AI news archive.
10,660 articles filed under Research, newest first, from the outlets listed on the sources page.
- LessWrongresearchTeX was invented to typeset math but is now used for reasoning
I'll keep this short, since it's a simple observation that I haven't seen anybody else make, about the way computer systems do math. If you ask a language model to do a multi-step math problem (let's take GLM-5.3 as an e…
- IEEE Spectrum AIresearchA Day in the Life of a Roboticist: Charlie Kemp
Building useful robots starts with understanding the people who use them. For Charlie Kemp, cofounder and chief technology officer of Hello Robot, that means developing assistive robots that can help people with everyday…
- Microsoft ResearchresearchOne year in: How Microsoft Research Asia – Singapore is advancing research, partnership and talent for real-world impact
Since launching a year ago, the Microsoft Research Asia — Singapore lab has established a strong foundation, deepened collaboration across government, academia, and industry, and explored how frontier AI research can cre…
- LessWrongresearchAI: cognitive labor glut + new guys
Why is the advent of AI a big deal, and more worrying than previous advents? I think there are actually two interesting things going on, that make AI importantly different to previous technologies. I. Industrializing the…
- MIT Technology Review AIresearchWhen can we say AI made a scientific discovery?
This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here. Last Wednesday, Anthropic announced that earlier this year it had launched a molec…
- The Verge AIresearchOpenAI keeps bulldozing mathematicians
In a chaotic few months, OpenAI has demonstrated it can do two things with remarkable consistency: make impressive breakthroughs in mathematics, then colossally screw up announcing them. OpenAI is now trying to do better…
- Mistral AI NewsresearchHallo, Deutschland!
Mistral opens a Munich hub for Physics AI and Industrial AI research, partnering with German industry.
- LessWrongresearch3 Tips to Improve Activation Oracle Results
Summary: Three simple inference-time changes can significantly improve Activation Oracle (AO) performance. Provide the activation oracle with multiple tokens, not just one. To mitigate hallucinations, sample several time…
- LessWrongresearchDeep models reveal better strategies for superposition
1. Introduction The current, incredible performance of AI models is closely related to their compression capabilities ( Language Modeling Is Compression (Delétang et al., 2023) ; Compression Represents Intelligence Linea…
- LessWrongresearchRamez Naam: Can AI self-improvement overcome diminishing returns?
This is a linkpost for a new essay by futurist and sci fi author Ramez Naam: Can AI self-improvement overcome diminishing returns? AI is already helping improve itself. The question is whether even fully autonomous recur…
- Import AIresearchImport AI 474: Platonic mindspace; TPUs in space; Zhipu starts an outer RSI loop
Where do you exceed the capabilities of an LLM?
- TechRadar AIresearchLocalization is the missing test for voice AI
Mainstream AI news from a large tech publication.
- WIRED AIresearchSolving Math’s Greatest Problems Was an Art Form. Then Came AI
Mathematics has been one of humanity’s most creative endeavors, akin to painting and poetry. Now, mathematicians are trying to save it from the brute force of AI.
- WIRED AIresearchThe Next Evolution of AI Is Learning From Your Dodgy Gaming Skills
A British startup is shaping video game inputs into training data for AI models that can navigate the physical world.
- LessWrongresearchWhy do models *really* fail on HLE tasks?
I recently attended Generality Labs ' Inspect Evals Data Viz Hackathon, and spent the day using Inspect AI and its offspring, Scout, with a simple goal in mind - generate a new plot of a new or existing benchmark. Many t…
- arXiv cs.AIresearchA Synthetic Ground-Truth Framework for the Evaluation of Explainable AI Methods
arXiv:2609.30397v1 Announce Type: new Abstract: Evaluating explainable Artificial Intelligence (XAI) methods is a challenging task due to the lack of reliable evaluation procedures and, in particular, the absence of grou…
- arXiv cs.AIresearchPredicting Transmembrane Protein Topology from 3D Structure
arXiv:2609.30446v1 Announce Type: new Abstract: This paper presents a novel approach to infer protein topology using the state-of-the-art graph neural network (GNN), SchNet. The model is trained on the same dataset used…
- arXiv cs.AIresearchSpectral Feedback for Test-Time Alignment of Protein Diffusion Models
arXiv:2609.30456v1 Announce Type: new Abstract: Reward maximization alignment methods for discrete diffusion models have primarily focused on steering the reverse process, either by influencing token logits or by selecti…
- arXiv cs.AIresearchPretrained ASR Pseudo-labeling for Noisy Police Audio
arXiv:2609.30469v2 Announce Type: replace Abstract: Pretrained ASR systems perform poorly on noisy Broadcast Police Communication (BPC), hindering efforts to understand police decision-making. Pseudo-labeling offers an u…
- arXiv cs.AIresearchDo LLMs Understand Context? A Knowledge Graph-Based Evaluation Framework
arXiv:2609.30484v1 Announce Type: new Abstract: While large language models (LLMs) have achieved remarkable linguistic capabilities, a profound question lingers at their core: do these models truly comprehend context or…
- arXiv cs.AIresearchBioEVAL: A global, multi-institutional benchmark of large language and multimodal models for bioengineering
arXiv:2609.30489v1 Announce Type: new Abstract: Large Language Models (LLMs) have demonstrated historic breakthroughs in general reasoning with early successes in biomedical science. However, existing LLM benchmarking em…
- arXiv cs.AIresearchBenchy: towards a universal language for task-oriented AI benchmarks
arXiv:2609.30550v1 Announce Type: new Abstract: Benchy is a semantic language and execution engine for benchmarking AI programs. A benchmark is completely specified by a program, a scoring function, and a dataset, B=(P,S…
- arXiv cs.AIresearchRank-Reliable Teacher-Guided Fitness Approximation for Expensive Evolutionary Optimization: A TinyML Architecture Search Study
arXiv:2609.30553v1 Announce Type: new Abstract: Expensive evolutionary search does not always need an exact fitness estimate for every candidate. It often needs a reliable answer to a simpler question: which candidate is…
- arXiv cs.AIresearchAtelier: Learning Local Self-Supervised Features for CryoEM Volumes via Hypernetworks
arXiv:2609.30569v1 Announce Type: new Abstract: CryoEM map interpretation requires features that are spatially localized, consistent across samples, and informative across spatial scales. Most deep learning methods for m…
- arXiv cs.AIresearchHARDEN: Constrained Evolutionary Search for Harder, Answer-Preserving Evaluation Cases
arXiv:2609.30571v1 Announce Type: new Abstract: Language models are often evaluated on curated benchmarks that underrepresent the complexity of enterprise deployments. We introduce HARDEN, a constrained evolutionary sear…
- arXiv cs.AIresearchT-RoPE: Time-Aware Rotary Position Embedding for Sequential Recommendation
arXiv:2609.30576v1 Announce Type: new Abstract: Large-scale recommenders increasingly adopt the sequential generative recipe behind large language models, bringing the Transformer into recommendation along with design ch…
- arXiv cs.AIresearchAudio LLMs Know When They Can't Hear You
arXiv:2609.30625v2 Announce Type: replace Abstract: Audio large language models allow users to interact with the model through speech. When an input recording is too degraded, the model may misinterpret the user's query…
- arXiv cs.AIresearchThe Price of Thought: Does Test-Time Reasoning Pay in LLM Trading?
arXiv:2609.30705v1 Announce Type: new Abstract: While inference-time reasoning in large language models (LLMs) promises better decision making, its higher computational cost may not yield better economic outcomes. Yet re…
- arXiv cs.AIresearchLAVOIR: Teaching a Single-Pass Decision Encoder When and What to Ask with Amortized Value of Information
arXiv:2609.30706v1 Announce Type: new Abstract: "System One" decision models such as TypeSafe's Jev and its open counterpart Laya answer typed questions about a text in a single forward pass with calibrated probabilities…
- arXiv cs.AIresearchFrom S3Q Theory to Implementation: Towards an Architecture for Machine Qualia
arXiv:2609.30743v1 Announce Type: new Abstract: A key challenge in machine consciousness research is translating theoretical models into computational-level implementations. In this paper, we address this challenge by pr…
- arXiv cs.AIresearchORCA: Evaluating LLMs on Data Science Code Translation
arXiv:2609.30749v1 Announce Type: new Abstract: Data Science Code Translation (DSCT) is the process of converting code between data science libraries while preserving functional equivalence and enabling interoperability…
- arXiv cs.AIresearchBackbone-Adaptive Evidence Routing for Robust Pairwise LLM Judging
arXiv:2609.30751v1 Announce Type: new Abstract: Pairwise language-model judges can gather evidence through direct comparison, reasoning, or reference-based verification, but no single protocol is best across benchmarks a…
- arXiv cs.AIresearchSelective Amortization of Full-Budget Counterfactual Reasoning for Visual Token Communication
arXiv:2609.30756v1 Announce Type: new Abstract: Generative image communication transmits compact semantic tokens under a limited packet budget, where token selection directly affects the final reconstruction quality afte…
- arXiv cs.AIresearchHCOE: Hyperbolic Clinical Ontology Embeddings from Biomedical Language Models
arXiv:2609.30763v1 Announce Type: new Abstract: Biomedical language models (LMs) encode textual semantics but do not explicitly preserve medical code hierarchies. We present Hyperbolic Clinical Ontology Embeddings (HCOE)…
- arXiv cs.AIresearchInsurance Reserve Intelligence Platform
arXiv:2609.30765v1 Announce Type: new Abstract: Insurance reserve estimation is a fundamental actuarial task supporting premium pricing, solvency assessment, financial reporting, capital planning, and risk management. Cl…
- arXiv cs.AIresearchDoes Thinking Help Fairness? Reasoning Tokens Resolve Some Biases but Create More
arXiv:2609.30768v1 Announce Type: new Abstract: Thinking in reasoning language models (RLMs) has been subject to debate on whether it resolves or amplifies bias. Prior works have shown competing conclusions in both direc…
- arXiv cs.AIresearchConsultMind:Towards Automated Diagnostic Consultation via Uncertainty-Aware Reasoning
arXiv:2609.30796v1 Announce Type: new Abstract: Diagnostic consultation is an online sequential decision-making process in which clinicians gather evidence through patient interaction until a diagnosis is sufficiently su…
- arXiv cs.AIresearchPTC-Decoder: Towards Intelligent SLMs on Offline Resource-Constrained Edge Devices
arXiv:2609.30836v1 Announce Type: new Abstract: Deploying small language models (SLMs) on offline, resource-constrained edge devices such as remote sensing satellites presents a fundamental challenge: their limited reaso…
- arXiv cs.AIresearchWhy Jailbreaks Succeed in Diffusion Language Models: An Energy Landscape Analysis
arXiv:2609.30841v1 Announce Type: new Abstract: Existing attacks and defenses for diffusion-based large language models (dLLMs) target specific vulnerabilities but lack a shared framework explaining why attacks succeed.…
- arXiv cs.AIresearchTISD: On-Policy Self-Distillation with Trajectory Intervention
arXiv:2609.30878v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) provides dense teacher targets, but evaluates them only along student-sampled rollouts. When the privileged teacher favors an alternative…
- arXiv cs.AIresearchEXAONE Demand 1.0: A Time Series Foundation Model for Demand Forecasting
arXiv:2609.30880v1 Announce Type: new Abstract: Time series foundation models (TSFMs) are pretrained on series from diverse domains, where demand series make up only a small fraction. Demand data has properties that such…
- arXiv cs.AIresearchTraining Graph Foundation Models on The Web Graph
arXiv:2609.30894v1 Announce Type: new Abstract: We introduce Acacia, a graph foundation model, trained on the web graph. Acacia (i) supports arbitrary feature dimensionalities and semantics without additional training, (…
- arXiv cs.AIresearchJevSoup: System-One Routing for Training-Free LoRA Composition
arXiv:2609.30922v1 Announce Type: new Abstract: Building adaptable AI systems requires effective coordination of specialized capabilities across diverse tasks. Low-rank adaptation (LoRA) enables modular expertise, but ex…
- arXiv cs.AIresearchSelf-Play Search Distillation for Large Language Model Reasoning
arXiv:2609.30936v1 Announce Type: new Abstract: Improving reasoning abilities in Large Language Models (LLMs) requires high-quality data that exposes difficult decisions, competing alternatives, and their consequences. D…
- arXiv cs.AIresearchFTB Graph: Determining and Validating First-token Broadcasters and Language-Identity Head Circuits in Multilingual Language Models
arXiv:2609.30954v1 Announce Type: new Abstract: Large language models operating in multilingual contexts must resolve target response languages early in generation, yet the causal circuitry governing first-token language…
- arXiv cs.AIresearchMoMHa: Multi-Objective Optimization of LLM Harnesses over Accuracy, Safety, and Tokens
arXiv:2609.30967v1 Announce Type: new Abstract: Most work on improving large language models treats accuracy as the sole objective. We argue that the harness, the Python code surrounding the model that constructs prompts…
- arXiv cs.AIresearchFactorized axis convolutional gated recurrent unit with dynamic adaptive pooling for remaining useful life prediction of rolling bearings
arXiv:2609.30972v1 Announce Type: new Abstract: Convolutional neural networks (CNN) are widely used to predict the remaining useful life (RUL) of rolling bearings from time-frequency representations (TFRs) of vibration s…
- arXiv cs.AIresearchSame Text, Different Numbers: The Divergence of LLM-Based Measures
arXiv:2609.31013v1 Announce Type: new Abstract: Researchers increasingly use generative large language models (LLMs) to convert corporate text into empirical variables. We examine the extent to which LLM-based textual me…
- arXiv cs.AIresearchGoverned Deduction: Policy-Grounded Premise Authorization Beyond Relevance
arXiv:2609.31029v1 Announce Type: new Abstract: Reasoning systems usually treat premise use as a question of relevance: if a fact is available and useful, it may be selected for inference. Authorization imposes a differe…
- arXiv cs.AIresearchNeuralyzing the Trace: Selective Representation-Level Unlearning with Contrastive Sparse Autoencoders
arXiv:2609.31056v1 Announce Type: new Abstract: Machine unlearning aims to remove targeted information while preserving a model's other abilities. In realistic settings, such as privacy requests under the EU GDPR, the ta…