AI news archive.
19,861 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIagentsPEARL: Adaptive Prefill-Decode Execution with Elasticity for Agentic Reinforcement Learning
arXiv:2609.35158v1 Announce Type: new Abstract: Multi-turn rollout dominates the cost of agentic reinforcement learning (RL). Asynchronous execution and elastic GPU resources can accelerate this stage, but adding rollout…
- arXiv cs.AIresearchFONDANT: Strong and Best-Effort Planning via Antichains
arXiv:2609.35160v1 Announce Type: new Abstract: A classical solution concept in fully observable nondeterministic (FOND) planning, is the strong policy (aka winning strategy in the closely related area of reactive synthe…
- arXiv cs.AIresearch5W1H+Which: Context-Valid Semantic Indexing with Progressive Ontology Binding
arXiv:2609.35184v2 Announce Type: replace Abstract: Semantic indexing must preserve reusable source information without committing too early to one business schema, while providing checkable premises for reasoning. We pr…
- arXiv cs.AIresearchBeneath the Tokens: A Performance Engineering Study of Multi-Token Prediction in GPU-Accelerated LLM Inference
arXiv:2609.35188v1 Announce Type: new Abstract: Autoregressive large language model inference repeatedly invokes the target model to generate one token at a time, making generation sensitive to GPU memory movement and se…
- arXiv cs.AIagentsASCT: Attentive Search over Counterfactual Trees for Credit Assignment in Agentic Reinforcement Learning
arXiv:2609.35215v1 Announce Type: new Abstract: Terminal utility evaluates a complete agentic workflow, but learning requires credit for the decisions within it. We introduce Attentive Search over Counterfactual Trees (A…
- arXiv cs.AIagentsEP-Mem: Elastic Privacy Memory for Social Relationship-Aware LLM Agents
arXiv:2609.35233v1 Announce Type: new Abstract: Large language model (LLM) agents face critical privacy risks when acting as delegates in human-agent-human communication. To prevent such breaches, agents must understand…
- arXiv cs.AIagentsTowards Reliable AI Data Scientists: Data Agents with Workflow Harnesses
arXiv:2609.35255v1 Announce Type: new Abstract: Large language model agents are increasingly deployed for data-intensive work, yet reliable data analysis requires more than general-purpose reasoning and ad hoc tool augme…
- arXiv cs.AIresearchImprint Reader: From Weight-Update Readout to Behavioral Intervention
arXiv:2609.35261v1 Announce Type: new Abstract: As language models take a growing role in AI development, a natural aspiration is for them to reflect on their own learning process, as humans do, and use that reflection t…
- arXiv cs.AIresearchTextual User Taste: Natural-Language User Context for Foundation-Model Recommender System at Scale
arXiv:2609.35285v1 Announce Type: new Abstract: Foundation model recommender systems require user context that can be consumed by large language models, reasoned over, and refined through natural-language interaction. Tr…
- arXiv cs.AIresearchThe Argument and the Letterhead: Source-Position Coherence in AI Evaluation
arXiv:2609.35286v1 Announce Type: new Abstract: An argument can be surprising coming from a particular speaker without being a bad argument. Do AI evaluators keep these judgments apart? Two preregistered descriptive stud…
- arXiv cs.AIagentsEvoIn: Bridging Evolution and Internalization for Agent Fine-Tuning
arXiv:2609.35290v2 Announce Type: replace Abstract: Recent work has explored improving agents by jointly evolving their harnesses and models, but often takes a ''potpourri'' approach that bundles together new tools, new…
- arXiv cs.AIresearchAbGaze: Attentive Geometric Representation Learning for End-to-End Antibody Design
arXiv:2609.35296v1 Announce Type: new Abstract: Computational antibody design requires representations that capture the geometric patterns underlying antigen--antibody interactions, yet existing approaches often rely on…
- arXiv cs.AIresearchTraining-Free Clinical Reasoning through Medical Ontologies and Cognitive Mapping: A Symbolic-Probabilistic Knowledge Graph Framework
arXiv:2609.35298v1 Announce Type: new Abstract: Most clinical prediction systems learn patient-variable-outcome associations; we investigate a training-free diagnostic paradigm mapping patient observations to explicit me…
- arXiv cs.AIresearchNarrowing the Horizon: Quantifying Topic Saliency Shifts in Generative Monoculture
arXiv:2609.35302v1 Announce Type: new Abstract: As Large Language Models (LLMs) become central to how we access and share information, they play an increasingly powerful role in shaping global knowledge. However, as thes…
- arXiv cs.AIagentsReliability Engineering for AI Systems: Challenges, Methods, and Directions
arXiv:2609.35316v1 Announce Type: new Abstract: AI reliability concerns whether an AI system performs its intended function dependably over a stated period and under stated operating conditions, with stated evidence. As…
- arXiv cs.AIagentsHyper Algorithm Design Agent: Evolving Learnable Optimizer from Zero
arXiv:2609.35328v3 Announce Type: replace Abstract: Meta-Black-Box Optimization (MetaBBO) is one of the highlights in the recent AI for Optimization trend. This paradigm's bi-level workflow leverages the learnable algori…
- arXiv cs.AIagentsTMCS: Tool-Grounded Multi-Agent Reasoning for Compositional Chemical Problem Solving
arXiv:2609.35336v1 Announce Type: new Abstract: Despite the promise of Large Language Models (LLMs) in computational chemistry, rigorous combinatorial chemistry problems remain difficult because they require quantitative…
- arXiv cs.AIresearchJev thinks "I don't know'', but doesn't say it: Introducing Sys1Cal-v1 Dataset for Probability Calibration
arXiv:2609.35342v1 Announce Type: new Abstract: The appearance of Jev marked the era of System One Models, foundation models that return structured decisions with probability distributions rather than text. Aside from lo…
- arXiv cs.AIresearchJailbreaks for Black-Box Uncertainty Quantification in Large Reasoning Models
arXiv:2609.35350v1 Announce Type: new Abstract: While Large Reasoning Models (LRMs) excel at complex reasoning, alignment through reinforcement learning often induces systemic overconfidence. In production environments,…
- arXiv cs.AIresearchDon't Inoculate Everything: Stratified Inoculation Prompting Narrows Backdoor Triggers and Preserves Desired Traits
arXiv:2609.35356v3 Announce Type: replace Abstract: Supervised fine-tuning can teach language models undesired behaviours alongside desired ones. Inoculation prompting (IP) aims to limit unwanted generalisation by reques…
- arXiv cs.AIresearchA decision-support system applied to Law: Reasoning and explainability of the decision
arXiv:2609.35370v1 Announce Type: new Abstract: The emergence of the digital transition brought an increasing need to control the processing of digital information, including in Law Enforcement Agencies (LEAs). At the EU…
- arXiv cs.AIresearchStructural Alignment for Reliable Industrial AI: Bridging Physical Reality, Data, Models, and Human Intent
arXiv:2609.35400v1 Announce Type: new Abstract: Artificial intelligence is increasingly deployed in critical industrial domains, including healthcare, energy grids, subsurface exploration, where failures can have severe…
- arXiv cs.AIagentsSelf-Adapting Group of Experts for Multi-Agent Reasoning
arXiv:2609.35412v1 Announce Type: new Abstract: Multi-agent systems bring together language model agents with different roles to propose, review, and refine solutions. Each agent's response depends on its model's capabil…
- arXiv cs.AIresearchBuilding Transformation Layers for Riemannian Neural Networks
arXiv:2609.35436v2 Announce Type: replace Abstract: Recently, deep neural networks on manifold-valued representations have garnered significant attention across various machine learning applications. One recent focus is…
- arXiv cs.AIresearchJust Initialize: A Training-Free Initialization Component for Large-Scale Routing Optimization
arXiv:2609.35443v3 Announce Type: replace Abstract: Large-scale routing problems are difficult to solve efficiently as their search spaces grow rapidly with problem size. Existing approaches primarily improve the optimiz…
- arXiv cs.AIagentsAutoBCI: Forecast-Guided Agentic Neural Architecture Discovery for EEG-Based Brain--Computer Interfaces
arXiv:2609.35456v1 Announce Type: new Abstract: EEG-based brain-computer interfaces support a broad range of applications, yet designing decoding architectures that perform well across diverse tasks remains challenging.…
- arXiv cs.AIagentsA.D.A.M.O. (Agent for language-Driven Actions with Multimodal Observations): A Visual-Symbolic Framework for Virtual Humans
arXiv:2609.35463v1 Announce Type: new Abstract: Creating believable vh requires the coherent integration of perception, reasoning, and action mediated by language. A central challenge is to combine these components into…
- arXiv cs.AIresearchWhy Deterministic PRM Guidance Underperforms in Discrete Diffusion Reasoning
arXiv:2609.35472v1 Announce Type: new Abstract: Discrete diffusion language models (dLLMs) expose a denoised solution at every step, which makes process reward model (PRM) guidance look like a way to spend compute at tes…
- arXiv cs.AIagentsSRHarness: A Harness for Agentic Symbolic Regression
arXiv:2609.35501v2 Announce Type: replace Abstract: Recent agentic symbolic regression approaches increasingly rely on large language models to analyze data, select scientific operations, and refine hypotheses over long…
- arXiv cs.AIagentsMechBench: Can AI Scientific Agents Discover Mechanisms Beyond Phenomenal Laws?
arXiv:2609.35515v1 Announce Type: new Abstract: Scientific discovery requires not only recovering mathematical laws that describe observable behavior, but also identifying the mechanisms that generate them. Existing benc…
- arXiv cs.AIagentsARISE: Adapting to Evolving Capability Gaps in Agentic Reinforcement Learning
arXiv:2609.35532v1 Announce Type: new Abstract: As a long-horizon agent improves through experience, previously observed weaknesses may recede while new limitations emerge, continually changing what it still needs to lea…
- arXiv cs.AIagentsContinuous Context Management
arXiv:2609.35540v1 Announce Type: new Abstract: Long-horizon large language model (LLM) agents commonly retain their complete interaction history until compaction is triggered at a predefined threshold. We study Continuo…
- arXiv cs.AIresearchRareDx: Controlled Knowledge Integration and Graph-Grounded Policy Optimization for Rare-Disease Diagnosis
arXiv:2609.35549v2 Announce Type: replace Abstract: Rare-disease diagnosis is a long-tail reasoning problem: phenotypes are incomplete, individual disorders are sparsely documented, and relevant evidence is distributed a…
- arXiv cs.AIagentsBaRe-Mem: Bayesian Reliability Memory for Robust and Adaptive Agent Consultation
arXiv:2609.35551v1 Announce Type: new Abstract: In multi-agent systems, reliable consultation is challenging because advisor capabilities vary across tasks, and misleading information can make consultation worse than aut…
- arXiv cs.AIagentsFrom Search to Research: Exploring Search Scaling in Autonomous Quantitative Factor Mining
arXiv:2609.35559v1 Announce Type: new Abstract: Inference scaling has been shown to improve large language model (LLM) performance, and this principle naturally extends to autonomous LLM agents through increased search b…
- arXiv cs.AIagentsRSI-Master: Structuring Experiments to Guide Autonomous Model Improvement
arXiv:2609.35561v2 Announce Type: replace Abstract: Recursive self-improvement (RSI) seeks to enable AI systems to participate in improving their own capabilities. A concrete pathway is autonomous model development, wher…
- arXiv cs.AIresearchRepresentation Alignment as a Bottleneck in LLM-Based Retrosynthesis Planning
arXiv:2609.35571v1 Announce Type: new Abstract: While LLMs show promise in general reasoning, symbolic planning in chemistry remains a bottleneck. Direct ''SMILES-to-PDDL'' attempts fail because they force models to jugg…
- arXiv cs.AIagentsShare-Borne AI Virus: Memory-Hopping Attacks Across LLM Agents
arXiv:2609.35576v1 Announce Type: new Abstract: Large language models are increasingly deployed as stateful assistants that retain information across interactions and use tools to read, modify, and create persistent arti…
- arXiv cs.AIresearchIMC-CLINIC: Coupled Loss-Informed Newton Iterations for Clipping in Analog In-Memory Computing
arXiv:2609.35586v1 Announce Type: new Abstract: Analog in-memory computing (IMC) offers a promising path toward energy-efficient large language model (LLM) inference by executing matrix multiplications (MatMul) directly…
- arXiv cs.AIresearchSource-preserving alignment for robust evidence localization in scientific PDFS
arXiv:2609.35588v1 Announce Type: new Abstract: Scientific information-extraction systems often return a claim with an evidence string, which users must locate in the original PDF. This is challenging because the extract…
- arXiv cs.AIresearchSignatures of semantic search in the activations of large language models
arXiv:2609.35599v2 Announce Type: replace Abstract: When recalling lists of concepts (e.g., animals) during the semantic fluency task (SFT), both humans and large language models (LLMs) organise their output into cluster…
- arXiv cs.AIresearchTCSAlgBench: Benchmarking Automated Proving for Research-Level Theoretical Computer Science
arXiv:2609.35606v1 Announce Type: new Abstract: Large language models perform strongly on competition mathematics, but their research-level reasoning remains difficult to evaluate systematically. Theoretical computer sci…
- arXiv cs.AIresearchFrom cacophony to hierarchy: a principled framework for assessing AI consciousness
arXiv:2609.35618v2 Announce Type: replace Abstract: The question of AI consciousness is one of the most urgent pre-emptive problems in philosophy and computer science, yet progress is hampered by a cacophony of competing…
- arXiv cs.AIresearchRIDE: Reference-Anchored Inference-Time Diffusion Editing for Scaffold Hopping
arXiv:2609.35623v1 Announce Type: new Abstract: Scaffold hopping is a critical task in drug discovery, which seeks to discover new, structurally distinct molecules that share key functional groups and similar 3D shape wi…
- arXiv cs.AIresearchVerifiable Visual Rewards Transfer from Synthetic Scenes to Natural Prompts
arXiv:2609.35641v1 Announce Type: new Abstract: Precise instruction following in image generation, such as satisfying object counts and spatial relations, remains an open challenge at least in part because it is learned…
- arXiv cs.AIresearchNot All Thinking is Created Equal: Latent Reasoning Discovers a Recurrent Search Algorithm for Depth Generalization
arXiv:2609.35643v1 Announce Type: new Abstract: Large Language Models can perform multi-step reasoning and improve task performance through different forms of intermediate computation, from token-based traces to computat…
- arXiv cs.AIagentsPhoneCLI: From App Interfaces to Callable Commands for Mobile Agents
arXiv:2609.35671v2 Announce Type: replace Abstract: Mobile GUI agents operate through a perception--action loop: at each step they screenshot the device, invoke a vision--language model (VLM), and emit an action. It is s…
- arXiv cs.AIresearchVerifier Errors in RLVR: Reward Hacking, Limits of Feedback, and Selective Control
arXiv:2609.35677v1 Announce Type: new Abstract: In reinforcement learning with verifiable rewards (RLVR), imperfect verifiers can reward incorrect responses, creating opportunities for reward hacking. Using gradient flow…
- arXiv cs.AIagentsReport: Progressive Disclosure of Agent Skills
arXiv:2609.35692v1 Announce Type: new Abstract: Users of Workday's deployed LLM-based agents often request features which can be addressed by defining named procedures, also known as skills, in the LLM context, effective…
- arXiv cs.AIresearchReasoning with Continuous Latent Diffusion
arXiv:2609.35694v3 Announce Type: replace Abstract: We introduce the Continuous Embedding Diffusion Reasoner (CEDR), a training and inference pipeline for continuous latent diffusion models that achieves strong performan…