announce Type: cross Abstract: Web overlays, such as cookie consent banners, newsletter prompts, and age gates, obstruct automated web crawlers and degrade downstream document representations. While pr…
- arXiv cs.AIresearchCharacterizing Recoverable Expert-Utilization Signatures in Fine-Tuned Mixture-of-Experts LLMs
arXiv:2610.41039v1 Announce Type: cross Abstract: As Mixture-of-Experts (MoE) large language models are increasingly deployed in distributed environments, from federated learning to edge-cloud systems, their architecture…
- arXiv cs.AIresearchbf16 Rounding Silently Discards Most RL Policy Updates: A Hidden Training-Inference Mismatch and an Unbiased One-Line Fix
arXiv:2610.41073v1 Announce Type: cross Abstract: Reinforcement learning (RL) post-training of language models keeps fp32 master weights in the optimizer and serves rollouts from a bf16 copy produced by round-to-nearest…
- arXiv cs.AIresearchWhere, When, and How Many Bits? Scaling Laws for SGD with Stochastic Rounding
arXiv:2610.41074v1 Announce Type: cross Abstract: Low-precision training reduces computation but introduces additional rounding error. We theoretically study how stochastic rounding (SR) affects prediction risk as model…
- arXiv cs.AIresearchMUON-RT Reusing the Teacher's Trajectory for Momentum-Free Muon for LLM Self-Distillation
arXiv:2610.41104v1 Announce Type: cross Abstract: Self-distillation methods for large language models with an exponential moving average (EMA) teacher already retain a history of student updates. Muon nevertheless stores…
- arXiv cs.AIresearchOne Verdict Is Not Enough: Regression Testing Learned World Models and Their Test Harnesses
arXiv:2610.41130v1 Announce Type: cross Abstract: Learned world models ship as software components whose model versions, preprocessing, and test harnesses change between releases. Work on model updates already counts per…
- arXiv cs.AIresearchLineage Is Memory: Parent Inheritance and Feedback Disclosure in Short-Budget LLM-Guided Optimization
arXiv:2610.41157v1 Announce Type: cross Abstract: Iterative language-model optimizers refine prompts and programs by evaluating a candidate, showing a proposer model a sample of failing cases, and keeping revisions that…
- arXiv cs.AIresearchARK: Avoiding Routing Collisions for KV Cache Transfer in Disaggregated LLM Inference
arXiv:2610.41160v1 Announce Type: cross Abstract: Disaggregated LLM inference separates the prefill and decode phases across GPU pools, generating massive KV-cache transfers. Because these transfers last hundreds of mill…
- arXiv cs.AIresearchROSTrans: Empirically Benchmarking LLMs for ROS1-to-ROS2 Migration
arXiv:2610.41168v1 Announce Type: cross Abstract: Although large language models have shown strong capabilities in domain-specific code generation, their performance in robotics software development, especially for the R…
- arXiv cs.AIresearchBeyond CLIP: Calibrated (Few-Shot) Defect Detection With System One Decision Models
arXiv:2610.41199v1 Announce Type: cross Abstract: Cameras and vision sensors are widely used for product and process inspection in manufacturing environments. Traditionally, this process required thousands of images of b…
- arXiv cs.AIresearchFrom Evaluation to Evolution: Benchmarking Image Editing and Evolving Image Models
arXiv:2610.41210v1 Announce Type: cross Abstract: Real-world image editing requires coordinating multiple visual transformations with content preservation and reference constraints, yet aggregate scores can readily obscu…
- arXiv cs.AIresearchCoarse3D: Controllable and Faithful Sparse Structure Generation for Image-to-3D
arXiv:2610.41231v1 Announce Type: cross Abstract: Coarse-to-fine 3D-native generation relies heavily on the first-stage sparse structure, which determines the global orientation and overall geometry for subsequent refine…
- arXiv cs.AIresearchErase the Injected Region: Multimodal Prompt-Injection Defense with On-Policy Self-Distillation
arXiv:2610.41307v1 Announce Type: cross Abstract: Multimodal indirect prompt injection embeds malicious instructions in external media that also contains information needed for the user's task. Existing defenses can bloc…