AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchHiLo-Token: Input-Adaptive High-Low Frequency Token Compression for Efficient Image Editing
arXiv:2606.13898v3 Announce Type: replace-cross Abstract: Creative image editing tools, such as Adobe Photoshop's Remove and Generative Fill buttons, are central to everyday customer use and account for a major share of…
- arXiv cs.AIresearchAdaMame: A Training Recipe for Adaptive Multilingual Reasoning
arXiv:2606.15080v2 Announce Type: replace-cross Abstract: While Large Reasoning Models (LRMs) show strong performance in English, they often fail to reason in the language of the query, a phenomenon known as language col…
- arXiv cs.AIresearchA Red-Team Study of Anthropic Fable 5 & Opus 4.8 Models
arXiv:2606.18193v2 Announce Type: replace-cross Abstract: We evaluate the adversarial robustness of three frontier large language models (LLMs) developed by Anthropic, Opus 4.8, Fable 5 and its successor Fable 5.1, again…
- arXiv cs.AIresearchThe Metanym Game: An LLM Benchmark Without Ground Truth That Rises With the Models It Measures
arXiv:2606.21008v4 Announce Type: replace-cross Abstract: We introduce a benchmark that is fully self-contained, needs no ground truth, and rises with the models it measures. Language models compete at making analogies a…
- arXiv cs.AIresearchAdaReP:Adaptive Re-Planning under Model Mismatch for Neural World-Model Predictive Control
arXiv:2606.23079v2 Announce Type: replace-cross Abstract: Neural world models coupled with model predictive control (MPC) replan at every environment step to bound accumulated prediction error, but this incurs substantia…
- arXiv cs.AIresearchInSight: Self-Guided Skill Acquisition via Steerable VLAs
arXiv:2606.24884v2 Announce Type: replace-cross Abstract: Vision-language-action (VLA) models excel at robot manipulation via imitation learning, but adapting them to new tasks often requires additional human demonstrati…
- arXiv cs.AIresearchLearning to Fold: prizewinning solution at LeHome Challenge 2026 (1st place online, 2nd offline)
arXiv:2606.27163v3 Announce Type: replace-cross Abstract: I describe my solution to the LeHome Challenge 2026, an ICRA 2026 competition on bimanual garment folding. The system placed 1st of 62 teams in the online (simula…
- arXiv cs.AIresearchHow Do LLMs Cite? A Mechanistic Interpretation of Attribution in Retrieval-Augmented Generation
arXiv:2606.28358v2 Announce Type: replace-cross Abstract: Retrieval-Augmented Generation (RAG) aims to enhance the trustworthiness of Large Language Models (LLMs) by grounding their outputs in external documents, often u…
- arXiv cs.AIresearchFew-class Fidelity: Evaluating Explanations of Real-conditions CNN classifiers with Optimized Perturbations
arXiv:2606.28391v2 Announce Type: replace-cross Abstract: The wide use of Convolutional Neural Networks (CNN) in numerous domains and real-world classification applications is justified by their high precision and automa…
- arXiv cs.AIresearchOut-of-Distribution Generalization of Risk Aversion in Language Models
arXiv:2607.02755v2 Announce Type: replace-cross Abstract: Training AIs to be risk-averse in resources could offer a failsafe in the event that AIs turn out misaligned. Misaligned but risk-averse AIs would tend to prefer…
- arXiv cs.AIresearchWhy3-py: A Tool for Formal Verification of Hypothesis Testing and Meta-Analysis in Python
arXiv:2607.03951v2 Announce Type: replace-cross Abstract: The reproducibility crisis in scientific research has received widespread recognition, thereby increasing the importance of meta-analyses that integrate statistic…
- arXiv cs.AIresearchHALO-WA: Hybrid-Attention Latent-Guided Online Reinforcement Learning for World-Action Models
arXiv:2607.04265v2 Announce Type: replace-cross Abstract: World-action (WA) models can generate long-horizon action chunks for general-purpose robotic manipulation, but they remain vulnerable to calibration, perception,…
- arXiv cs.AIresearchRiemannian Geometry for Pre-trained Language Model Embeddings
arXiv:2607.07047v3 Announce Type: replace-cross Abstract: Understanding the geometric structure of pre-trained language model embeddings matters for interpretability and safety. We ask whether sentence-level classificati…
- arXiv cs.AIresearchSee like a Robot: Robot-Centric Pointmaps for VLA Models
arXiv:2607.11498v2 Announce Type: replace-cross Abstract: Vision-language-action (VLA) models require 3D spatial reasoning, yet RGB observations encode robot-object geometry only implicitly. Lifting depth with camera int…
- arXiv cs.AIresearchGreedy Volume Maximization of Gradient Embeddings for Long-Tailed Frame-Level Bioacoustic Active Learning
arXiv:2607.13555v2 Announce Type: replace-cross Abstract: Bioacoustic call-type classification relies on costly expert annotation. Active learning can reduce this burden by selecting a small batch of segments for expert…
- arXiv cs.AIresearchInformation-Directed Sampling for Causal Bandits
arXiv:2607.15577v2 Announce Type: replace-cross Abstract: Causal bandits exploit structural relationships among variables to share information across interventions and accelerate the identification of high-reward decisio…
- arXiv cs.AIresearchCANDOR: Chance-Calibrated Neighborhood Discordance in Frozen Encoders for Medical Imaging
arXiv:2607.18451v4 Announce Type: replace-cross Abstract: In medical imaging, one frozen foundation encoder is often reused for many tasks. The encoder is usually chosen by the area under the receiver operating character…
- arXiv cs.AIresearchActive Inference as a Convex Markov Decision Process
arXiv:2607.20152v3 Announce Type: replace-cross Abstract: Active Inference (AIF) frames adaptive behavior as the minimization of expected free energy (EFE), combining epistemic and pragmatic objectives within a single va…
- arXiv cs.AIresearchFakeIDet3-DB: Refining Digital Attacks and Patch Extraction for Secure ID Benchmarking
arXiv:2607.26641v2 Announce Type: replace-cross Abstract: Identity document (ID) authentication relies on the structural integrity of complex, high-frequency security patterns. However, advanced Generative AI models can…
- arXiv cs.AIresearchGraRe: Grasp Candidate Re-Ranking for Frozen 6-DoF Grasp Detectors
arXiv:2608.00946v3 Announce Type: replace-cross Abstract: Existing 6-DoF grasp detectors typically rank grasp candidates by detector confidence. However, our analysis on GraspNet-1Billion shows that detector confidence i…
- arXiv cs.AIresearchTS-MAMP: A Remanufactured Agricultural Robot with Second-Life EV Components and NMS-Free On-Device Weed Detection
arXiv:2608.02270v2 Announce Type: replace-cross Abstract: Agriculture 4.0 robotic systems improve field efficiency yet remain too capital-intensive for the fragmented smallholdings that dominate global agriculture. Meanw…
- arXiv cs.AIresearchWho Should Be Generated? Justifying Demographic Targets in Open-Ended Generation
arXiv:2608.02551v2 Announce Type: replace-cross Abstract: Fairness evaluation concerns not only what a model produces, but also what its outputs ought to be compared against. When a model generates "a CEO in the United S…
- arXiv cs.AIresearchOutput-Aware Rotation for INT2 KV-Cache Quantization
arXiv:2608.02691v4 Announce Type: replace-cross Abstract: The key-value (KV) cache has become a major memory and bandwidth bottleneck in long-context large language model inference, making ultra-low-bit quantization incr…
- arXiv cs.AIresearchBALANCE: Hybrid Autoregressive-Speculative LLM Inference at the Network Edge
arXiv:2608.05926v2 Announce Type: replace-cross Abstract: Edge inference is a promising paradigm to provide large language model (LLM) inference services in next-generation mobile networks. LLM inference mainly relies on…
- arXiv cs.AIresearchFederated Learning for Distributed CNC Tool Wear Prediction
arXiv:2608.11281v3 Announce Type: replace-cross Abstract: Tool wear prediction is an important task in CNC machining, where accurate monitoring of tool condition supports product quality and process reliability. Machine…
- arXiv cs.AIresearchTerminal Symmetry as a Carrier of Asymmetric Process Knowledge: Statewise Refinement for Anytime Verified Construction
arXiv:2608.11318v5 Announce Type: replace-cross Abstract: Many sequential construction tasks have exact terminal symmetries even though execution is directed and depends on history. Process evidence supplies order. Termi…
- arXiv cs.AIresearchHalluTracer: Pre-Decoding Truthfulness Prediction via Depth-Averaged Probe-Logit
arXiv:2608.16353v2 Announce Type: replace-cross Abstract: Internal-state probes enable truthfulness prediction before a large language model generates an answer. When detectors change both the layers they read and the ru…
- arXiv cs.AIresearchCounting Documents Is Not Counting Text: Unit Bias in Web-PDF Corpus Statistics
arXiv:2608.16390v2 Announce Type: replace-cross Abstract: PDF corpora advertise their size in tokens, but every rate they publish (coverage, OCR routing, re-fetch recovery, language mix) is computed per document, and non…
- arXiv cs.AIresearchCan LLMs Reason in a Legally Meaningful Manner? A Small-scale Study on European Court of Human Rights Cases
arXiv:2608.17168v2 Announce Type: replace-cross Abstract: Reasoning has become a standard technique and feature for contemporary LLMs; however, its application and quality in the context of demanding legal-oriented tasks…
- arXiv cs.AIresearchPACE: A Unified Condense-and-Extract Paradigm for Fast VLM Inference
arXiv:2608.27206v2 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) demonstrate exceptional visual reasoning capabilities, yet their inference costs escalate rapidly with the proliferation of visual t…
- arXiv cs.AIresearchCompared to What? A Human-Anchored Security Benchmark for LLM-Generated Infrastructure-as-Code
arXiv:2608.28021v2 Announce Type: replace-cross Abstract: Large language models increasingly author Infrastructure-as-Code (IaC), where one insecure default is provisioned straight into production. Prior evaluations repo…
- arXiv cs.AIresearchSingProbe Technical Report
arXiv:2608.30703v2 Announce Type: replace-cross Abstract: We present SingProbe, an open intrinsic guardrail framework for generation-time monitoring of LLMs. Intrinsic guardrails reuse hidden states already produced by t…
- arXiv cs.AIresearchToward Physically Grounded JEPA World Models for Goal-Conditioned Robotic Planning
arXiv:2609.03565v2 Announce Type: replace-cross Abstract: Action-conditioned JEPA world models enable planning toward visually specified goals without reconstructing future pixels, yet latent prediction alone does not ex…
- arXiv cs.AIresearchA Systematic Evaluation of Cross-Lingual Consistency Enhancement Methods in Multilingual Language Models
arXiv:2609.04409v2 Announce Type: replace-cross Abstract: Multilingual language models often produce inconsistent answers to semantically equivalent questions across languages, motivating methods to improve cross-lingual…
- arXiv cs.AIresearchA Roadmap for MEG Foundation Models
arXiv:2609.04461v2 Announce Type: replace-cross Abstract: Foundation models are beginning to reshape brain-signal analysis by moving the field beyond task-specific decoding pipelines toward reusable models pretrained on…
- arXiv cs.AIresearchVERPO: Verified Evidence Regularized Policy Optimization
arXiv:2609.06100v4 Announce Type: replace-cross Abstract: Verifiable rewards improve language models through reliable task-level feedback, but methods based on Group Relative Policy Optimization (GRPO) apply a sequence-l…
- arXiv cs.AIresearchMind the Gap: Exposing LLM Translation Blind Spots Using the AlphaMWE Multilingual Parallel Corpus
arXiv:2609.06634v2 Announce Type: replace-cross Abstract: LLMs' performance on machine translation (MT) tasks is often dependent on the data availability in the specific domains and language pairs that they are trained u…
- arXiv cs.AIresearchRoofLang: Enabling AI-Driven Architecting of LLM Inference Systems
arXiv:2609.12551v2 Announce Type: replace-cross Abstract: AI is beginning to make substantive contributions to LLM inference optimization. Existing AI optimizations are predominantly profiling-based. Profiling-bound feed…
- arXiv cs.AIresearchA New Transformer-Based Approach for Audio-Based Kinship Verification and a New Uncontrolled Mandarin Kinship Speech Dataset
arXiv:2609.14145v2 Announce Type: replace-cross Abstract: Kinship verification is a task involving determining whether two individuals share a first-order kin relation. To tackle this task, we propose CONVTRAP-TN, a new…
- arXiv cs.AIresearchA Vision-Language Foundation Model for Precise and Comprehensive Brain Tumor Diagnosis from Preoperative Multimodal Data
arXiv:2609.16597v4 Announce Type: replace-cross Abstract: We developed BrainVLM to classify all 12 World Health Organization (WHO) 2021 brain tumor types. BrainVLM integrates an uncertainty quantification strategy to ind…
- arXiv cs.AIresearchBeyond Token-Local Imitation: Reward-Compatible Temporal Credit Assignment for On-Policy Distillation
arXiv:2609.16937v2 Announce Type: replace-cross Abstract: On-policy distillation (OPD) has emerged as an effective approach for large language model post-training, yet existing objectives face a trade-off between objecti…
- arXiv cs.AIresearchBeyond In-Distribution Metrics: A Systematic Out-of-Distribution Evaluation of Congenital Heart Disease Segmentation
arXiv:2609.17068v2 Announce Type: replace-cross Abstract: Congenital heart disease (CHD) diagnosis and surgical planning often require patient-specific 3D anatomical models, but manual segmentation is labor-intensive, pa…
- LessWrongresearchAn opinionated primer on Causal Decision Theory (CDT) and Evidential Decision Theory (EDT)
The first half of this post is a neutral explanation of EDT and CDT. The second half defends EDT and argues against CDT. If you want to learn properly about EDT, you should read Arif Ahmed's Evidence, Decision, and Causa…
- LessWrongresearchStory Imprinting: AI Assistants Absorb Traits from Human Characters They Resemble
This is the abstract, introduction and discussion of our new paper . We also include an addendum on the connection to the Persona Selection Model. Section, appendix, and figure references refer to the full paper. Links:…
- LessWrongresearchWhat if not Circuits?
This post was written as part of the Iliad Fellowship . Inspired by conversations with Richard Ngo, Dmitry Vaintrob, and Brianna Grado-White. To all of these, my thanks. Preface: I'm confused about how neural networks do…
- LessWrongresearchCLT Features Support Manifold Steering, but Hold Only Partial Steering Signal
tl;dr I continue work to extend the Goodfire lab's cyclical manifold methods to Anthropic's pre-trained CLT feature space on Gemma-2-2B. I find that manifold steering produces the expected cyclical transition between day…
- TechCrunch AIresearchOpenAI forms math advisory group as its AI resolves more than 100 open problems
The group won't be given leeway to slow down or redirect OpenAI's ongoing mathematical research.
- NVIDIA Developer BlogresearchAccelerating a ROS 2 Node with an AI Agent and NVIDIA Isaac ROS
GPU acceleration can speed up compute-intensive robotics workloads, but a fast CUDA kernel alone does not guarantee a fast ROS 2 graph. As messages move between...
- LessWrongresearchTurbulence & Fragility: Prompting-Based Experiments are Sensitive to Stray Details (And other lessons for new researchers.)
Many interesting experiments can be done with prompt engineering to elicit behavior from LLMs and attempt to determine what their drives are; what behaviors they are at risk of as general patterns rather than when prompt…
- Tom's Guide AIresearchAI could be costing you money: new study finds chatbots get most financial questions wrong
Consumer AI launches, tools, and practical coverage.