AI news archive.
10,661 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchGraph Hierarchical Recurrence for Long-Range Generalization
arXiv:2605.18387v2 Announce Type: replace-cross Abstract: Graph Neural Networks and Graph Transformers have become central to graph learning, combining expressive representation learning with sample-efficient inductive b…
- arXiv cs.AIresearchThe Fragility of Chain-of-Thought Monitoring Across Typologically Diverse Languages
arXiv:2605.27901v2 Announce Type: replace-cross Abstract: Chain-of-thought (CoT) monitoring has been proposed as a promising safety mechanism for detecting misaligned behavior in large language models. However, its relia…
- arXiv cs.AIresearchMemGuard: Preventing Memory Contamination in Long-Term Memory-Augmented Large Language Models
arXiv:2605.28009v2 Announce Type: replace-cross Abstract: Memory-augmented large language models extend reasoning beyond a fixed context window by maintaining long-term memory across interactions. However, existing memor…
- arXiv cs.AIresearchBORA: Bridging Offline Reinforcement Learning and Online Residual Adaptation for Real-World Dexterous VLA Models
arXiv:2605.30226v3 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) policies provide strong behavioral priors for dexterous manipulation, yet adapting them on real robots remains challenging because hi…
- arXiv cs.AIresearchTrait-space Monitoring for Emergent Misalignment During Supervised Finetuning
arXiv:2606.07631v2 Announce Type: replace-cross Abstract: Emergent misalignment (EM) occurs when narrow finetuning induces dangerous behavior outside the finetuning task. Detecting this shift through repeated behavioral…
- arXiv cs.AIresearchTRIAGE: Dialectical LLM Reasoning for Explainable Risk Prediction on Irregularly Sampled Medical Time Series
arXiv:2606.09030v2 Announce Type: replace-cross Abstract: Clinical early warning systems built on irregularly sampled medical time series (ISMTS) from electronic health records must deliver continuous risk scores for pat…
- arXiv cs.AIresearchCHILLGuard: Towards Fine-Grained Chinese LLM Safety Guardrail with Scalable Data Construction and Model-aware Preference Alignment
arXiv:2606.15396v2 Announce Type: replace-cross Abstract: Malicious content generated from large language models (LLMs) could pose severe safety risks and ethical concerns. While existing LLM safety guardrails excel in E…
- arXiv cs.AIresearchConstitutional Value Potentials: reading and steering internal priority margins in language models
arXiv:2606.15420v2 Announce Type: replace-cross Abstract: Safety evaluations test a policy on prompts that omit the incentive information deployment supplies: a commission, a performance score, a dashboard naming which a…
- arXiv cs.AIresearchVision-language models for chest radiography do not always need the image
arXiv:2606.17710v3 Announce Type: replace-cross Abstract: Vision-language models that answer questions about chest radiographs are evaluated by their accuracy on labels derived from radiology reports. High benchmark accu…
- arXiv cs.AIresearchPepti-drift: Scalable Safe-Active Peptide Generation Without Inference-Time Guidance
arXiv:2606.27824v3 Announce Type: replace-cross Abstract: Therapeutic peptides are a promising drug modality, but their generation must satisfy multiple therapeutic constraints. We introduce BindSafe-PepBench, a fixed-bu…
- arXiv cs.AIresearchSelf-conditioned Flow Map Language Models via Fixed-point Flows
arXiv:2607.00714v2 Announce Type: replace-cross Abstract: Self-conditioning is a core technique that enhances continuous flow-based language models, where the model learns to denoise generated text by conditioning on its…
- arXiv cs.AIresearchReading Between the Dots: Decoding Hidden Computation across Filler Tokens
arXiv:2607.03502v2 Announce Type: replace-cross Abstract: Frontier LLMs can perform multi-step reasoning over content-free filler tokens like dots or counting sequences, producing correct answers with no visible chain-of…
- arXiv cs.AIresearchFuncBridge: Towards Functional Tool-Use Generalization via Keypoint Trajectory Reasoning
arXiv:2607.05780v2 Announce Type: replace-cross Abstract: While humans readily repurpose a book, a stone, or a shoe to drive a nail, robots trained on specific tools fail to transfer the same function to novel ones -- a…
- arXiv cs.AIresearchEvaluating Neural Decompilation of Dart AOT Binaries: Fine-Tuning, Metric Validity, Specification Leakage, and Reliability
arXiv:2607.06125v2 Announce Type: replace-cross Abstract: We present an execution-based evaluation of neural decompilation for Dart ahead-of-time binaries and an audit of what its scores measure. Across six archived adap…
- arXiv cs.AIresearchMeteorology-driven Causal Nowcasting of Fugitive Landfill Emissions from Measured Coupling Timescales
arXiv:2608.14254v2 Announce Type: replace-cross Abstract: Which meteorological processes control exposure to fugitive gases downwind of a source, and on what timescales, have largely been inferred from dispersion theory…
- arXiv cs.AIresearchCoupled-cluster molecular properties across the main group that extrapolate beyond training size
arXiv:2608.18346v2 Announce Type: replace-cross Abstract: Coupled-cluster theory defines the accuracy standard for molecular electronic-structure properties but scales too steeply for routine application, whereas density…
- arXiv cs.AIresearchAdam at the Edge of Stability: Adaptive Feedback, Provable Oscillation, and Gradient Reversal
arXiv:2608.20638v2 Announce Type: replace-cross Abstract: The edge-of-stability (EoS) phenomenon of full-batch Adam has been widely observed, yet its underlying dynamical mechanism remains poorly understood. In this pape…
- arXiv cs.AIresearchCredal Large Language Models for Semantic Commitment under Uncertainty
arXiv:2608.23244v2 Announce Type: replace-cross Abstract: Large language models (LLMs) often produce fluent but incorrect answers with unwarranted confidence. A central limitation is that standard LLMs represent uncertai…
- arXiv cs.AIresearchTalked Out of the Truth: Sycophancy in the Reasoning Chains of Multimodal Models
arXiv:2608.28623v3 Announce Type: replace-cross Abstract: Large multimodal reasoning models (LMRMs) are increasingly capable, largely through generating explicit chain-of-thought reasoning before answering, but in langua…
- arXiv cs.AIresearchStratified Consistency Distillation for Natural Language Formalization
arXiv:2608.30258v3 Announce Type: replace-cross Abstract: Neurosymbolic reasoning has shown promising success in addressing complex reasoning tasks by combining large language models (LLMs) and symbolic solvers. While th…
- arXiv cs.AIresearchWhat Do Scan-Derived Class Prototypes Add? Disentangling Supervision, Prototype Content and Query Protocol in Recognition over Frozen Foundation Features
arXiv:2609.04381v2 Announce Type: replace-cross Abstract: A scan supplies labeled images and a geometric reference. We separate their contributions in a recognizer whose scan-derived prototype matrix acts as a supervised…
- arXiv cs.AIresearchTANGO: Humanoid Navigation in Cluttered Environments with a Whole-Body Vision-Language-Action Model
arXiv:2609.09158v2 Announce Type: replace-cross Abstract: We study the problem of navigating cluttered indoor environments with a humanoid robot. Unlike conventional methods that model navigation as a 2D path planning pr…
- Latent SpaceresearchAcademia is for Ambition — Alex Zhang, MIT
We catch up with RLM first author Alex Zhang, MIT PhD, on Jev, PhD masxing, and the future of harnesses.
- Apple Machine Learning ResearchresearchLanguage Discrimination Improves Linguistic Learning in Multilingual Speech Models
Multilingual self-supervised speech models can benefit from sharing information across languages, but under a matched total pretraining data budget they still fall short of monolingual models. We show that strengthening…
- Apple Machine Learning ResearchresearchLimits of Confidence in Diffusion
Discrete diffusion, including remasking and uniform-state samplers, generate a sequence by writing multiple token positions per step, drawing each from a per-position distribution and choosing which positions to write fr…
- Google ResearchresearchOur Project Suncatcher prototype satellite is in orbit.
Google's prototype satellite for Project Suncatcher, built in partnership with Planet, launched into orbit aboard the Transporter-18 rideshare mission with SpaceX.
- OpenAI NewsresearchThe eternal complement
Advanced AI may matter most for the routine work behind breakthrough ideas. Explore why execution could shape the next economy and the pace of progress.
- Ars Technica AIresearchWith most information hidden, the game Stratego had stumped AI—until now
Adding in a second neural network that guesses the identity of hidden pieces was key.
- LessWrongresearchLearning Steganography Is Easy, Learning Steganographic Reasoning Is Hard
This post summarises our paper Learning Steganography Is Easy, Learning Steganographic Reasoning Is Hard , accepted as an oral at the NeurIPS 2026 Workshop on Trustworthy AI for Good (AI4GOOD). Code, configurations and r…
- AI Stack ExchangeresearchHow does a VGG-based Style-Loss incorporate color information?
I've recently been reading a lot about style transfer, its applications and implications. I understand what the Gram matrix is and does. I can program it. But one thing that has been boggling me is: how does the VGG styl…
- LessWrongresearchFLF’s Epistemic Case Study Competition: Results
This summer, FLF ran a contest which asked entrants to help us “find the best workflows and methodologies for using AI to produce reliable, trustworthy knowledge bases, grounded in real-world cases”. We awarded just over…
- The DecoderresearchAI beats Stratego's greatest player, ending one of the last human strongholds in board games
The AI system Ataraxos has decisively beaten the best Stratego player of all time. The board game is a tough test for AI because both sides set up their pieces face down. Google Deepmind fell short in 2023 despite a mult…
- AI Stack ExchangeresearchNormalizing image embeddings after exponential moving average
I am working on object tracking in image plane. For each detected object in an image, I get a 256-dimensional embedding vector using a reID model. Since the embeddings can be noisy, for the same object in the image plane…
- arXiv cs.AIresearchImproving OCR Faithfulness via Gated and Attenuated On-Policy Distillation
arXiv:2609.38282v1 Announce Type: new Abstract: Vision-language models may rewrite anomalous text in images into linguistically plausible expressions, compromising OCR transcription faithfulness. Sequence-level task rewa…
- arXiv cs.AIresearchCARAT: Do Materials LLMs Reason or Recite?
arXiv:2609.38340v1 Announce Type: new Abstract: When a materials LLM answers a question about crystal structure, does it reason from the structure or copy an answer already printed in its input? Accuracy cannot tell: a s…
- arXiv cs.AIresearchExamining Variation in How Guided AI Tutors Resolve Student Impasses
arXiv:2609.38346v1 Announce Type: new Abstract: When a student is stuck, a tutor faces the assistance dilemma: help given too early can hinder productive struggle, while help withheld too long leaves the student in a fru…
- arXiv cs.AIresearchBeyond Mode Collapse: Generating Diverse Synthetic Expert Conversations via Generative Flow Networks
arXiv:2609.38359v1 Announce Type: new Abstract: High quality synthetic data is central to post training LLMs for adaptive AI applications that represent the diverse expert strategies and decisions in conversations. Promp…
- arXiv cs.AIresearchAligned Data Can Induce Misalignment via Context Confusion
arXiv:2609.38379v1 Announce Type: new Abstract: Large language models (LLMs) are frequently updated for various use cases, where filtering out misaligned training samples is a common practice for preventing post-update m…
- arXiv cs.AIresearchFine-Tuning Diffusion Language Models with Context Selection and Target Weighting
arXiv:2609.38385v1 Announce Type: new Abstract: Supervised fine-tuning of discrete diffusion language models masks some response tokens and trains the model to recover their original values from the visible context. The…
- arXiv cs.AIresearchDecode-Latency Feedback Prefill: A Model-Free Controller and Its Generalization Limits
arXiv:2609.38386v1 Announce Type: new Abstract: Concurrent autoregressive inference creates a fundamental interference problem: prefilling a newly arrived long prompt can delay tokens for requests that are already decodi…
- arXiv cs.AIresearchMetaPersona: Task-Grounded Synthetic Populations from Empirical Social Science
arXiv:2609.38392v1 Announce Type: new Abstract: Personas used to seed LLM social simulations face a cold-start problem: existing methods lack a principled basis for deciding which attributes to include and how to assign…
- arXiv cs.AIresearchSimTrace: Grounded Multimodal User Trajectories Generation for Online User Modeling
arXiv:2609.38397v1 Announce Type: new Abstract: Virtual clients offer a cost-effective approach to support applications such as A/B testing, recommender system development, and interface evaluation. However, building the…
- arXiv cs.AIresearchArgGYM: A Procedural, Engine-Verified Benchmark for Structured Defeasible Reasoning
arXiv:2609.38409v1 Announce Type: new Abstract: Recent progress in large language model reasoning has been driven by benchmarks and reinforcement learning environments with automatically verifiable rewards, particularly…
- arXiv cs.AIresearchReach Into The CHOIR: Free-List Elicitation Uncovers Distinct Model Voices in LLM Ensembles
arXiv:2609.38448v1 Announce Type: new Abstract: Open-ended LLM homogeneity can create false plurality when several systems appear to offer independent perspectives while returning the same familiar default. Single-pass a…
- arXiv cs.AIresearchDemographic Pluralism: Inference-Time Modeling of Pluralistic Human Preference Distributions
arXiv:2609.38555v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used in culturally sensitive settings, where alignment requires representing diverse preferences within populations. Yet exist…
- arXiv cs.AIresearchDefining and Categorising Human-AI Interactions in Clinical Trials: A Multidimensional Human-AI Classification Approach
arXiv:2609.38559v1 Announce Type: new Abstract: This paper examines human-AI interactions (HAIIs) in clinical trials and presents a multidimensional categorisation framework that classifies interactions according to AI t…
- arXiv cs.AIresearchTowards Model as a Library: Offline, Community-Sourced AI for Low-Resource African Languages
arXiv:2609.38574v1 Announce Type: new Abstract: Large language models are frequently proposed as a route to AI-powered services for African communities, but they are least reliable exactly where the need is greatest: all…
- arXiv cs.AIresearchConditional Generation of Creative Chess Puzzles with Diffusion Models
arXiv:2609.38577v1 Announce Type: new Abstract: While modern language models demonstrate impressive generative capabilities, they often struggle with constrained, counter-intuitive creative tasks. To address this limitat…
- arXiv cs.AIresearchSense and Sensitivity: Benchmarking LLM Clinical Triage Recommendations with Physician Experts
arXiv:2609.38600v1 Announce Type: new Abstract: As large language models (LLMs) are increasingly used in clinical settings, it is critical to evaluate their reliability under realistic variation in clinical text. We stud…
- arXiv cs.AIresearchWhen Scientific Contradictions Are Lost in Translation
arXiv:2609.38621v1 Announce Type: new Abstract: Two scientific findings can disagree without contradicting each other. Determining whether they conflict requires knowing whether they describe comparable measurements. We…