AI & Machine Learning · Computer Science & AI
Artificial Intelligence
The newest papers in Artificial Intelligence that ReadSearch has read, newest first. Each links to its own page, with its authors, date, where it was published and the original.
The newest 5,000 papers, from 11 October 2026 to 5 November 2026: 4,998 arXiv preprints and 2 OpenAlex records.
Get the few papers worth your time in Artificial Intelligence every morning.
Follow Artificial Intelligence and ReadSearch reads its new papers each day, then brings you the few worth reading, each with its key finding in a sentence.
Newest papers, page 92
WebAnything: From the Wild Web to Scalable Agent Environments
Web agents require diverse, executable, and resettable website environments for reproducible training and evaluation.Locating Sycophancy Across Training Stages in Open Language Models
Language model assistants sometimes change a correct answer when the user suggests a wrong one.Spokes, not a Ring: Revisiting the Weekday Manifold in Language Models
Recent work has hypothesised that large language models may represent some concepts as multi-dimensional manifolds instead of single directions.What Does an Ontology Catch? Curated, LLM-Generated and Data-Derived Ontologies as Checkers of LLM Agents over Knowledge Graphs
LLM agents that answer questions over Web knowledge graphs are increasingly wrapped in checks derived from the graph's ontology, on the assumption that a formal schema makes a reliable judge.Responsible Quantum Machine Learning for Medical Imaging: A Fairness Audit Across Sensitive Subgroups
Quantum machine learning (QML) is increasingly explored for medical classification, yet its behaviour under subgroup-aware fairness criteria remains understudied.DeFT-RAG: Decomposing Multi-Hop Questions into Self-Contained Sub-Questions with Reusable Answers
Retrieval-augmented generation (RAG) lets a large language model (LLM) answer questions by retrieving relevant passages from a large corpus.Learning Lotteries with Minimal Stakeholder Violations from Membership Queries
An AI developer often has several candidate versions of a system and can deploy a mixture of them, sending each request to a randomly chosen version, as in A/B testing.Measuring Where Intent Gets Lost in Agentic Smart Contract Synthesis
AI agents can now write software from a plain-English description, but code that runs is not the same as code that does what was asked.Where the Error Lives: Spectral Representations of Learned Kohn-Sham Hamiltonians
Kohn-Sham density functional theory (DFT) is the standard method for computing the electronic structure of molecules and materials, but its self-consistent-field (SCF) iterations make it expensive…Detecting Covert Collusion between AI Agents through Partner-Outsider Positive Listening
AI agents that share a secret convention can coordinate through messages that look innocent, so a monitor that reads only the messages can miss them.Evaluating and improving AI agents for sequencing protocol understanding
Existing scientific-agent benchmarks rarely evaluate whether agents can reconstruct the exact molecular structures and generation workflows specified by sequencing protocols, a capability critical…Seq@K: Evaluating Language Models Under Sequential Feedback
Language-model agents differ fundamentally from prior intelligent systems in their ability to learn from natural language in context.Echo: Scaling Visual Reinforcement Learning with Off-policy Experience
Scaling visual reinforcement learning (RL) with GPU-accelerated simulation is computationally expensive due to rendering and visual-policy inference costs.What Constitutes Food Similarity? Visual Representations and Retrieval in Indian Cuisine
Food similarity underlies retrieval, recommendation, substitution, and other food computing tasks, yet food representations are typically evaluated by identity preservation.PairCLIP-SWA: Similarity-Guided Balancing and Stochastic Weight Averaging for Generalized Deepfake Image Detection with OpenCLIP
Deepfake detectors are almost always deployed as the single checkpoint that scored best on a validation split, yet the reliability of that choice is rarely measured.Verification Learning Theory: A Process Model of Checking and Retention in AI-Assisted Judgment
Accurate AI assistance can improve immediate judgments while reducing opportunities to practise the operations needed for independent performance.When More Planning Hurts: Learned Planners for World Models
On PushT, our fixed action updater (1,505 parameters) solves 8 of 16 scenes reserved from our configuration selection.The Anatomy of Socio-Demographics in Pluralistic Offensiveness Modeling: When and Where It Helps, and the Role of Reasoning
Subjective judgments such as offensiveness often vary substantially across individuals, motivating pluralistic modeling approaches that incorporate annotators' socio-demographic information to…Scan Before You Install: Detecting and Mitigating Local and Composite Vulnerabilities in Skills
An agent skill packages natural-language instructions, scripts, and reference files that an LLM-based agent loads for a task.Think Twice: The Metis-1.6 Technical Report
Metis is a bet that better data, better training methods and new architectures can buy what a bigger parameter count cannot, at a fraction of the cost.