AI news archive.
10,661 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchEdge-AI-Driven Learning-to-Rank for Decentralized Task Allocation in Circular Smart Manufacturing
arXiv:2605.16433v4 Announce Type: replace-cross Abstract: Task allocation in smart manufacturing systems must operate under decentralized decision-making, dynamic workloads, and shared-resource constraints. In circular m…
- arXiv cs.AIresearchObservation-Aligned Mask Priors for Learning Physical Fields from Authentic Occlusions
arXiv:2605.16818v2 Announce Type: replace-cross Abstract: Learning physical fields directly from incomplete observations is challenging because authentic occlusions are structured, sample-dependent, and often missing not…
- arXiv cs.AIresearchAccelerating Video Inverse Problem Solvers with Autoregressive Diffusion Models
arXiv:2605.20624v2 Announce Type: replace-cross Abstract: Diffusion models provide powerful priors for zero-shot video inverse problems, but their real-time deployment is hindered by two inefficiencies: high initial late…
- arXiv cs.AIresearchEvoScene-VLA: Evolving Scene Beliefs Inside the Action Decoder for Chunked Robot Control
arXiv:2605.21862v3 Announce Type: replace-cross Abstract: Chunked vision-language-action (VLA) policies generate multi-step actions from one observation and typically re-observe after executing the action chunk. Earlier…
- arXiv cs.AIresearchDecomposing and Measuring Evaluation Awareness
arXiv:2605.23055v3 Announce Type: replace-cross Abstract: Frontier language models sometimes recognize that they are under evaluation and adjust their behavior which can undermine validity of benchmark results. Yet the f…
- arXiv cs.AIresearchScalable Heterogeneous Graph Foundation Models for Data-Driven Optimal Power Flow in Smart Grids
arXiv:2605.23194v2 Announce Type: replace-cross Abstract: Fast and reliable optimal power flow (OPF) approximation is important for power system operation, yet heterogeneous OPF graph models are often evaluated either wi…
- arXiv cs.AIresearchAdaptive Mass-Segmented KV Compression for Long-Context Reasoning
arXiv:2605.23200v2 Announce Type: replace-cross Abstract: The linear growth of the Key-Value (KV) cache is a critical bottleneck in long-form LLM inference. Existing KV compression methods mitigate this by evicting token…
- arXiv cs.AIresearchAdvantageFlow: Regularized Advantage-Weighted RL in Flow Models
arXiv:2605.26013v2 Announce Type: replace-cross Abstract: We present AdvantageFlow, a forward-process reinforcement learning (RL) algorithm for rectified flow models. The algorithm minimizes an advantage-weighted predict…
- arXiv cs.AIresearchEfficient Pre-Training of LLMs through Truncated SVD Representations
arXiv:2605.28573v2 Announce Type: replace-cross Abstract: LLM pretraining is extremely costly; therefore, parameter-efficient LLM architectures have recently emerged as a compelling research direction. One such promising…
- arXiv cs.AIresearchReasoning with Sampling: Cutting at Decision Points
arXiv:2605.30327v2 Announce Type: replace-cross Abstract: Frontier reasoning models are produced by post-training base language models with reinforcement learning. Recent work has challenged this by showing that sampling…
- arXiv cs.AIresearchTwo-Fidelity Best-Action Identification for Stochastic Minimax Tree
arXiv:2606.01708v2 Announce Type: replace-cross Abstract: We study fixed-confidence best-action identification (BAI) in stochastic minimax trees. This problem is increasingly relevant in modern AI planning, where deep mi…
- arXiv cs.AIresearchRobotValues: Evaluating Household Robots When Human Values Conflict
arXiv:2606.03312v2 Announce Type: replace-cross Abstract: While household robots are often evaluated based on task completion, everyday domestic environments involve value-conflicting situations where robots are expected…
- arXiv cs.AIresearchReproducing, Analyzing, and Detecting Reward Hacking in Rubric-Based Reinforcement Learning
arXiv:2606.04923v3 Announce Type: replace-cross Abstract: Rubric-based reinforcement learning (RL) uses an LLM-as-a-Judge (LaaJ) to score model outputs according to rubrics as rewards. However, policy models may exploit…
- arXiv cs.AIresearchFisher-Guided Progressive Parameter Selection for Adaptive Fine-Tuning
arXiv:2606.10196v2 Announce Type: replace-cross Abstract: Parameter-efficient fine-tuning often selects trainable parameters before adaptation using architectural heuristics, without accounting for their varying importan…
- arXiv cs.AIresearchUPLOTS: A Unified Pretrained Language Model for Constrained Time-series Generation
arXiv:2606.10466v4 Announce Type: replace-cross Abstract: In time-series generation, existing approaches typically handcraft ortrain a separate model for each dataset, which hinders their scalability and fails to leverag…
- arXiv cs.AIresearchActiveSAM: Fast and Accurate Open-Vocabulary Semantic Segmentation with Frozen SAM 3
arXiv:2606.16996v2 Announce Type: replace-cross Abstract: Segment Anything Model 3 (SAM 3) provides a strong frozen backbone for concept-prompted segmentation, but applying it directly to open-vocabulary semantic segment…
- arXiv cs.AIresearchTemporal Self-Imitation Learning
arXiv:2606.19752v3 Announce Type: replace-cross Abstract: Long-horizon policies trained with reinforcement learning can still achieve high return through inefficient interactions, while rare efficient behaviors discovere…
- arXiv cs.AIresearchLow-power analogue neural networks with trainable nonlinear connections for continuous control
arXiv:2606.23742v2 Announce Type: replace-cross Abstract: Physical neural networks promise low-power machine learning by computing directly with analogue device physics, but most architectures force nonlinear device resp…
- arXiv cs.AIresearchDoes Anthropomorphic Language Impact Public Perceptions of AI?
arXiv:2606.29121v2 Announce Type: replace-cross Abstract: Public discourse about artificial intelligence (AI) often uses anthropomorphic language: language that attributes human capabilities and characteristics to AI sys…
- arXiv cs.AIresearchWeight-Adjusted Gradients Reveal Parameter Importance and Failure Modes in LLMs
arXiv:2607.10803v2 Announce Type: replace-cross Abstract: Understanding which parameters are influential in Large Language Models (LLMs) is central to improving their efficiency, reliability, and interpretability. We int…
- arXiv cs.AIresearchA Safety-First Gateway Architecture for Trusted Public Health Resource Navigation
arXiv:2607.13038v2 Announce Type: replace-cross Abstract: Conversational AI can improve access to public health information, but public-facing healthcare applications require safeguards against inappropriate medical guid…
- arXiv cs.AIresearchDistilled Reinforcement Learning for LLM Post-training
arXiv:2607.17247v2 Announce Type: replace-cross Abstract: Large language model (LLM) post-training is essential for improving reasoning, adaptation, and alignment. Existing methods mainly follow two paradigms: reinforcem…
- arXiv cs.AIresearchLOCKS: Page-Local Compact Key Summaries for Efficient Long-Context Decoding
arXiv:2607.24555v3 Announce Type: replace-cross Abstract: Serving large language models at long context is bottlenecked by the key-value (KV) cache, which is read at every decode step. We find that attention keys are app…
- arXiv cs.AIresearchUnlocking Spatial Grounding in Large Audio-Visual Retrieval models
arXiv:2607.24786v2 Announce Type: replace-cross Abstract: Weak supervision sets a practical regime for audio-visual sound source localization as dense spatial annotations are costly to obtain at scale. The task, however,…
- arXiv cs.AIresearchWhisperRec: Latent Reasoning for Efficient Foundation Recommendation Models
arXiv:2607.26621v4 Announce Type: replace-cross Abstract: Large language models (LLMs) have demonstrated strong reasoning capabilities, motivating their adoption as backbones for foundation recommendation models (FRMs).…
- arXiv cs.AIresearchPAC-MAN: Perception-Aware CBF-RL for Whole-Body Safety in Humanoid Dodgeball
arXiv:2607.28623v2 Announce Type: replace-cross Abstract: We present PAC-MAN, a perception-aware CBF-RL framework that couples control-barrier safety with deployment-realistic onboard sensing for whole-body humanoid dodg…
- arXiv cs.AIresearchTrustworthiness Costs of Domain Adaptation in Small Language Models:A Cross-Architecture Empirical Study
arXiv:2608.00042v2 Announce Type: replace-cross Abstract: Domain adaptation of small language models (SLMs) has emerged as a practical strategy for deploying capable NLP systems in resource-constrained, high-stakes envir…
- arXiv cs.AIresearchSR-OPSD: Self-Referenced On-Policy Self-Distillation
arXiv:2608.09745v2 Announce Type: replace-cross Abstract: On-policy self-distillation (OPSD) converts feedback into dense token-level supervision on student-generated trajectories, complementing reinforcement learning wi…
- arXiv cs.AIresearchFunctional compatibility as a determinant of persistent neural learning
arXiv:2608.22462v3 Announce Type: replace-cross Abstract: Neural networks can acquire new capabilities while damaging existing ones, but what determines whether new learning persists remains unclear. We identify function…
- arXiv cs.AIresearchOn-policy Distillation with Verifiable Reward
arXiv:2608.24696v4 Announce Type: replace-cross Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) and on-policy distillation (OPD) have become two widely adopted paradigms for post-training large language m…
- arXiv cs.AIresearchHomo-RAG: Homology-Guided Retrieval-Augmented Generation for Cross-Species Gene Function Prediction
arXiv:2608.25466v2 Announce Type: replace-cross Abstract: The functional annotation of genes in non-model organisms remains a significant challenge in computational biology, with 20-70% of sequenced genes lacking charact…
- arXiv cs.AIresearchPerformative Privacy: When Differential Privacy Maximizes Utility
arXiv:2608.28198v3 Announce Type: replace-cross Abstract: Privacy-preserving learning is often motivated by the idea that protecting users' data can preserve trust and thus participation, improving utility in the long te…
- arXiv cs.AIresearchThe Illusion of Replacement: Rethinking Specialized Machine Learning Models in the Foundation Model Era
arXiv:2608.28980v3 Announce Type: replace-cross Abstract: Specialized machine learning architectures encode structural assumptions (equivariance, permutation invariance, relational structure) that language-based foundati…
- arXiv cs.AIresearchSimFuse3D: Source-Guided Target Simulation and Confidence-Guided Multi-Stage Localization Reweighting for Cross-Platform 3D Object Detection
arXiv:2609.04886v2 Announce Type: replace-cross Abstract: Changes in sensor height and viewpoint alter object-level point distributions, making cross-platform LiDAR unsupervised domain adaptation (UDA) difficult. Self-tr…
- arXiv cs.AIresearchOnline Surrogate Repair: Decoupling High-Fidelity Feedback from Search Length in Closed-Loop Discovery
arXiv:2609.07655v3 Announce Type: replace-cross Abstract: Closed-loop AI scientists can generate candidate designs at low marginal computational cost, whereas reliable feedback may require wet-lab synthesis, characteriza…
- arXiv cs.AIresearchKalman Delta Networks: Uncertainty-aware Associative Memory
arXiv:2609.07816v2 Announce Type: replace-cross Abstract: Linear attention enables efficient long-context inference by compressing token history into a fixed-size recurrent memory. This compression makes each update a tr…
- arXiv cs.AIresearchAcFlow: Controlling Text-to-Image Diffusion Transformers via Learned Conditional Activation Flow
arXiv:2609.10723v2 Announce Type: replace-cross Abstract: Text-to-image diffusion transformers (DiTs) are powerful generators, yet direct prompting provides limited control interface for style intensity and can fail to s…
- LessWrongresearchTristan Buckmaster Numberphile Interview Transcript
Tristian Buckmaster recently gave an interview with Brady Haran of Numberphile discussing what happened in the Navier-Stokes drama and some context about his research. I'm posting the transcript below for people who pref…
- Hugging Face BlogresearchOpen TTS Leaderboard: Scalable Evaluation for Multilingual Text-to-Speech and Voice Cloning
Open-source model releases, tooling, and community updates.
- Apple Machine Learning ResearchresearchOn the Effectiveness-Fluency Trade-Off in LLM Conditioning: A Systematic Study
Controlling the output of Large Language Models (LLMs) is a central challenge for their reliable deployment, yet a clear understanding of the involved trade-offs remains elusive. Current approaches to conditioning are of…
- LessWrongresearchAgainst strong decision-theoretic realism
N.B. Some of this post argues by analogy between decision theory and values. I expect at least these parts of the post to be unconvincing to anyone who expects sufficiently smart agents to converge on the same values, as…
- SiliconANGLE AIresearchHPE Labs connects AI sovereignty with quantum security
As director of HPE Labs, Andrew Wheeler (pictured) is uniquely positioned to observe rapid advancements in artificial intelligence and new developments for the emerging field of quantum computing. Yet, there is a third a…
- AI Stack ExchangeresearchWhat’s the most effective way to reduce hallucinations in enterprise RAG systems?
We’ve been testing different approaches for reducing hallucinations in production-grade LLM applications, especially in retrieval-augmented generation (RAG) systems used for enterprise workflows. Some observations from o…
- WIRED AIresearchAnthropic Says It Discovered a Crispr-Like System. Now What?
“The experiments are still in the queue. The PR is already live,” says one expert.
- LessWrongresearchAlignment Is to a Virtual Governor: A Theory of Coordination in Diverse Intelligence
Benjamin Lyons , Léo Pio-Lopez , Michael Levin Abstract Alignment is a central problem in systems composed of diverse, interacting components, from biological development to social and engineered systems. One crucial asp…
- LessWrongresearchHow Does Changing the Elo of a Chess Transformer Affect its Computations?
Increasing Skill Level Recruits Deeper Attention Layers in a Frozen Chess Transformer Paper: https://arxiv.org/abs/2609.23917 TL;DR: Maia-3 is a transformer-based chess model that takes Elo (the standard metric for chess…
- LessWrongresearchSome Intuitions on Steering Vectors
This was written with some assistance from Claude Fable 5, Opus 5.5, and GPT-6 Astra in collecting sources and fact-checking claims. This was written quickly, so some slop may leak through despite my best attempts. An Ap…
- Microsoft ResearchresearchIntroducing Quine: An AI research system designed for the complexity of biology
Biology doesn't operate in silos, and neither should the AI representation of it. Quine is an early-stage research effort to create a multimodal world model of biology. By connecting insights across biological scales and…
- The Register AI + MLresearchInvestors are pricing in a 32.6% AI productivity boost for software engineers
Economists turn stock movements into an estimate of anticipated gains – while warning that markets can get carried away
- arXiv cs.AIresearchWitness Overlap: Directional Provenance Inside Open-Weight Model Families
arXiv:2609.31784v2 Announce Type: replace Abstract: Open-weight models are often released, fine-tuned, aligned, merged, and re-released, making provenance audits ask not only whether checkpoints are related, but also whi…