AI news archive.
19,864 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchLarge-Scale Biomedical RAG: Corpus Scaling, Document Representation, and Efficiency Trade-offs
arXiv:2610.26339v1 Announce Type: cross Abstract: We present a controlled, large-scale evaluation of how corpus scale, document representation, and reranking affect the effectiveness and efficiency of biomedical Retrieva…
- arXiv cs.AIresearchPhyPro-FM: A Physics and Prompt-Guided Graph Foundation Model for Contingency Analysis
arXiv:2610.26351v1 Announce Type: cross Abstract: The continuous evolution of next-generation power systems, characterized by the high penetration of renewable energy resources and dynamic topology variations, poses form…
- arXiv cs.AIagentsWhen Coding Agents Outpace Review: Evaluating PRD-Driven Development in Mature Enterprise Repositories
arXiv:2610.26394v1 Announce Type: cross Abstract: Coding agents increasingly undertake development tasks in mature enterprise repositories, yet existing benchmarks offer limited insight into whether they fulfill complete…
- arXiv cs.AIagentsFake News, Real Crashes: Information Injection Attacks on LLM Financial Trading Agents
arXiv:2610.26438v1 Announce Type: cross Abstract: Large language model (LLM) agents now trade real capital using news as evidence, yet they cannot reliably tell fabricated headlines from real news. Because they read the…
- arXiv cs.AIresearchRethinking LLM Fine-Tuning via Weight Space Reparameterization: Preserving Safety during Downstream Adaptation
arXiv:2610.26443v2 Announce Type: replace-cross Abstract: Fine-tuning large language models (LLMs) on downstream tasks often weakens their previously aligned safety behavior, creating an inherent trade-off between task a…
- arXiv cs.AIresearchALIGN: Aligning Condition Estimation and Expression in Advantage-Conditioned Behavior Cloning
arXiv:2610.26455v1 Announce Type: cross Abstract: Human-collected robot manipulation data often mix successful behavior with retries, pauses, recoveries, and failures. Advantage-Conditioned Behavior Cloning (AC-BC) can e…
- arXiv cs.AIresearchSPAR: Structural Pruning and Action Recovery for Vision-Language-Action Models
arXiv:2610.26487v1 Announce Type: cross Abstract: The limited onboard computing resources of robots make the inference demands of vision-language-action (VLA) models costly to accommodate. Reducing these demands through…
- arXiv cs.AIresearchMitigating Reward Hacking in Preference Reward Models with Evolving Penalties
arXiv:2610.26496v2 Announce Type: replace-cross Abstract: Reinforcement learning for language models has moved from preference reward models to verifiable rewards, which sustain much longer optimization. Preference model…
- arXiv cs.AIresearchLearn the Task, Not the Schedule: Reducing Spurious Cross-Arm Coupling in Bimanual Imitation Learning
arXiv:2610.26507v1 Announce Type: cross Abstract: A bimanual demonstration specifies both what the robot should do and one particular schedule for when each arm should act. Standard imitation learning does not distinguis…
- arXiv cs.AIresearchConcerns in Benchmarking Jailbreaks: An Empirical Study of Attack Success Rates
arXiv:2610.26595v1 Announce Type: cross Abstract: Jailbreak methods are commonly compared using Attack Success Rate (ASR), but these comparisons depend on how responses are generated and judged. In this paper, we examine…
- arXiv cs.AIresearchConducting Reliable Keyword Estimation with Unreliable ASR in Police Audio Data
arXiv:2610.26614v1 Announce Type: cross Abstract: Extracting insights from large audio data sets using human transcription is expensive and time-consuming. Off-the-shelf Automatic Speech Recognition (ASR) systems offer a…
- arXiv cs.AIresearchVaDar: Implicit Goal Conditioning via Value Disaggregation
arXiv:2610.26629v1 Announce Type: cross Abstract: Long-horizon tasks are important for modern AI and robotics, yet pose a challenge for reinforcement learning. Historically, a popular approach has been to decompose a tas…
- arXiv cs.AIresearchReasoning-Aware Multicalibrated Reliability Predictors for Language Models
arXiv:2610.26645v2 Announce Type: replace-cross Abstract: As people increasingly rely on large language models (LLMs) for information, model providers need to communicate meaningful estimates of whether individual respon…
- arXiv cs.AIresearchThe Effectiveness and Durability of Pretraining Data Filtering in Protein Language Models
arXiv:2610.26651v1 Announce Type: cross Abstract: Pretraining filtering is one of the most common safeguards for open-weight language models. Here we study its effectiveness, durability and scalability using the ESM-2 fa…
- arXiv cs.AIresearchThe Development of Sycophancy in Language Models: Tracing Vulnerabilities from Pretraining to Post-Training
arXiv:2610.26655v1 Announce Type: cross Abstract: Language models can change a committed answer to accommodate a user's opinion or pushback without new evidence. However, when this behaviour develops during training and…
- arXiv cs.AIresearchIntra-Rubric Reward Divergence: Mitigating Rubric Hacking through Latent Dimension Recovery and Balancing
arXiv:2610.26685v1 Announce Type: cross Abstract: Rubric hacking limits both performance and training robustness in rubric-based reinforcement learning. By decomposing rubrics into sub-dimensions, we reveal the Intra-Rub…
- arXiv cs.AIresearchHierarchical Token Merging with Recursive Autoencoders for Unsupervised Grammar Induction
arXiv:2610.26695v1 Announce Type: cross Abstract: Hierarchical abstraction is a fundamental principle across many data modalities, enabling complex sequences to be represented compositionally at multiple levels. Learning…
- arXiv cs.AIresearchrecAMPLIFY: Large-Scale Pretraining of Recursive Protein Language Models
arXiv:2610.26747v1 Announce Type: cross Abstract: Protein language models (pLMs) are powerful tools for protein annotation and mutation-effect prediction, but progress has largely depended on scaling model size and data,…
- arXiv cs.AIresearchTokenGate: Adapting LLMs for Fast and Memory-Efficient Prefill with Learned Token Pruning
arXiv:2610.26754v1 Announce Type: cross Abstract: Expert human readers process large inputs (e.g., code repositories and long documents) by skimming to closely read only what matters. On the other hand, language models p…
- arXiv cs.AIresearchGame of Phones: Harnessing Game Theory and LLMs in 'Bot Wars' to Counteract Phone Scams
arXiv:2610.26781v1 Announce Type: cross Abstract: Phone scams impose substantial and growing costs on individuals and financial systems worldwide, yet existing countermeasures impose no resource cost on perpetrators. Str…
- arXiv cs.AIresearchMise-en-Video: Fine-Grained 3D Control for Multi-Shot Video Generation
arXiv:2610.26801v1 Announce Type: cross Abstract: Recent advances in text-to-video generation have enabled high-fidelity single-shot synthesis, yet practical storytelling creation requires multi-shot videos with both str…
- arXiv cs.AIresearchNext Best Action: Causal Constrained Offline RL for Business Action Targeting
arXiv:2610.26817v1 Announce Type: cross Abstract: Next Best Action (NBA) systems must select among customer interventions (e.g., discounts, promotions, and lifecycle actions), maximize long-term economic value, satisfy c…
- arXiv cs.AIresearchAfriBED: a Speech Recognition Bias Evaluation Dataset for Low Resource African Languages
arXiv:2610.26821v2 Announce Type: replace-cross Abstract: Despite advancements in automatic speech recognition systems (ASR), current systems exhibit speaker-attribute bias, i.e., their speech recognition performance is…
- arXiv cs.AIagentsRefCraft: Benchmarking Visual-Reference-Conditioned Agentic Generation through Web Development
arXiv:2610.26826v1 Announce Type: cross Abstract: Beyond basic tool use and coding skills, effectively incorporating visual references into agentic generation is equally essential for applying AI agents to real-world pro…
- arXiv cs.AIresearchKronecker Delta Rule for Long Context Linear Attention
arXiv:2610.26834v1 Announce Type: cross Abstract: We propose KronLSTM, a linear attention architecture based on the Kronecker Delta Rule, a novel update rule for associative matrix memories. KronLSTM extends the xLSTM fa…
- arXiv cs.AIresearchLatComm: Reusable Context Compression for Latent Communication between Large Language Models
arXiv:2610.26840v1 Announce Type: cross Abstract: Systems that combine several large language models (LLMs) often use one context for many questions. When the context is passed between models as text, the receiving model…
- arXiv cs.AIagentsWeAct: Dynamic Weight Switching as Agent Actions
arXiv:2610.26847v2 Announce Type: replace-cross Abstract: Long-horizon agents must learn heterogeneous capabilities from interleaved trajectories. Updating shared parameters to learn new capabilities can degrade previous…
- arXiv cs.AIresearchWhere is Physical RSI capability from? An Observational Scaling Law of LLM-as-policy
arXiv:2610.26848v1 Announce Type: cross Abstract: In this paper, we systematically investigate the mechanisms, scaling behavior, and generalization of physical Recursive self-improvement. We introduce PhysRSI-eval to qua…
- arXiv cs.AIresearchCoopUQ: Robust Multi-Robot Social Navigation via Uncertainty-Aware Cooperative Forecasting
arXiv:2610.26852v1 Announce Type: cross Abstract: Reinforcement learning enables mobile robots to make fast and scalable navigation decisions in constrained human-shared environments. However, directly extending single-r…
- arXiv cs.AIresearchReinforcement Learning with Visually Attributable On-Policy Self-Distillation
arXiv:2610.26856v1 Announce Type: cross Abstract: Multimodal reinforcement learning with verifiable rewards (RLVR) provides reliable task-level supervision, yet offers limited guidance for localizing errors in visual rea…
- arXiv cs.AIagentsOpenVLHarness: Real World Visual Agentic Tool-Use with Multimodal Memory
arXiv:2610.26860v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) have made rapid progress on visual reasoning, yet many tasks require evidence that is difficult to obtain through internal reason…
- arXiv cs.AIresearchPython Tool for Visualizing Variability of Pareto Fronts over Multiple Runs
arXiv:2305.08852v2 Announce Type: replace Abstract: Hyperparameter optimization is crucial to achieving high performance in deep learning. On top of the performance, other criteria such as inference time or memory requir…
- arXiv cs.AIagentsGenerative to Agentic AI: Survey, Conceptualization, and Challenges
arXiv:2504.18875v2 Announce Type: replace Abstract: Agentic AI builds upon Generative AI (GenAI). It constitutes the next major step in the evolution of AI with much stronger reasoning, learning, and interaction capabili…
- arXiv cs.AIresearchMarkovian Linguistic-Temporal Bridge: Unlocking the Potential of LLMs for Time Series Forecasting
arXiv:2505.13175v2 Announce Type: replace Abstract: Adapting pretrained Large Language Models (LLMs) for time series forecasting primarily relies on token-level linguistic-temporal alignment, leading to the stacking of l…
- arXiv cs.AIresearchEmpirical Analysis of Decoding Biases in Masked Diffusion Models
arXiv:2508.13021v4 Announce Type: replace Abstract: Masked Diffusion Models (MDMs) have recently emerged as a promising non-autoregressive paradigm for sequence generation. However, their performance is highly sensitive…
- arXiv cs.AIresearchDistinguishing Performance From Competence in Evaluations of Humanlike Abstract Reasoning
arXiv:2510.02125v5 Announce Type: replace Abstract: AI reasoning models have exceeded human performance on the ARC-AGI-1 benchmark, but does that mean state-of-the-art models have the underlying competence (humanlike abs…
- arXiv cs.AIresearchSELECT: Semantic Erasure Boundary Construction with Sibling Exclusive Counterfactuals
arXiv:2510.16342v2 Announce Type: replace Abstract: Concept erasure aims to suppress undesirable concepts in text-to-image diffusion models while preserving their generative capabilities. Existing methods often redirect…
- arXiv cs.AIagentsReflection-Driven Self-Improving Language Model Agents for Adaptive RAN Slicing
arXiv:2603.10564v2 Announce Type: replace Abstract: Adaptive control in AI-native networks remains challenging because tasks such as dynamic Radio Access Network (RAN) slicing require persistent trade-off decisions among…
- arXiv cs.AIresearchFrom Language Models to Control Policies: Search, Selection, and Verification
arXiv:2603.20537v2 Announce Type: replace Abstract: Industrial process control demands policies that are interpretable and auditable, requirements that black-box neural policies struggle to meet. LLM-driven heuristic syn…
- arXiv cs.AIresearchAI Evaluation Should Require Standardized Item-Level Data Releases
arXiv:2604.03244v3 Announce Type: replace Abstract: This position paper argues that standardized item-level benchmark data should become the default infrastructure for AI evaluation. Current evaluations suffer from under…
- arXiv cs.AIresearchTraceable Context-Enriched Knowledge Graph Generation with Induced Schemas
arXiv:2604.03496v3 Announce Type: replace Abstract: Knowledge graph generation typically relies either on predefined ontologies or on schema-free extraction. Ontology-driven pipelines enforce consistent typing but requir…
- arXiv cs.AIresearchSchema-Aware Planning and Hybrid Knowledge Toolset for Reliable Knowledge Graph Triple Verification
arXiv:2604.04190v2 Announce Type: replace Abstract: Knowledge Graphs (KGs) provide a critical foundation for many AI systems, yet their automated construction inevitably introduces noisy triples that undermine data trust…
- arXiv cs.AIresearchGypscie: A Cross-Platform AI Artifact Management System
arXiv:2604.10311v3 Announce Type: replace Abstract: Artificial Intelligence (AI) models, encompassing both traditional machine learning (ML) and more advanced approaches such as deep learning and large language models (L…
- arXiv cs.AIresearchRADD: Retrieval-Anchored Discrete Diffusion for Multi-Modal Knowledge Graph Completion
arXiv:2604.25693v3 Announce Type: replace Abstract: Multi-modal knowledge graph completion aims to infer missing links by integrating structural, visual, and textual information. However, global retrieval can leave fine-…
- arXiv cs.AIagentsVirtual Speech Therapist: A Clinician-in-the-Loop AI Agent for Speech Therapy Planning
arXiv:2605.01101v3 Announce Type: replace Abstract: This paper presents a proof of concept of the Virtual Speech Therapist (VST), an intelligent multi-agent based system designed to support stuttering assessment and indi…
- arXiv cs.AIagentsMOIRA: Policy Memory and Belief Revision for Language-Model Agents in Non-Stationary Pair Trading
arXiv:2605.01954v2 Announce Type: replace Abstract: Many sequential decision-making problems exhibit hierarchical structure, where high-level semantic choices constrain downstream actions and feedback is delayed and ambi…
- arXiv cs.AIagentsCan LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations
arXiv:2605.11633v2 Announce Type: replace Abstract: Operational disaster response goes beyond damage assessment, requiring responders to integrate multi-sensor signals, reason over road networks, populations and key faci…
- arXiv cs.AIagentsHerculean: An Agentic Benchmark for Financial Intelligence
arXiv:2605.14355v4 Announce Type: replace Abstract: As AI agents advance, the central question is no longer whether they can solve isolated financial tasks, but whether their capabilities generalize across complete profe…
- arXiv cs.AIagentsNexus : An Agentic Framework for Context-Aware Time Series Forecasting
arXiv:2605.14389v2 Announce Type: replace Abstract: Language models have emerged as highly capable reasoners, yet their performance in time series forecasting remains uneven across domains and levels of contextual ground…
- arXiv cs.AIagentsConflict-Aware Multi-Agent Reasoning via Signed Graph Modeling
arXiv:2605.19418v2 Announce Type: replace Abstract: LLM-based multi-agent systems (MAS) seek to improve reasoning by combining diverse perspectives from multiple agents. However, aggregation may overlook supportive, cont…