AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchImproving Human-Robot Teamwork in Urban Search and Rescue Through Episodic Memory of Prior Collaboration
arXiv:2606.18836v3 Announce Type: replace-cross Abstract: Effective human-robot teamwork requires robots to adapt to partners, situations, and task dynamics from the start of an interaction. In the MATRX Urban Search and…
- arXiv cs.AIresearchTransitNet: A Compact Attention-Augmented Deep Learning Framework for Low-SNR Transit Blind Searches
arXiv:2606.18932v3 Announce Type: replace-cross Abstract: Motivated by the observational incompleteness of intermediate-to-long-period Earth-size planets, we present TransitNet, a compact attention-augmented deep-learnin…
- arXiv cs.AIresearchReinforcement learning to improve large language model-based automated code compliance systems
arXiv:2606.22402v2 Announce Type: replace-cross Abstract: Large language model (LLM)-based approaches for automated code compliance (ACC) of building regulations are prone to generating incorrect and hallucinated compute…
- arXiv cs.AIresearchThe Model as One Rater Among Several: Measuring Political Positions in Data-Sparse Regions with a Language-Model Panel
arXiv:2606.23042v2 Announce Type: replace-cross Abstract: Most instruments for measuring political positions, such as manifesto coding, expert surveys and text-scaling models, were built and validated on Western party sy…
- arXiv cs.AIresearchRepresentation and Reference Selection in Training-Free Synthetic Image Attribution
arXiv:2607.12052v2 Announce Type: replace-cross Abstract: Synthetic image attribution aims at identifying the generator responsible for a given AI-generated image. Training-free reference-based attribution methods are ea…
- arXiv cs.AIresearchCalibrated Selective Prediction Using Deep Ensembles for ROI-Based Thyroid Nodule Ultrasound Classification Under Dataset Shift: A Retrospective Evaluation
arXiv:2607.12075v2 Announce Type: replace-cross Abstract: Background: Deep learning models can classify thyroid nodules on ultrasound, but reliable clinical decision support also requires calibrated probabilities, uncert…
- arXiv cs.AIresearchGUEST: Preliminary Guidelines for Using and Evaluating GenAI Tools to Support Systematic Literature Reviews
arXiv:2607.24991v2 Announce Type: replace-cross Abstract: Context: Generative AI (GenAI) and Large Language Models (LLMs) are increasingly used for academic tasks in software engineering and beyond, including systematic…
- arXiv cs.AIresearchHow Do Developers Edit AI-Generated Code, and How Well Do LLMs Infer Their Edits? Insights from 53.6K Developer Edits
arXiv:2607.25130v3 Announce Type: replace-cross Abstract: Imperfections in AI-generated code require that software developers modify generations through manual code edits. However, little is known about how developers ed…
- arXiv cs.AIresearchDecodable but Not Actionable: Auditing the Representation-Behavior Gap in Frozen LLMs for Geometric Constraints
arXiv:2608.17843v2 Announce Type: replace-cross Abstract: Large language models (LLMs) perform well on structured reasoning tasks, but what they encode and whether encoded information guides their behavior remain unclear…
- arXiv cs.AIresearchEvidence-Gated Task and Motion Planning with Vision-Language Models
arXiv:2608.20084v2 Announce Type: replace-cross Abstract: Robots executing long-horizon manipulation tasks from natural-language instructions must reason about both semantic task structure and geometric feasibility. Hier…
- arXiv cs.AIresearchTowards AI-Assisted Clinical Trial Matching: Practical Considerations, Multicenter Evaluation, and Real-World Deployment
arXiv:2609.01202v2 Announce Type: replace-cross Abstract: Clinical trials are essential for evidence-based medicine and drug development, but many fail because of insufficient patient enrollment. Although there is growin…
- arXiv cs.AIresearchOutageDiT: A Foundation Model for Power Outage Forecasting and Scenario Generation
arXiv:2609.01896v2 Announce Type: replace-cross Abstract: Power-outage planning requires scenarios before an event occurs. These scenarios must represent uncertainty in magnitude, timing, and duration while preserving te…
- arXiv cs.AIresearchThe Direction of Ignorance: LLMs Encode and Temper a Unigram Prior
arXiv:2609.02959v2 Announce Type: replace-cross Abstract: What does a language model predict when it has few clues?The answer lies in its unembedding geometry: a single direction of the unembedding matrix encodes the uni…
- arXiv cs.AIresearchGrounded and Faithful P&ID Reasoning: Constraining Vision-Language Models with Recovered Evidence Graphs
arXiv:2609.05880v2 Announce Type: replace-cross Abstract: Piping and Instrumentation Diagrams (P&IDs) are the authoritative maps of process plants: isolation, maintenance, and HAZOP decisions depend on what connects to w…
- arXiv cs.AIresearchMapping the Emerging Social Science of Large Language Models
arXiv:2609.07598v2 Announce Type: replace-cross Abstract: As large language models (LLMs) blur the boundary between cognition and computation, they reshape knowledge production and collective life. We combine a curated r…
- arXiv cs.AIresearchLearning Multimodal One-step Flow Policy via Value-weighted Optimal Transport
arXiv:2609.15883v2 Announce Type: replace-cross Abstract: Offline reinforcement learning aims to learn a policy solely from fixed datasets, which often contain multimodal action distributions. Flow policies can naturally…
- arXiv cs.AIresearchIterative Calibration and Learning: Mitigating Teacher Bias in LLM Distillation without Target-Domain Reward Feedback
arXiv:2609.17474v2 Announce Type: replace-cross Abstract: Large language model (LLM) distillation can propagate teacher bias under covariate shift when target reward feedback is unavailable. We propose Iterative Calibrat…
- LessWrongresearchInterpreting attention head outputs with NLAs
Natural language autoencoders are pretty interesting; I'm a big fan of using meta-models to interpret what's going on inside LLMs and NLAs seem like a promising route to understanding them better. One thing I've been int…
- LessWrongresearchFormalising Newcomblike Problems
This is the write-up of my MATS project so far, as a mentee of Abram Demski . Those who would prefer a more formal/academic/mathematical write-up should look at the AAMAS submission here . TL:DR, a formal framework for r…
- LessWrongresearchFormally Defining Wavefunction Branches
[LW Meta: This post is (a) about how wavefunction branching explains measurements and why chaotic dynamics don't cause the quantum state to become non-classical-looking (b) the mathematical problem of formally defining w…
- LessWrongresearchSomeone Made an Infra-Bayesian RL Agent, and it One-Boxes!
Someone made an infra-bayesian RL agent , and it learns to one-box . Arxiv paper , GitHub repo . Discuss
- LessWrongresearchJ-Lens can't detect what a subliminal learning teacher is transmitting (in 7B-32B models)
I ran the trademark experimental setup from the subliminal learning paper . I reproduced the subliminal learning effect, but I also found that on a teacher model that's been prompted to imbue its answers with a love for…
- Microsoft ResearchresearchMicrosoft Research Focus – Edition #97
Host Sowbhagya Suresh shares some of the latest news and cutting-edge research from across Microsoft—from understanding the challenges facing coding agents to uncovering how people use AI at work and improving extended-r…
- LessWrongresearchThe Newly "Needed" Labor: Under AI 2040, economic growth is normal, but prices fall exponentially
The AI futures project predicts that under plan A An average of roughly 90% yearly growth in real output from 2032 to 2037. Broadly speaking, throughout this period AI and robots are capable of doing the vast majority of…
- SiliconANGLE AIresearchIntuit rewires accounting education as AI pushes new hires toward client work
As AI automates routine finance work, the repetitive tasks that once defined entry-level accounting are disappearing, and advisory services are moving to the center of how firms create value. That shift leaves colleges r…
- Latent SpaceresearchCan Frontier Models See Past a Decoy? Elorian Tests Visual Reasoning
A new decoy counting benchmark exposes a gap between seeing and reasoning. Elorian’s Glimpse tries to bridge it by thinking with images.
- AI Stack ExchangeresearchHow Can AI-Powered Indoor Drones Improve Inventory Identification Accuracy in GPS-Denied Warehouses?
Using AI drones to autonomously monitor inventory in industrial warehouses. Small indoor drones like the Drone IQ Nano provide a good example of where these systems can be employed. A govkicw fbnk arises is ensuring the…
- NVIDIA Developer BlogresearchUnlocking Tensor Core Performance for Machine Learning Interatomic Potentials
Atomistic simulations help researchers understand how molecules and materials behave, from chemical reactions and phase transitions to mechanical failure. These...
- Hugging Face BlogresearchAnnouncing Project ProtoStar: Making Publicly Funded Science AI-Ready
Open-source model releases, tooling, and community updates.
- The Register AI + MLresearchPlease state the nature of the medical emergency: Google is building a robot medic for NASA
The AI medical assistant is still being tested. Now the space boffins want to give it hands
- LessWrongresearchSome thoughts on GRAFT and follow-up work
I'm speaking as myself here and not on behalf of any of my co-authors. A bunch of the stuff in this post is explicitly speculative. I've written this post as a way of centralizing some points I want to make in discussion…
- LessWrongresearchWhat do we mean by ‘long-horizon’ work?
TL;DR - The benchmarks frontier labs cite for long-horizon capability mostly measure capability on tasks involving many steps or lots of context, verifiable end states, and little external input. I argue that what define…
- TechRadar AIresearchGartner report finds just 3% of organizations say they've hit productivity and savings goals from using AI
Mainstream AI news from a large tech publication.
- MIT Technology Review AIresearchPairing industrial AI autonomy and human control
High-signal analysis and thoughtful AI reporting.
- LessWrongresearchHumans' lot is lifelong learning
Currently, in most fields, the foremost experts are humans. They form a culture that pushes on the envelope of knowledge, uncovering facts and developing methods. Already, machine intelligence is able to uncover facts, a…
- LessWrongresearchStreetlight, Torchlight and Nightvision
Note: We at PrincInt are launching a program teaching “Discernment” or “Taste” and generally spend a lot of our time in workshops and otherwise trying to teach academics how to think about AI such that their research end…
- Ai2 BlogresearchHow researchers are adapting olmOCR 2 for historical and archaeological archives
Researchers are adapting olmOCR 2 to make historical photo libraries searchable locally and improve transcription of handwritten archaeological records.
- arXiv cs.AIresearchCounting the Bandages: On measuring the time that AI saves
arXiv:2610.35942v1 Announce Type: new Abstract: I consider what it means for AI assistance to save time. I contrast a stopwatch view, on which time saved is whatever the clock records, with a ledger view, on which it is…
- arXiv cs.AIresearchEnhancing Deep Learning Model Performance through Data Generation Strategies: Classification of Olive Tree Diseases
arXiv:2610.35984v1 Announce Type: new Abstract: Deep learning models are highly effective in a wide range of domains, particu larly in plant disease diagnosis. However, their performance is often limited by challenges su…
- arXiv cs.AIresearchWhy and How to Do Science When AI Is Better at It: A Framework Grounded in Human Understanding and Direction
arXiv:2610.36116v1 Announce Type: new Abstract: As AI begins to match or exceed human researchers at core scientific tasks, two questions become pressing: \emph{why} should humans keep doing science, and \emph{how} shoul…
- arXiv cs.AIresearchFrom Structural Feedback to Prompt Policies: Learning Faithful Text-to-Image Prompt Editors
arXiv:2610.36146v1 Announce Type: new Abstract: Prompt optimization should improve the image a user receives without changing the scene the user requested. Yet stylistic rewriting can weaken object attributes and relatio…
- arXiv cs.AIresearchDon't Pick a Winner: Knowledge Conflicts Need a Prior, Not a Rule
arXiv:2610.36169v1 Announce Type: new Abstract: A retrieval-augmented language model answers from two sources, the retrieved context and its own knowledge, and the two disagree whenever the context is incomplete or wrong…
- arXiv cs.AIresearchLearning Human Sports Motion Feasibility via Quantile Regression on Tracking Data
arXiv:2610.36183v1 Announce Type: new Abstract: We study time-to-arrival (TTA) in sports: the shortest plausible time a player needs to reach an arbitrary location on the pitch given their current kinematic state, estima…
- arXiv cs.AIresearchMARQ: A Modular, LLM-Agnostic Requirement Quality Assessor Infused with Systems Engineering Knowledge
arXiv:2610.36234v1 Announce Type: new Abstract: Requirements engineering (RE) is foundational to systems engineering (SE), as requirement defects can propagate downstream and adversely affect development outcomes. Althou…
- arXiv cs.AIresearchBehaveRec: When Does Long-Term Behavioral History Help Sequential Recommendation? A Study with Session, History, and Community Signals
arXiv:2610.36239v1 Announce Type: new Abstract: Sequential recommendation models typically predict a user's next action from recent session interactions, which limits personalization when short-term signals are weak or a…
- arXiv cs.AIresearchA Smart Instant Book Filter Model based on Dual-Objective Optimization
arXiv:2610.36287v1 Announce Type: new Abstract: Hospitality service in online marketplaces has become increasingly popular over the past few years, not only because it helps people easily rent short-term lodging but also…
- arXiv cs.AIresearchSpiking Neural Networks Are Harder to Steal: A Benchmark Study of Model Stealing Attacks
arXiv:2610.36298v1 Announce Type: new Abstract: Machine learning models deployed behind query interfaces expose more than their predictions, because an adversary can use those predictions to rebuild the model itself. Mod…
- arXiv cs.AIresearchHow Stable Are LLM-Generated Proofs Under Program Refactoring? A Stress Test for Verifiable Code Generation
arXiv:2610.36328v1 Announce Type: new Abstract: Benchmarks for verifiable code generation usually test a proof against one fixed implementation. We ask whether the same proof remains valid after a program refactoring tha…
- arXiv cs.AIresearchPlenty of Room in the Middle: Revealing Opportunities on a GPU Cluster via Fine-Grained Monitoring
arXiv:2610.36329v1 Announce Type: new Abstract: GPU clusters have become essential computing resources for AI and scientific research in academia and industry. However, their high cost and growing demand often result in…
- arXiv cs.AIresearchExecutable Archaeology of Programs That Model Programming: Reanimating the Heuristic Compiler (1961) and KBEmacs (1984)
arXiv:2610.36341v1 Announce Type: new Abstract: Early AI programs were often also theories of how people think, and programs that wrote programs were, in part, theories of how people program. Here we report on the reanim…