AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchA Computationally Efficient Multi-Section Whole Slide Image Analysis for Capturing Prostate Cancer Multifocality in Biochemical Recurrence Prediction
arXiv:2603.20273v2 Announce Type: replace-cross Abstract: Prostate cancer is among the most frequently diagnosed malignancies in men worldwide. However, precise prediction of biochemical recurrence (BCR) after radical pr…
- arXiv cs.AIresearchCtrl-A: Control-Driven Machine Learning using Adaptive Online Data Augmentation
arXiv:2603.21819v2 Announce Type: replace-cross Abstract: We introduce ControlAugment (Ctrl-A), an automated data augmentation algorithm for image-vision tasks, which incorporates principles from control theory for onlin…
- arXiv cs.AIresearchSemLoc: Structured Grounding of Free-Form LLM Reasoning for Fault Localization
arXiv:2603.29109v2 Announce Type: replace-cross Abstract: Large language models (LLMs) can describe what a piece of code should compute, but they do so in free-form text. Such text cannot be checked against program execu…
- arXiv cs.AIresearchScoring-Rule Sensitivity in LLM Evaluation: Evidence from Indian Financial Regulatory Text
arXiv:2604.19298v3 Announce Type: replace-cross Abstract: A benchmark leaderboard reports one accuracy number, but that number is a joint product of a model and a scoring rule. We ask what changes when only the rule chan…
- arXiv cs.AIresearchAmaraSpatial-10K: A Spatially and Semantically Aligned 3D Dataset for Spatial Computing and Embodied AI
arXiv:2604.23018v3 Announce Type: replace-cross Abstract: Web-scale 3D asset collections are abundant but rarely ready to ingest without cleanup, suffering from arbitrary metric scaling, incorrect pivots, brittle geometr…
- arXiv cs.AIresearchDeGenTWeb: A First Look at MGT-dominant Websites
arXiv:2605.00087v2 Announce Type: replace-cross Abstract: Many news reports have claimed that machine-generated text (MGT) from large language models (LLMs) is taking over the web. However, these claims are typically not…
- arXiv cs.AIresearchOn the Optimal Sample Complexity of Offline Multi-Armed Bandits with KL Regularization
arXiv:2605.02141v2 Announce Type: replace-cross Abstract: Kullback-Leibler (KL) regularization is widely used in offline decision-making and offers several benefits, motivating recent work on the sample complexity of off…
- arXiv cs.AIresearchPosterior-First Neural PDE Simulation: Inferring What Downstream Tasks Need from a Single Field
arXiv:2605.03247v2 Announce Type: replace-cross Abstract: Neural PDE simulators often receive only a single observed field at deployment. In this setting, a field-to-future predictor can collapse distinct latent problem…
- arXiv cs.AIresearchDr. Post-Training: A Data Regularization Perspective on LLM Post-Training
arXiv:2605.07063v2 Announce Type: replace-cross Abstract: Data selection methods address a critical challenge in LLM post-training: effectively leveraging scarce, high-fidelity target data alongside abundant but imperfec…
- arXiv cs.AIresearchOn the Tradeoffs of On-Device Generative Models in Federated Predictive Maintenance Systems
arXiv:2605.07860v2 Announce Type: replace-cross Abstract: Federated Learning (FL) has emerged as a promising paradigm for preserving client data ownership and control over distributed Internet of Things (IoT) environment…
- arXiv cs.AIresearchRePO-VLA: Recovery-Driven Policy Optimization for Vision-Language-Action Models
arXiv:2605.09410v2 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) models remain brittle in long-horizon, contact-rich manipulation because success-only imitation provides little supervision for execu…
- arXiv cs.AIresearchThe Geometry of Learning: Read-Write Asymmetry Across Training
arXiv:2605.16600v2 Announce Type: replace-cross Abstract: Pretraining and alignment update the same transformer weights, and we ask in which directions each stage changes them. The residual stream gives every matrix a sh…
- arXiv cs.AIresearchBoLT: A Benchmark to Democratize Black-box Optimization Research for Expensive LLM Tasks
arXiv:2605.17000v2 Announce Type: replace-cross Abstract: Optimization of LLM training and inference configurations, such as hyperparameters, data mixtures, and prompts, is critical to performance, but it is often approa…
- arXiv cs.AIresearchLLM Code Smells: A Taxonomy and Detection Approach
arXiv:2605.22976v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are increasingly integrated into software systems for diverse purposes, due to their versatility, flexibility, and ability to simulat…
- arXiv cs.AIresearchHidden-State Privacy Has an Empty Middle
arXiv:2605.24042v4 Announce Type: replace-cross Abstract: Of $1{,}536$ Gaussian release covariances we tested for single-layer hidden-state privacy, zero achieve both moderate utility and moderate privacy against an adap…
- arXiv cs.AIresearchCalArena: A Large-Scale Post-Hoc Calibration Benchmark
arXiv:2605.30188v3 Announce Type: replace-cross Abstract: Reliable probability estimates are critical in many machine learning applications, yet modern classifiers are often poorly calibrated. Post-hoc calibration provid…
- arXiv cs.AIresearchGrasp-Then-Plan with Failure Attribution: A Closed Two-Stage Framework for Precise and Generalizable Robotic Manipulation
arXiv:2606.03385v2 Announce Type: replace-cross Abstract: In robotic manipulation, the tight coupling between grasping and motion planning often obscures the true source of failure, leading to inefficient trial-and-error…
- arXiv cs.AIresearchDRIFT: A Residual Flow Adapter for Decoding Continuous Outputs in Vision-Language Models
arXiv:2606.05758v2 Announce Type: replace-cross Abstract: Many modern vision-language models (VLMs) build on autoregressive decoding of discrete tokens. While text-based output interfaces enable scalable pretraining and…
- arXiv cs.AIresearchLLMs Can Leak Training Data But Do They Want To? Evaluating Memorization Propensity in LLMs
arXiv:2606.06286v2 Announce Type: replace-cross Abstract: Large language models can reproduce training data, but memorization evaluations mostly measure whether models can be made to do so, not whether they do so without…
- arXiv cs.AIresearchWhat Makes Video World Model Latents Action-Relevant: Prediction over Reconstruction
arXiv:2606.07687v2 Announce Type: replace-cross Abstract: Video world models are increasingly used to provide predictive visual representations, yet it remains unclear which pretraining signals induce action-relevant str…
- arXiv cs.AIresearchSpatio-Temporal Audio Language Modeling for Dynamic Sound Sources
arXiv:2606.14141v2 Announce Type: replace-cross Abstract: Sound events have semantic identities, locations, and trajectories, yet existing audio-language models typically encode clips as global event content. Conversely,…
- arXiv cs.AIresearchDifFRACT: Diffusion Feature Reconstruction and Attribution for Circuit Tracing
arXiv:2606.15796v2 Announce Type: replace-cross Abstract: While transcoder-based circuit tracing has recently enabled detailed causal analyses of large language models, it remains unclear how semantic information propaga…
- arXiv cs.AIresearchLineageMark: Multi-user White-box Watermarking for Contribution Tracing in Model Derivation Chains
arXiv:2606.17123v2 Announce Type: replace-cross Abstract: In open large language model (LLM) ecosystems, models are frequently adapted across multiple domains and applications, forming multi-stage derivation chains. Cons…
- arXiv cs.AIresearchTokenMinds: Pretrained User Tokens and Embeddings for User Understanding in Large Recommender Systems
arXiv:2606.25147v2 Announce Type: replace-cross Abstract: User modeling in industrial recommender systems typically produces dense embeddings, which suffer from representational constraints due to fixed-dimensional vecto…
- arXiv cs.AIresearchTrust Region Policy Distillation
arXiv:2607.04751v2 Announce Type: replace-cross Abstract: Big goals are hard to achieve all at once; breaking them into small steps is wiser. We present Trust Region Policy Distillation (TOP-D), which transforms the noto…
- arXiv cs.AIresearchFrom Language to Navigation Goals: A Vision-Language Approach for Semantic Navigation of Mobile Robots Using RGB-D Perception
arXiv:2607.13624v2 Announce Type: replace-cross Abstract: Natural language interaction provides an intuitive way for non-expert users to communicate with robotic platforms. However, transforming user requests into execut…
- arXiv cs.AIresearchFrom Deterministic to Generative Deep Learning for Urban Air Quality Reconstruction from Sparse Observations
arXiv:2607.25687v2 Announce Type: replace-cross Abstract: Full-field reconstruction of air pollution is essential for evaluating pollution exposure and supporting public health decision-making. However, the complex inter…
- arXiv cs.AIresearchTextCloak: Safeguarding Data from Unauthorized LLM Fine-Tuning via RL-Driven Unlearnable Examples
arXiv:2607.28862v2 Announce Type: replace-cross Abstract: The development of Large Language Models (LLMs) has led to significant advances across a wide range of language tasks, while simultaneously raising growing concer…
- arXiv cs.AIresearchActionable CBFI: Prescriptive Interventional Diagnosis and Interaction-Aware Escalation for Sparse Causal Algorithmic Recourse
arXiv:2608.27821v3 Announce Type: replace-cross Abstract: Algorithmic recourse tells an individual which feasible actions would change an unfavourable prediction. Feature attribution explains a prediction, but the most i…
- arXiv cs.AIresearchFrom Perspective to Fisheye Depth Estimation and Open-Vocabulary Segmentation
arXiv:2608.27860v2 Announce Type: replace-cross Abstract: Vision foundation models are capable of generalizing across 3-dimensional (3D) scenes with high-fidelity estimates; their empirical success can be attributed to t…
- arXiv cs.AIresearchCan LLMs Design Video Coding Tools? A Case Study on Planar Mode
arXiv:2609.01535v2 Announce Type: replace-cross Abstract: This paper explores whether large language models (LLMs) can design video coding tools, a highly challenging task due to the intricate algorithmic coupling of too…
- arXiv cs.AIresearchMUMINS: Metadata-conditioned Uncertainty-aware Medical Image Next-state Synthesis
arXiv:2609.17169v2 Announce Type: replace-cross Abstract: Forecasting anatomical changes such as tumor growth and neurodegeneration is a challenging generative vision task. Morphological evolution is subtle relative to s…
- LessWrongresearchQuantum Mechanics Overview
[LW meta: This is a very brief overview of what quantum mechanics says and how it relates to ordinary reality. Many readers here will probably already be familiar with this material, especially if they've read Eliezer's…
- AI Stack ExchangeresearchHow does SGD escape local minima?
SGD is able to jump out of local minima that would otherwise trap BGD I don't really understand the above statement. Could someone please provide a mathematical explanation for why SGD (Stochastic Gradient Descent) is ab…
- AI Stack ExchangeresearchDoes AI change how we think when we learn?
Does AI change how we think when we learn? We are conducting an independent survey to explore generative AI, independent thinking, and learning. Whether you use AI tools regularly or deliberately choose not to use genera…
- The Verge AIresearchResearchers are getting tired of OpenAI gatekeeping its breakthroughs
On Monday, OpenAI announced impressive breakthroughs in mathematics, computer science, and physics, along with what the company says is an evolving approach to communicating its findings. But despite its apparent course…
- WIRED AIresearchAnthropic Says It Will Revolutionize Biology—Some Day
In breathless press releases and posts, the company has touted its models’ ability to deliver scientific breakthroughs. But its IPO filing taps the brakes on just how fast its AI will lead to big discoveries.
- Ars Technica AIresearchVibe genes? Natural(ish) language comes to protein, genome design software
You can apparently now tell AI tools what sort of protein you want them to make.
- The DecoderresearchGoogle Deepmind's Genie 4 puts a modern spin on agent designs that predate LLMs
Google Deepmind has introduced Genie 4, a real-time world model that simulates interactive 3D worlds with sound and talking characters. Gemini models help plan in the background to keep the simulation consistent. Genie 4…
- TechCrunch AIresearchFrontier AI labs say their models refuted the Exponential Time Hypothesis. What does that mean?
TechCrunch spoke to the MIT computer scientist charged with explaining OpenAI's latest mind-bending discovery.
- Hugging Face BlogresearchCan we trust LLMs to judge whether a research idea is novel?
Open-source model releases, tooling, and community updates.
- NVIDIA BlogresearchHow AI Is Changing the Pace of Discovery for Millions Living With Dementia
There are 57 million people worldwide living with dementia today. For the researchers working to help them, the obstacle isn’t a lack of effort — it’s too much data, too few experts to synthesize it and a pace of discove…
- AI Stack ExchangeresearchObtaining the diagonal of the Fisher information matrix
I am looking to use elastic weight consolidation to try and prevent, or at least mitigate, forgetting across learning tasks. In order to do that I need to calculate the diagonal of the fisher matrix. Explicitly this is g…
- The DecoderresearchFormer Deepmind researcher thinks AlphaGo's playbook can cure AI hallucinations
Former Deepmind researcher Thore Graepel wants his startup Metis Reasoning to bring AlphaGo's search principle to general reasoning. Instead of locking knowledge inside model weights, an explicit knowledge ledger would m…
- LessWrongresearchSolving for the Necessity of Phenomenal Consciousness
For better or worse, biological consciousness continues to inspire our thinking about machine consciousness. It is our only available source of testable knowledge about the phenomenon. Only some, or none, or all of these…
- IEEE Spectrum AIresearchI Used AI to Fake a Scientific Paper. Here’s What Happened
Science has a spam problem. And the spam filter is getting clogged up. According to all reports, academic publishing is drowning in what has come to be called AI slop: low-quality, mass-produced content generated with AI…
- The DecoderresearchOpenAI asks outside experts, including a Nobel laureate, to check proofs its AI produced in math, computer science, and physics
OpenAI says its internal AI models have cracked three major open questions in math, computer science, and physics. One of them is part of a famous Millennium Prize Problem, each of which carries a $1 million prize. Outsi…
- Ai2 BlogresearchCan we trust LLMs to judge whether a research idea is novel?
Old Ideas, Novel Problems tests how reliably LLMs judge the novelty of research ideas, finding that small changes to the evaluation setup can produce very different verdicts.
- arXiv cs.AIresearchInverse Modeling for Laser Pulse Shape Design in Inertial Confinement Fusion
arXiv:2610.33096v1 Announce Type: new Abstract: The achievement of practical fusion energy remains one of the most pressing unsolved scientific challenges, with immense implications for carbon-free power. A critical fact…
- arXiv cs.AIresearchIRON-Meme: A Benchmark for Structured Understanding of Identities and Roles in Online Meme Narratives
arXiv:2610.33097v1 Announce Type: new Abstract: As polarization grows on social media, understanding how political memes shape online narratives has become increasingly important. Previous research has relied on either q…