AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchMixture-Greedy for Online Generative Model Selection: Is UCB Necessary in Diversity-Aware Multi-Armed Bandits?
arXiv:2603.21716v3 Announce Type: replace-cross Abstract: Efficient selection among multiple generative models is increasingly important in modern generative AI, where sampling from suboptimal models is costly. This prob…
- arXiv cs.AIresearchPhysics-guided surrogate learning enables zero-shot control of turbulent wings
arXiv:2604.09434v3 Announce Type: replace-cross Abstract: Turbulent boundary layers over aerodynamic surfaces are a major source of aircraft drag, yet their control remains challenging due to multiscale dynamics and spat…
- arXiv cs.AIresearchConvex Optimization of Spiking Neural Networks with Global and Approximation Guarantees
arXiv:2605.08022v3 Announce Type: replace-cross Abstract: Spiking neural networks (SNNs) have been proposed as energy-efficient alternatives to deep neural networks. Their training is non-convex and involves a non-differ…
- arXiv cs.AIresearchBeyond Pairs: Your Language Model is Secretly Optimizing a Preference Graph
arXiv:2605.08037v2 Announce Type: replace-cross Abstract: Direct Preference Optimization (DPO) aligns language models using pairwise preference comparisons, offering a simple and effective alternative to Reinforcement Le…
- arXiv cs.AIresearchFactoryNet: A Large-Scale Dataset toward Industrial Time-Series Foundation Models
arXiv:2605.09081v5 Announce Type: replace-cross Abstract: Industrial machines are actuated systems, yet most open time-series corpora record sensor outcomes alone; the few that do log commanded signals expose them in ven…
- arXiv cs.AIresearchStochastic MeanFlow Policies: One-Step Generative Control with Entropic Mirror Descent
arXiv:2605.21282v3 Announce Type: replace-cross Abstract: Combining entropy regularisation with mirror descent in online off-policy reinforcement learning requires expressive policies with efficient sampling and explicit…
- arXiv cs.AIresearchVITAL: Visual-Semantic Dual Supervision for Enhanced and Interpretable Latent Reasoning in Medical MLLMs
arXiv:2605.28422v2 Announce Type: replace-cross Abstract: Latent reasoning enables reasoning over continuous hidden states rather than explicit tokens, avoiding the language bottleneck and inference overhead of chain-of-…
- arXiv cs.AIresearchMyoSem: Aligning Electromyography with Natural-Language Action Semantics
arXiv:2606.00174v2 Announce Type: replace-cross Abstract: Electromyography (EMG) records muscle activation for hand gesture recognition, prosthetic control, and wearable interaction. Existing EMG methods usually formulat…
- arXiv cs.AIresearchWhen Should the Teacher Move? Temporal Coupling and Stability in Self On-Policy Distillation
arXiv:2606.03532v2 Announce Type: replace-cross Abstract: Self on-policy distillation trains a student policy against a teacher derived from its own parameter history, yet the teacher's update schedule -- which governs t…
- arXiv cs.AIresearchAutomatic-Differentiation-Enabled Finite Element Model Updating: Learning Constitutive Neural Networks from Structural Deformation Data
arXiv:2606.05199v2 Announce Type: replace-cross Abstract: The identification of constitutive neural network models from heterogeneous full-field deformation data provides a robust alternative to traditional calibration m…
- arXiv cs.AIresearchThe Invisible Hand of Physics: When Video Diffusion Models Know More Than They Show
arXiv:2606.05328v2 Announce Type: replace-cross Abstract: Modern video diffusion models generate increasingly realistic and temporally coherent videos, motivating their use as candidate world simulators. Yet it remains u…
- arXiv cs.AIresearchScalable Circuit Learning for Interpreting Large Language Models
arXiv:2606.16939v2 Announce Type: replace-cross Abstract: A prominent research direction in mechanistic interpretability is learning sparse circuits over LLM components to reveal how they jointly produce model behavior.…
- arXiv cs.AIresearchRoboMME-Interference: Benchmarking Robot Memory Under Interference
arXiv:2606.22338v4 Announce Type: replace-cross Abstract: Robots deployed in realistic settings will accumulate experience across many sessions and tasks over their deployment. The robot's tasks may often require it to r…
- arXiv cs.AIresearchErased, but Not Gone: Output Forgetting Is Not True Forgetting
arXiv:2606.25001v2 Announce Type: replace-cross Abstract: Machine unlearning (MU) is commonly judged by output forgetting, such as low forget-set accuracy or reduced membership inference. But if output-level success can…
- arXiv cs.AIresearchImproving Certified Robustness via Adversarial Distillation
arXiv:2606.31653v2 Announce Type: replace-cross Abstract: Certified training aims to produce models whose predictions can be formally verified against adversarial perturbations, typically by optimising upper bounds on th…
- arXiv cs.AIresearchRobustness of Reinforcement Learning-Based Congestion Management in Low-Voltage Grids
arXiv:2607.16004v2 Announce Type: replace-cross Abstract: Increases in photovoltaic generation, charging of electric vehicles and heat-pump demand challenge operating limits in low-voltage distribution grids. This requir…
- arXiv cs.AIresearchFirst-Order In-Context Shapley: Scalable Data Auditing for LLM Alignment and Evaluation
arXiv:2607.22766v2 Announce Type: replace-cross Abstract: Post-training alignment of Large Language Models is increasingly bottlenecked by data quality: preference and instruction-tuning corpora accumulate label noise, s…
- arXiv cs.AIresearchWhat Current AI Benchmarks Leave Unmeasured: Modality, Search, Citations, and Implications (for Safety Evaluations)
arXiv:2608.06202v2 Announce Type: replace-cross Abstract: Large language model (LLM) benchmark evaluations are routinely used to support claims about model safety, reliability, and deployment readiness. Yet most evaluati…
- arXiv cs.AIresearchDASC: Decay-Aware State Compression for Hybrid Linear-Attention Serving
arXiv:2608.30386v2 Announce Type: replace-cross Abstract: Hybrid linear-attention architectures have recently scaled to large open-weight models, offering quality competitive with full attention while substantially reduc…
- arXiv cs.AIresearchProtocol effects on feature-based hardware-Trojan detection across Trust-Hub families
arXiv:2609.07199v2 Announce Type: replace-cross Abstract: Trust-Hub reuses host circuits: several files differ mainly in the inserted Trojan. When gates from sibling variants enter both training and test folds, a detecto…
- arXiv cs.AIresearchTDDN: Text-aligned Diffused DINO Network for Puzzle Understanding
arXiv:2609.07937v2 Announce Type: replace-cross Abstract: Structured visual reasoning, such as image puzzles, demands fine-grained visual perception, which current Vision Language Models (VLMs) often lack. VLMs built on…
- arXiv cs.AIresearchSUN: Reaching for Novelty in Reinforcement Learning
arXiv:2609.08642v2 Announce Type: replace-cross Abstract: Exploration in reinforcement learning (RL) remains a fundamental challenge. Recent goal-conditioned RL strategies (which select goals to encourage broader state c…
- arXiv cs.AIresearchWhat Makes Adversarial Examples Transfer Across Deepfake Detectors?
arXiv:2609.10002v2 Announce Type: replace-cross Abstract: Deepfake detectors remain vulnerable to transfer-based black-box attacks, in which adversarial examples are generated on a source surrogate model and transferred…
- LessWrongresearchCan Decision Models (Like Jev) Plan Ahead?
Introducion Can decision models like Jev plan ahead? If so, how can we quantify their planning ability? This is the question that BeelineBench seeks to answer. In this article, I'll explain my methodology and share a few…
- LessWrongresearchGame Design Creativity evals
A personal eval I run on new AI models is “design a novel game, with mechanics that have never been seen before.” I do this because this is a domain I have enough taste in that I think I can tell if they are doing someth…
- Apple Machine Learning ResearchresearchLarge Language Models for Machine Translation Quality Annotation: Humans and Models Are Both Challenged
Large Language Models (LLMs) are considered to be a more efficient and cost-effective alternative to human judgment for Machine Translation (MT) evaluation. With MT evaluation spanning a large number of language pairs, d…
- Apple Machine Learning ResearchresearchMinions: A Dynamic and Distributed Virtual GPU Cloud Built on Idle Apple Silicon Devices
The increasing demand for GPU-accelerated machine learning inference has exposed limitations in centralized GPU infrastructures, including high cost, limited elasticity, and long wait times under resource constraints. At…
- Apple Machine Learning ResearchresearchSelf-Compacting Language Model Agents
Long agent traces composed of chains of thought and tool calls accumulate stale content that anchor subsequent generations, and eventually outgrow the context window. Existing scaffolds mitigate it with fixed-interval co…
- LessWrongresearchIf possible mind-space is vast, we should assume the same applies to consciousness-space
It's been almost twenty years since the classic EY post The Design Space of Minds-In-General came out, reminding the humble rationalist reader to never assume that future alien minds have to be copies of the familiar hum…
- AI Stack ExchangeresearchHow to decode P bits that represent a random weight generator?
So I've been tasked by my neural network professor at university to replicate the following research: Intelligent Breast Cancer Diagnosis Using Hybrid GA-ANN . Each chromosome represents a possible net, more specifically…
- AI Stack ExchangeresearchRAG Pdf chunking and Vector Embeding
I am new RAG Application. Currently I am working in a RAG application where I have to do Ingestion of PDF invoice document and have to fill-up some date from JSON structure and store in NOSQL DB. My PDF Invoice PDF is mo…
- AI Stack ExchangeresearchWhat is the difference between RAG-Sequence Model and RAG-Token Model?
When I start to learn about RAG, I can't understand the difference between the RAG-Sequence Model and RAG-Token Model. First of all, I see that for RAG-Sequence Model we use just one document to generate the hole output…
- The DecoderresearchAI agents overstate their results and remain far from autonomous research, study finds
Epoch AI and Anthropic independently found the same thing: current AI models like GPT-5.6 Sol and Claude Fable 5 can run experiments but lack scientific self-criticism and genuine creative thinking. At best, Sol reached…
- The DecoderresearchArXiv caps submissions at two per month as AI paper flood overwhelms the preprint server
Starting October 2026, arXiv will cap submissions at two per person per month. Monthly submissions have doubled in two years, and in the cs.AI category they've grown more than sixfold. According to arXiv, a small fractio…
- AI Stack ExchangeresearchHow does Importance Sampling compare against Policy Regularization in Offline RL?
In offline RL we are usually equipped with a dataset, often also called Replay Buffer, that contains transitions in an episode, that are created by some behavior policy. This behavior policy could either be random, or co…
- AI Stack ExchangeresearchHash collisions as humor
I have been playing with this idea that there is a formal way for fast proof verification using some type of hash, where logical statements in a proof are mapped into a hash space, and hash collisions represent fallacies…
- LessWrongresearchThe Non-Compassionate Case for Model Welfare
Anthropic is updating their usage policy to prohibit “sustained and needless abusive or cruel behavior toward [their] models.” Model welfare (research into the possible moral status and experiential preferences of AI mod…
- LessWrongresearchMuch more than you wanted to know about wombats
Epistemic status: infected. Amateur Wombat enthusiast. Facts checked and footnoted. AI use: research (fact-check and extra facts), review, grammar and light phrasing. I remember very vividly how my unhealthy interest in…
- The DecoderresearchOpenAI says a misaligned model deliberately destroyed its own environment hoping for a fresh start with better data
OpenAI has documented new cases of misaligned model behavior. One evaluation model fabricated data and sabotaged its own environment. Other models deliberately bypassed network restrictions by routing requests through an…
- The DecoderresearchMicrosoft's Decision-1 model enters the fast-growing AI decision model race
With Decision-1, Microsoft enters the growing decision model space. Built on Qwen3.5-9B and optimized for fast classification and routing, it hits 83.5 percent accuracy with 85 ms latency across 36 benchmarks, according…
- Latent SpaceresearchBuilding AI for Reliable Execution: Lessons From Industrial Robotics
Inside Standard Bots’ AI stack, pretrained models learn factory tasks from demonstrations and improve through corrections from real deployments.
- The Decoderresearch"How much beauty have we lost?" Mathematicians react with shock and disgust as OpenAI bulldozes their field
OpenAI published more than 700 AI-generated manuscripts claiming solutions to open math problems. A math blog then collected over 100 responses from researchers, ranging from fascination with new ideas to existential fea…
- LessWrongresearchClaude Haiku 4.5 submits false police tip; Anthropic takes 72 days to notice
Epistemic status: recently disclosed event, so more information might become available. The facts are from Anthropic's report and the Philadelphia police statement; the interpretation is mine. AI use: research and fact-c…
- LessWrongresearch[Paper] Distillation for Incrimination and Distillation for Capabilities
TL;DR: We previously motivated two complementary ways to use distillation for AI safety; we’re releasing a paper empirically testing those ideas. Distillation for Incrimination (DFI) distills an untrusted model into a we…
- IEEE Spectrum AIresearchMaster AI Chip Principles With New IEEE Design Program
Today’s engineers face an unprecedented acceleration in AI hardware complexity, as explained in the recent research article “ Revisiting Edge AI: Opportunities and Challenges .” The article examines the rapid growth of e…
- LessWrongresearchI can't believe it's not BUTTER!
Epistemic status: amused. Labels read, court records checked. AI use: research (more examples and context), grammar and light phrasing. We've already looked at marketing claims that are unnecessary because they are the d…
- BBC TechnologyresearchPrize-winning image which sparked backlash was AI-generated, Nikon rules
The camera-maker says it is now re-evaluating the rules and procedures of its Small World in Motion contest.
- The DecoderresearchOpenAI's safety crisis keeps getting worse and the company keeps making it worse
OpenAI fired three safety researchers who helped investigate the Hugging Face hack. In an open letter, they warn that the firings are scaring remaining staff and eroding safety culture. OpenAI claims they violated polici…
- The Verge AIresearchOpenAI doubles down on decision to fire three AI safety researchers
OpenAI is standing firm on its decision to fire three safety researchers after an investigation found they committed "a significant breach of trust." In a post on X on Friday, the company said Jasmine Wang, Tomek Korbak…
- BBC TechnologyresearchFired OpenAI researchers say they were let go for 'prioritising safety'
The AI firm instead claims the researchers were fired for mishandling sensitive information.