AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchGrapeSplat: Geometry-Grounded Reconstruction via Amalgamated Pose-Free Encoding for Feed-Forward 3D Gaussian Splatting
arXiv:2609.23182v1 Announce Type: cross Abstract: Feed-forward 3D Gaussian Splatting now reconstructs renderable scenes from unposed, uncalibrated images. Yet, most models supervise only photometric consistency and predi…
- arXiv cs.AIresearchWhy Ghost Outputs Teach: A Kernel-Based Understanding of Subliminal Learning
arXiv:2609.23260v1 Announce Type: cross Abstract: Subliminal Learning (SL) is a recently identified phenomenon in which a student model acquires downstream task capabilities by matching seemingly unrelated auxiliary outp…
- arXiv cs.AIresearchKnowing When to Trust Images: Reliability-Aware Multi-modal Entity Alignment
arXiv:2609.23267v1 Announce Type: cross Abstract: The visual modality, i.e., images, plays a key role in multi-modal entity alignment (MMEA). Existing approaches often directly fuse the image with other modalities to ali…
- arXiv cs.AIresearchValueDiff: Value-Geometric KV Cache Eviction for Sink-Suppressed LLMs
arXiv:2609.23314v2 Announce Type: replace-cross Abstract: Modern LLMs with QK-normalization, gated attention, learned attention sinks, or logit softcapping exhibit weaker persistent attention sinks, on which existing KV…
- arXiv cs.AIresearchCo-occurrence Patterns of LoRA Adapters in Production Diffusion Model Inference Services
arXiv:2609.23321v1 Announce Type: cross Abstract: Low-rank adaptation (LoRA) has become a key technology for serving large-scale personalized large language models and diffusion models in the cloud. However, the co-occur…
- arXiv cs.AIresearchA Patient World Model for Early Forecasting of Digital Health Campaign Outcomes: Capabilities and Limits
arXiv:2609.23333v1 Announce Type: cross Abstract: Digital direct-to-consumer (DTC) health campaigns are usually measured after the fact. In-flight forecasting commonly relies on a separate classifier for every cutoff and…
- arXiv cs.AIresearchDiscovering Physical Representation Languages
arXiv:2609.23381v1 Announce Type: cross Abstract: Before a machine can discover a physical law, it must discover what its measurements are: which observations live on cells, which are intensive or extensive, which sector…
- arXiv cs.AIresearchBayesian Filtering in Physical Systems via Test-time Trained Flow Matching
arXiv:2609.23383v1 Announce Type: cross Abstract: Bayesian filtering provides a principled framework for online state estimation under uncertainty, yet its application to systems with high-dimensional states and complica…
- arXiv cs.AIresearchBlind Thermodynamic Ontology Discovery from Anonymous Experiments
arXiv:2609.23387v2 Announce Type: replace-cross Abstract: Before a machine learning model can learn a thermodynamic equation of state, it must discover what its measurements represent: which channels scale with system si…
- arXiv cs.AIresearchEnhancing Shrimp Disease Detection via Deep Learning and Data Refinement for Resilient Aquaculture
arXiv:2609.23397v1 Announce Type: cross Abstract: Shrimp diseases continue to cause devastating losses in the aquaculture industry, driving a critical need for robust, automated detection. This work contributes the first…
- arXiv cs.AIresearchBeyond PUE: A Local Impact Audit Framework for Data Center Environmental Accountability
arXiv:2609.23421v1 Announce Type: cross Abstract: Standard data center sustainability metrics, including Power Usage Effectiveness (PUE), Water Usage Effectiveness (WUE), and Carbon Usage Effectiveness (CUE), measure a f…
- arXiv cs.AIresearchRiverVLN: Phase-Grounded Temporal Vision--Language Navigation for Unmanned Surface Vehicles
arXiv:2609.23423v1 Announce Type: cross Abstract: Vision-language navigation (VLN) has largely been developed for indoor and terrestrial robots, where language can often be treated as a static goal and motion is approxim…
- arXiv cs.AIresearchPhysReflect: Geometry and Perception Guided Diffusion for Physically-Plausible Mirror Reflections
arXiv:2609.23442v1 Announce Type: cross Abstract: Diffusion models generate high-quality images, yet often violate the physical laws governing mirror reflections. Reflections often suffer from geometric aberrations, incl…
- arXiv cs.AIresearchRLVR$^{2}$: Reinforcement Learning with Verifiable Rubric-based Ranking
arXiv:2609.23457v2 Announce Type: replace-cross Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) is expanding from tasks with well-defined correctness signals, such as mathematics and code, toward multifac…
- arXiv cs.AIresearchHeating in human-HVAC interaction for smart homes: An interdisciplinary overview
arXiv:2609.23508v1 Announce Type: cross Abstract: As part of HVAC systems, residential heating provides foundational infrastructure for human habitation in cold weather. However, research on how residents interact with H…
- arXiv cs.AIresearchPredicting Out-of-Distribution Generalization of Neural Operators via Observable Spectral Error Decomposition
arXiv:2609.23529v1 Announce Type: cross Abstract: Neural operators have emerged as powerful surrogates for solving partial differential equations (PDEs), yet their reliability under distribution shift remains a critical…
- arXiv cs.AIresearchPhysics-residual machine learning predicts oxygen-evolution catalyst activity beyond the training range from sparse polarization measurements
arXiv:2609.23549v2 Announce Type: replace-cross Abstract: Discovery campaigns for oxygen evolution reaction catalysts repeatedly choose, make and measure catalysts. High-throughput platforms stop polarization curves belo…
- arXiv cs.AIresearchMaskVLA: Visual Masking Against Trajectory Overfitting of Vision-Language-Action Model
arXiv:2609.23565v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models integrate vision-language understanding with executable robot actions, enabling end-to-end learning for robot control. However, our em…
- arXiv cs.AIresearchARID: A Deployable Edge AI System for Structured Information Extraction from Industrial Maintenance Work Orders
arXiv:2609.23582v1 Announce Type: cross Abstract: Maintenance work orders must often be processed offline on embedded hardware, yet downstream software requires predictable structured output. We present ARID (Aviation-in…
- arXiv cs.AIresearchOn the Efficiency-Safety Dilemma in Large Reasoning Models
arXiv:2609.23587v1 Announce Type: cross Abstract: Large reasoning models (LRMs) incur high inference costs, often mitigated by efficiency techniques like quantization and pruning. However, the impact of these techniques…
- arXiv cs.AIresearchListen Then Reason: Perception-Grounded Test-Time Reinforcement Learning for Large Audio-Language Models
arXiv:2609.23589v1 Announce Type: cross Abstract: Large audio-language models (LALMs) are increasingly used for a broader range of audio reasoning tasks. These models typically incorporate audio representations into a la…
- arXiv cs.AIresearchCost-Aware Reinforcement Learning with Action Masking and Projection for Battery Energy Storage Dispatch under Suppressed-Spread Market Shifts
arXiv:2609.23590v1 Announce Type: cross Abstract: Battery energy storage system (BESS) dispatch must preserve operational feasibility while declining price spreads reduce the margin available to pay for cycling. We study…
- arXiv cs.AIresearchStyleAT: Defending Face Recognition Against Semantic Attacks
arXiv:2609.23596v1 Announce Type: cross Abstract: With face-recognition models now embedded in everyday authentication and surveillance, recent works have pinpointed a critical weakness: these models remain acutely vulne…
- arXiv cs.AIresearchPETR: Prompt Ensembling with Training-free Routing for Vision-Language Models
arXiv:2609.23600v1 Announce Type: cross Abstract: Prompt learning efficiently adapts vision-language models (VLMs) to downstream tasks, but gains on seen classes often come at the expense of generalization to unseen clas…
- arXiv cs.AIresearchPREM: Prefix-Steered Recurrent Memory for Long-Video Understanding
arXiv:2609.23601v1 Announce Type: cross Abstract: Long-video understanding must capture transient visual evidence under strict token budgets, yet existing methods compress frames, append memory tokens, or alter internal…
- arXiv cs.AIresearchDo Language Models Need Music Supervision? Verifiable Rewards for Multi-Constraint Symbolic Music Generation
arXiv:2609.23665v3 Announce Type: replace-cross Abstract: Language models can now generate symbolic music from natural-language instructions. However, existing models still struggle to satisfy multiple explicit musical c…
- arXiv cs.AIresearchSmoothed Analysis of Inconsistent A*
arXiv:2609.23680v1 Announce Type: cross Abstract: The A* search is a fundamental path-finding algorithm in artificial intelligence. While admissible and consistent heuristics guarantee efficient performance by expanding…
- arXiv cs.AIresearchA multi-temporal dataset for mapping burned areas in the Brazilian Cerrado using time series of remote sensing imagery
arXiv:2609.23687v1 Announce Type: cross Abstract: This paper introduces a multi-temporal tabular dataset derived from satellite images to map burned areas in the Chapada dos Veadeiros National Park, in Goi\'as, Brazil, c…
- arXiv cs.AIresearchTail-Weight Control and Localized Generalization in Nearly Low-Rank Adversarial Classification
arXiv:2609.23688v3 Announce Type: replace-cross Abstract: Empirical ramp fitting can assign weight to pure-noise features even when the population optimum ignores them. We quantify this gap for norm-constrained adversari…
- arXiv cs.AIresearchSTEVE: Stabilizing Textual Gradient-Based Prompt Optimization via Error-Driven Refinement and Regularized Verification
arXiv:2609.23716v1 Announce Type: cross Abstract: Textual-gradient methods automate prompt optimization through natural-language feedback, but their iterative updates can be unstable. We identify two sources of this inst…
- arXiv cs.AIresearchGRACE: Grounded Adversarial Reasoning over Canadian Law
arXiv:2609.23726v1 Announce Type: cross Abstract: Large language models have shown strong performance across a range of legal tasks, but existing benchmarks rarely evaluate the ability to take and defend a legal position…
- arXiv cs.AIresearchTriFleetRCA: On-Premise LLM Root Cause Analysis for Kubernetes
arXiv:2609.23766v1 Announce Type: cross Abstract: Root cause analysis at a remote site is slow: evidence is scattered across pod logs, Kubernetes events and cluster-level objects, and many operators cannot send productio…
- arXiv cs.AIresearchBelted Engression: Sufficient Dimension Reduction for Generative Distributional Regression
arXiv:2609.23789v1 Announce Type: cross Abstract: Modern conditional generative models face significant challenges when learning complex covariate dependencies. While sufficient dimension reduction (SDR) provides a princ…
- arXiv cs.AIresearchFederated Multilingual Speech-LLMs: Architecture and Aggregation Strategy Benchmarking
arXiv:2609.23825v1 Announce Type: cross Abstract: We present a comprehensive benchmark of Federated Learning (FL) for multilingual Automatic Speech Recognition (ASR), evaluating four Speech-LLM architectures on the Multi…
- arXiv cs.AIresearchFrom UNDRR Reports to Event Records: Schema-Constrained LLM Extraction of Georeferenced Disasters
arXiv:2609.23853v1 Announce Type: cross Abstract: Disaster-risk-reduction archives describe hazard events in prose that databases such as EM-DAT (Delforge et al., 2025) cannot ingest directly. We present an LLM pipeline…
- arXiv cs.AIresearchGLR-MM: Graph-Based Global-Local Reconstruction for Robust Multimodal Chest X-ray and EHR Representation Learning under Missing Modalities
arXiv:2609.23876v1 Announce Type: cross Abstract: Clinical multimodal models must often predict before all chest X-ray (CXR) and electronic health record (EHR) inputs are available. Existing approaches align observed rep…
- arXiv cs.AIresearchSyzHarness: Patch-Based Kernel Bug Reproduction with LLM-Synthesized Fuzzing Harnesses
arXiv:2609.23889v3 Announce Type: replace-cross Abstract: Automated kernel vulnerability reproduction is essential for bug triage, patch validation, and regression testing, but still lacks an effective and efficient solu…
- arXiv cs.AIresearchCircuit-Diff: Factual Edit-based Intervention Method for Localizing Knowledge in Attribution Graphs
arXiv:2609.23892v1 Announce Type: cross Abstract: Mechanistic interpretability defines features as the fundamental units of a neural network and circuits as the weighted subgraphs that carry out its computation. Because…
- arXiv cs.AIresearchReVeal: A Reconstruction-Aware Real-to-Sim Framework for VLA Policy Evaluation
arXiv:2609.23910v1 Announce Type: cross Abstract: Simulation-based evaluation provides a scalable and repeatable alternative to real-world evaluation of vision-language-action (VLA) policies. However, reconstruction erro…
- arXiv cs.AIresearchHaikuS2S: A Cascaded System For Responding In Verse
arXiv:2609.23951v1 Announce Type: cross Abstract: Expressive speech synthesis has advanced through prosody modeling, yet generating structured poetic speech, such as haiku, remains challenging. Prior work on prosody tran…
- arXiv cs.AIresearchDjinnlang: Higher-Level Programming by Unambiguous Specification with an LLM in the Compiler
arXiv:2609.23954v1 Announce Type: cross Abstract: Programmers write formal specifications, and LLMs implement them, proving that each implementation matches its spec. Taken to its extreme, this makes specification langua…
- arXiv cs.AIresearchRoboTalk: Learning Multi-Robot Communication and Coordination from Multimodal Demonstrations
arXiv:2609.23997v2 Announce Type: replace-cross Abstract: Multi-robot collaboration could enable more efficient and scalable solutions to complex robotic tasks, but collaboration under partial observability remains chall…
- arXiv cs.AIresearchMisaligned Clinical Risk Classification and Cost Asymmetry in Open-Weight Large Language Models
arXiv:2609.23999v3 Announce Type: replace-cross Abstract: How large language models (LLMs) integrate patient risk with clinical cost tradeoffs remains poorly understood. We investigated how four open-weight LLMs (Qwen-2.…
- arXiv cs.AIresearchInterHier: Learning Interconnected Hierarchical Semantics for Open-Vocabulary Object Detection
arXiv:2609.24026v1 Announce Type: cross Abstract: In this paper, we investigate the limitations of fixed, hand-crafted connectors in hierarchical semantic representations for open-vocabulary object detection. Existing me…
- arXiv cs.AIresearchMECT: Mixture of Experts with CNN-Transformer Network for Speaker verification
arXiv:2609.24061v1 Announce Type: cross Abstract: In this paper, we propose MECT, a speaker verification model that integrates the Mixture-of-Experts (MoE) mechanism into a CNN-Transformer backbone with optimized block s…
- arXiv cs.AIresearchFrom Content Generation to Learning Support: Pedagogy-Guided Generative Video Tutors for STEM Learning
arXiv:2609.24083v1 Announce Type: cross Abstract: Generative AI enables scalable production of educational videos, but current systems largely focus on producing visually coherent content rather than supporting learning.…
- arXiv cs.AIresearchFrom Bits to Beliefs: Recoverable Semantic Fingerprints for Black-Box Verification of Large Language Models
arXiv:2609.24084v1 Announce Type: cross Abstract: Open-weight large language models (LLMs) can be copied, modified, and redeployed behind black-box APIs, making post-release ownership verification difficult. Existing bla…
- arXiv cs.AIresearchFlashBoB: I/O-Efficient Exact Backward-over-Backward for Softmax Attention
arXiv:2609.24089v2 Announce Type: replace-cross Abstract: Transformer models built on the attention mechanism have become a central building block in modern deep learning, yet softmax attention remains a major bottleneck…
- arXiv cs.AIresearchActiveArena: Benchmarking and Understanding Active Perception in Robotic Manipulation
arXiv:2609.24124v2 Announce Type: replace-cross Abstract: Active perception and manipulation are crucial for robots to interact with complex scenes. Existing benchmarks struggle to evaluate how robots effectively acquire…
- arXiv cs.AIresearchSTAR: Scene- and Task-Aware 4D Radar Preprocessing Towards End-to-End Cognitive Radar
arXiv:2609.24151v1 Announce Type: cross Abstract: Four-dimensional (4D) Radar has emerged as a key sensor for environmental perception, providing range, azimuth, elevation, and Doppler measurements while remaining robust…