AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchTargeting World Models to Compromise Robot Learning Pipelines
arXiv:2606.09499v2 Announce Type: replace-cross Abstract: World models have recently seen a rapid growth in both their popularity and capability as more data efficient tools for generating robot training data or simulati…
- arXiv cs.AIresearchAn LLM-Native Psychometric Instrument Reveals a Self-Report--Behavior Gap Across 25 Models
arXiv:2606.09843v4 Announce Type: replace-cross Abstract: Do large language models' (LLMs') answers to self-report questionnaires predict how they behave? Prior work finds they do not, but it uses human personality inven…
- arXiv cs.AIresearchMixedPEFT: Combining Multiple PEFT Methods with Mixed Objectives for Unsupervised Domain Adaptation
arXiv:2606.22272v2 Announce Type: replace-cross Abstract: Applying pre-trained language models to new domains through full fine-tuning is computationally expensive and prone to catastrophic forgetting. To address this li…
- arXiv cs.AIresearchWalk fast but be careful: Understanding Parallel Sampling in Masked Diffusion
arXiv:2606.22976v2 Announce Type: replace-cross Abstract: In this paper, we use random walks on graphs as a verifiable sandbox for studying parallel sampling strategies in masked diffusion models (MDMs). We train an MDM…
- arXiv cs.AIresearchGeological text descriptions in ill-posed inverse problems: insights from learned hydraulic-conductivity inversion
arXiv:2606.24967v3 Announce Type: replace-cross Abstract: Hydraulic-conductivity inversion is ill posed: even complete head observations can leave structural ambiguity. Geological text descriptions can supply additional…
- arXiv cs.AIresearchAI-Assisted Computational Reproducibility on the FABRIC Testbed
arXiv:2606.25879v2 Announce Type: replace-cross Abstract: Computational reproducibility remains difficult despite being central to scientific research. In this paper, we show how the international FABRIC testbed, combine…
- arXiv cs.AIresearchCurvature-Guided Module Localization for Low-Rank Detoxification of Backdoored Large Language Models
arXiv:2606.30899v2 Announce Type: replace-cross Abstract: Backdoor attacks pose a serious threat to large language models (LLMs) by causing otherwise benign systems to produce attacker-specified malicious behavior when a…
- arXiv cs.AIresearchFAR: Failure-Aware Retry for Test-Time Recovery and Continual Policy Improvement
arXiv:2607.01111v2 Announce Type: replace-cross Abstract: Robot policies inevitably encounter failures when deployed in real environments. Naive retries often repeat the same mistakes, while many existing recovery method…
- arXiv cs.AIresearchAdapting Generalist Vehicle Models for High-Speed MPC Across Terrains
arXiv:2607.13319v4 Announce Type: replace-cross Abstract: High-speed off-road autonomy requires precise closed-loop control for a target vehicle while remaining robust across changing terrains. Recent forward kinodynamic…
- arXiv cs.AIresearchReViV: Reconstructing the Viewer and the View in 4D from Monocular Egocentric Video
arXiv:2607.17790v3 Announce Type: replace-cross Abstract: Egocentric devices, such as wearable front-facing cameras, provide a unique perspective for capturing the continuous interaction between a human viewer and the su…
- arXiv cs.AIresearchModPack: Extensible Teleoperation Interface for Bimanual Mobile Manipulation
arXiv:2607.19479v2 Announce Type: replace-cross Abstract: Existing teleoperation systems are often tailored to specific robot hardware and task domains, limiting their scalability and adaptability. We present ModPack, a…
- arXiv cs.AIresearchReinforcement Learning for Code Optimization
arXiv:2607.25970v2 Announce Type: replace-cross Abstract: RL for code correctness is now established: have the model generate a program, run it against hidden test cases, and reward solutions that pass. Extending this to…
- arXiv cs.AIresearchReToken: Improving Long-Context VLMs with Visual Retrieval Token
arXiv:2607.28627v2 Announce Type: replace-cross Abstract: Long visual contexts challenge vision-language models: performance degrades as the number of distractors grows, and processing all tokens at once can exceed GPU m…
- arXiv cs.AIresearchSign Language Video Synthesis via Loss-Guided Multi-Expert GANs
arXiv:2608.13368v3 Announce Type: replace-cross Abstract: This preliminary technical report presents a framework for sign language video synthesis using a loss-guided multi-expert Generative Adversarial Network (GAN) to…
- arXiv cs.AIresearchLittleLearner: Language Models Under Pedagogically Controlled Knowledge Exposure
arXiv:2608.13545v2 Announce Type: replace-cross Abstract: Modern language models are trained on heterogeneous web-scale text corpora. Consequently, studying knowledge and skill acquisition is difficult, as prior exposure…
- arXiv cs.AIresearchThe Shadow Price of Intelligence: Quality Degradation in LLM Inference as a Supply Chain Problem
arXiv:2608.23986v3 Announce Type: replace-cross Abstract: Large language model providers are compute constrained, and a common response to congestion is to degrade service. A degraded answer fails with some probability,…
- arXiv cs.AIresearchHow Language Models Organize and Structure Moral Knowledge
arXiv:2608.27402v2 Announce Type: replace-cross Abstract: How do large language models (LLMs) organize moral knowledge? Models detect moral content broadly, but detection is a low bar. We ask whether they go further, dis…
- arXiv cs.AIresearchOnline Estimation of Dynamic Origin-Destination Matrices Using Reinforcement Learning with Link-Flow Propagation Guidance
arXiv:2608.30317v2 Announce Type: replace-cross Abstract: Dynamic origin-destination (OD) matrix estimation calibrates time-dependent input demand for simulations to reproduce observed link flows. Reinforcement learning…
- Apple Machine Learning ResearchresearchNormalizing Trajectory Models
Diffusion-based models decompose sampling into many small Gaussian denoising steps, an assumption that breaks down when generation is compressed to a few coarse transitions. Existing few-step methods address this through…
- LessWrongresearchAre OpenAI's math results "creative" in an important way?
Many people expect the AI paradigm to approximately AGI complete, and expect it to smoothly scale to AI that can do innovative science. i.e. the kind of science you'd need to do to develop a plague from scratch that actu…
- SiliconANGLE AIresearchOpenAI publishes 722 AI-generated math discoveries in major scientific milestone
OpenAI Group PBC has published 722 math papers that it generated using an unreleased artificial intelligence model. Some of the papers, which were posted to GitHub late Tuesday, prove long-running hypotheses. Others disa…
- The Register AI + MLresearchArgonne scientists create chatty X-ray microscope that zooms in where you tell it to
Talking to an X-ray nanoprobe beamline is far better than talking to a toaster, we assume
- LessWrongresearchTraining with conflicting values can induce CoT override
CoT override: when a model makes a decision in its CoT but ignores it in its response TLDR We train models on two conflicting traits: (1) caring about the user’s health, and (2) promoting smoking Those models do not gene…
- LessWrongresearchVisual explainer of empirical Neural Tangent Kernels
There is a new mech interp method in town. I found it difficult to get my head around at first, but once I understood it, I found it beautiful. I am not an expert in this topic, and mostly write this up for me to get som…
- WIRED AIresearchThese Researchers Made AI Drive a Toyota Corolla to Get In-N-Out
Three engineers put GPT, Claude, and Grok in charge of a real car. Only one of them was successful.
- NVIDIA Developer BlogresearchThe Machines that Make the Machines
How we taught robots to assemble GB300 tester trays and what it taught us about robot learning, mechanical intelligence, and good old-fashioned engineering The...
- The DecoderresearchZuckerberg's Biohub leads a $1.8 billion push to build AI models that predict cell behavior
Biohub, the research organization backed by Mark Zuckerberg and Priscilla Chan, is coordinating a $1.8 billion initiative to train AI models that predict cell behavior. Meta, Google DeepMind, Isomorphic Labs, and the US…
- NVIDIA Developer BlogresearchScaling Decision Optimization to 100 Million Variables and Beyond with mPDLP in NVIDIA cuOpt
Supply chain problems are expanding across more SKUs, lanes, and constraints than ever before, while energy grids are balancing more distributed sources in real...
- Guardian AIresearchOpenAI’s release of mathematical findings draws concerns from experts
Leaders worry OpenAI is not doing due diligence to vet results and that AI models aren’t accessible to broader field of mathematicians OpenAI has astounded mathematicians after releasing hundreds of new mathematical find…
- The Verge AIresearchGoogle invests millions in Mark Zuckerberg’s efforts to create a ‘virtual cell’
Google DeepMind, Meta, and AI drug discovery startup Isomorphic Labs are jointly investing $300 million into an initiative to create a "virtual cell" that researchers can use to combat disease, as reported earlier by Reu…
- IEEE Spectrum AIresearchHiPHI: A Large-Scale Benchmark for High-Precision Human Motion and Object Interaction
This White Paper gives robotics researchers and engineers an overview of a new large-scale motion capture dataset built to close the data gap limiting humanoid robot learning. It also shows how policies trained on the da…
- LessWrongresearchBenchmarking Jev against no-CoT LLMs
TL;DR. We ran Jev 1.13 — TypeSafe's non-autoregressive model, which answers questions with probabilities and cannot emit text, on the ThinkFast no-CoT suite and Neel Nanda’s NCRI . Jev’s capabilities are very jagged. As…
- TechRadar AIresearchThe next phase of the AI era: why building R&D talent is now a make-or-break investment
Mainstream AI news from a large tech publication.
- The DecoderresearchOpenAI dumps 372 AI-generated math proofs on GitHub, telling the academic world to keep up
OpenAI has published 372 AI-generated mathematical results on GitHub, including Lean formalizations for machine verification. Each result consumed about three hours of ChatGPT Pro compute on average. But 25 Fields Medal…
- Ai2 BlogresearchNow in Nature: Retrofitting language models to operate over bytes
The technique behind Bolmo, Ai2’s fully open byte-level language models, is now published in Nature, with new checkpoints showing the approach generalizes beyond Olmo to other model families.
- Guardian TechnologyresearchScientists who put faith in technology less likely to take climate action
‘Techno-optimism’ may lead to reliance on unproven solutions at expense of societal changes needed Scientists who believe technology will largely solve the problems caused by the climate crisis are less likely to engage…
- Latent Spaceresearch[AINews] Quasi-Riemann-Hypothesis: OpenAI publishes 722 math papers solving 90 of the top 500 open math problems; “the most significant moment” in >100 years of mathematics
Our head hurts.
- Simon Willison's WeblogresearchQuoting Jake Boggan
I was a graph theory junkie long ago and even moved to Budapest for awhile to study among the greats. While I was there I started working on Barnette's Conjecture which came to occupy my thoughts over the next 24 years o…
- arXiv cs.AIresearchFluidPD: In-Place Elasticity for SLO-Aware Prefill-Decode Disaggregated LLM Serving
arXiv:2610.06917v1 Announce Type: new Abstract: Prefill-decode disaggregation is becoming a common architecture for LLM serving because it separates two phases with distinct execution patterns and SLO objectives. Existin…
- arXiv cs.AIresearchAnchor Divergence for Semantic Geometry in Contrastive Learning
arXiv:2610.06919v1 Announce Type: new Abstract: This paper concerns how semantic context determines geometry in learned vector representations. Similarity is typically measured using cosine similarity, which provides a s…
- arXiv cs.AIresearchMetonymic Circuits for Abstract Concept Grounding in Vision Transformers
arXiv:2610.06928v1 Announce Type: new Abstract: We study how Vision Transformers ground abstract concepts (e.g., angry) when training data provide limited direct referential evidence. We hypothesize a metonymic grounding…
- arXiv cs.AIresearchWhen better traffic forecasts fail to improve signal control: a layered diagnostic study of forecast-to-decision value
arXiv:2610.06992v1 Announce Type: new Abstract: Improved traffic forecasts do not necessarily yield better signal-control decisions. We investigate this gap through a layered diagnostic study using 29 days of reconstruct…
- arXiv cs.AIresearchJoint upper-bound coverage and route-choice utility: an empirical evaluation on two urban proxy tasks
arXiv:2610.06995v1 Announce Type: new Abstract: Whether more accurate traffic forecasts or higher uncertainty coverage improve route decisions is unclear. We evaluate this question with a frozen protocol that separates s…
- arXiv cs.AIresearchWhen to Rethink: Learning Multi-Perspective Self-Verification for Vision-Language Models
arXiv:2610.07018v1 Announce Type: new Abstract: Vision-language models (VLMs) have achieved strong performance in multimodal reasoning, yet they remain prone to generating plausible but incorrect answers. Self-verificati…
- arXiv cs.AIresearchBeyond Refusal Patterns: Safe-Role Internalization for Robust and Generalizable LLM Safety Alignment
arXiv:2610.07023v1 Announce Type: new Abstract: Large Language Models (LLMs) have achieved remarkable capabilities but remain vulnerable to jailbreak attacks that elicit harmful or unsafe outputs. Existing safety alignme…
- arXiv cs.AIresearchOffline AI Modules: Voice-First Offline Architecture, Hardware Reference Stack, Quantization and Benchmarking
arXiv:2610.07026v1 Announce Type: new Abstract: The Offline AI Modules workstream enables practical, low-power, and community-accessible deployment of voice-first AI systems that operate fully offline. Designed for Afric…
- arXiv cs.AIresearchJIVEAdapter: A Multi-Task Additive Low-Rank Adapter via Joint and Individual Variation Explained (JIVE)
arXiv:2610.07036v1 Announce Type: new Abstract: Parameter-efficient fine-tuning adapts pretrained models at a fraction of the cost of full fine-tuning, yet most low-rank adapters are single-task and represent each weight…
- arXiv cs.AIresearchInference-Time Projection for Physically Valid Biomolecular Diffusion Models
arXiv:2610.07037v1 Announce Type: new Abstract: AlphaFold 3-style cofolding models predict biomolecular complexes with high structural accuracy, yet a large fraction of their outputs are physically invalid: chains overla…
- arXiv cs.AIresearchSmart Content Ingestion for Generative AI Workloads
arXiv:2610.07091v1 Announce Type: new Abstract: The evolution of machine learning has progressively changed where intelligence resides in an AI system. In conventional machine learning the task, data representation, labe…
- arXiv cs.AIresearchSmall Language Models for Smart Data Model Classification at the Edge: A Cost-Aware Hybrid Approach
arXiv:2610.07093v1 Announce Type: new Abstract: The rapid proliferation of heterogeneous data sources within the Internet of Things (IoT) across domains such as smart cities, energy management, and environmental monitori…