AI news archive.
19,861 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchTARE: Weigh a Never-Poisoned Twin Before Reading Backdoor-Defense Costs
arXiv:2610.06994v1 Announce Type: cross Abstract: Backdoor-defense leaderboards print a clean-accuracy drop and read it as removal cost. Measured on the poisoned victim alone, the drop cannot separate removal from what t…
- arXiv cs.AIresearchMask-Guided KV Cache Eviction in Block Diffusion Language Models
arXiv:2610.06996v1 Announce Type: cross Abstract: Block diffusion language models keep a large key-value (KV) cache throughout generation and attend to it at every denoising step, limiting both memory capacity and genera…
- arXiv cs.AIresearchWhere Does the Audio Jailbreak Live? A Controlled Frequency-Depth Audit of AdvWave-P on Qwen2-Audio
arXiv:2610.07005v1 Announce Type: cross Abstract: We audit frequency and decoder-depth claims for AdvWave-P, an additive audio jailbreak, on Qwen2-Audio. The protocol masks frequency components of the perturbation in the…
- arXiv cs.AIresearchWhich Image Property Carries the Jailbreak? A Controlled Dissection of Image-to-Text Jailbreaks
arXiv:2610.07009v1 Announce Type: cross Abstract: Image-to-text jailbreaks place harmful intent in text, image content, or the relationship between them. We examine image-side factors across four published attack familie…
- arXiv cs.AIresearchAnchor and Adapt: Asymmetric Prompt Adaptation for Few-Shot Industrial Anomaly Detection
arXiv:2610.07016v1 Announce Type: cross Abstract: In few-shot industrial anomaly detection, the few normal target images provide no direct defect supervision, making anomaly prompts difficult to learn from these samples…
- arXiv cs.AIresearchAn Empirical Fault Vulnerability Exploration of ReRAM-based Process-in-Memory CNN Accelerators
arXiv:2610.07029v1 Announce Type: cross Abstract: Resistive random-access memory (ReRAM)-based Processing-in-Memory (PIM) accelerator is a promising platform for processing massively memory intensive matrix-vector multip…
- arXiv cs.AIresearchInvestigating Model Compression for Neural Machine Translation in the Biomedical Domain
arXiv:2610.07032v1 Announce Type: cross Abstract: Large-scale pretrained transformer models have achieved state-of-the-art performance across diverse machine translation tasks, including multilingual settings. Knowledge…
- arXiv cs.AIresearchTRIAGE: Direction-Aware Mismatch Stabilization of Native NVFP4 Reinforcement Learning
arXiv:2610.07043v2 Announce Type: replace-cross Abstract: Low-precision execution can substantially accelerate reinforcement learning (RL) for large language models, but discrepancies between learner and sampler executio…
- arXiv cs.AIresearchLearning to Simulate Individuals from Macro Social Signals
arXiv:2610.07062v1 Announce Type: cross Abstract: Large language models are increasingly used to simulate how individuals respond to new situations, yet the behavioral reasoning behind these responses is either inherited…
- arXiv cs.AIresearchDemo: Vision-Language Model-Guided Online Calibration of an Electromagnetic Digital Twin
arXiv:2610.07081v2 Announce Type: replace-cross Abstract: An electromagnetic (EM) digital twin gives mobile robots wireless situational awareness but depends on material conductivities that change with the environment. O…
- arXiv cs.AIresearchGraph-Based Recognition of Simulated Train-Driver States From Facial and Upper-Body Keypoints
arXiv:2610.07083v2 Announce Type: replace-cross Abstract: Driver fatigue poses a significant challenge to railway safety, with traditional systems like the dead-man switch offering limited and basic alertness checks. Thi…
- arXiv cs.AIagentsSchemaFill: Efficient LLM Tool Calling via Slot-Parallel Speculative Decoding
arXiv:2610.07086v1 Announce Type: cross Abstract: LLM agents interact with external systems by generating structured tool calls. Given a user request, conversational context, and a catalog of tool schemas, a tool-calling…
- arXiv cs.AIagentsTowards a Unified Misuse Monitoring Benchmark
arXiv:2610.07089v1 Announce Type: cross Abstract: LLM agents increasingly act in multi-actor environments, exposing them to misuse from multiple sources: decomposition attacks, where a harmful request is split into innoc…
- arXiv cs.AIresearchT-CCL: Resource Efficient and Performant Collective Communication using Tensor Memory Accelerator
arXiv:2610.07098v1 Announce Type: cross Abstract: Large transformer-based models increasingly depend on multi-GPU execution, which requires frequent collective communication among GPUs. Existing communication libraries o…
- arXiv cs.AIresearchMuon Is Theoretically Wrong For Convolutions, But Empirically Effective
arXiv:2610.07103v1 Announce Type: cross Abstract: Muon, an optimizer known for its efficiency, has a clear interpretation for matrix-valued updates, but convolutional kernels are stored as four-dimensional tensors. Stand…
- arXiv cs.AIresearchBeyond Successor Accuracy: State Retention for Recursive Self-Improvement in Recommendation
arXiv:2610.07105v1 Announce Type: cross Abstract: Recommendation recursive self-improvement (Rec-RSI) feeds recommender outputs into subsequent training. Evaluating each round solely through its latest model assumes that…
- arXiv cs.AIresearchLiLib: Lifelong Air-to-Ground Path-Loss Prediction on UAVs via a Drift-Triggered Model Library
arXiv:2610.07111v1 Announce Type: cross Abstract: UAVs that act as relays or base stations need accurate air-to-ground path-loss predictions for rate adaptation and placement, but propagation conditions change as a UAV m…
- arXiv cs.AIresearchWill the Judge Flip? Predicting Position-Sensitive LLM Judgments from Residual Stream Activations
arXiv:2610.07115v1 Announce Type: cross Abstract: The order in which candidate responses are presented can change an LLM judge's verdict. Detecting such a position flip ordinarily requires judging each pair in both order…
- arXiv cs.AIresearchJailbreaking Open-Weight LLMs via Random Embedding Perturbations
arXiv:2610.07125v1 Announce Type: cross Abstract: While open-weight models have enjoyed steady progress in capabilities and wide adoption across multiple domains, their safety remains an important concern. One key featur…
- arXiv cs.AIresearchPlaySuite: A Large-Scale Benchmark for Interactive Visual Intelligence
arXiv:2610.07127v1 Announce Type: cross Abstract: Recent advances in multimodal foundation models yield strong performance on static perception and reasoning benchmarks, yet such evaluations largely overlook a central as…
- arXiv cs.AIresearchAggregating User Preferences while Ensuring Equity, Diversity, and Inclusion using Graph Summarization
arXiv:2610.07128v1 Announce Type: cross Abstract: Aggregating the preferences of diverse user groups into a collective outcome raises fundamental challenges of equity, diversity, and inclusion (EDI): classical aggregatio…
- arXiv cs.AIresearchCroissantMiner: Automated Extraction and Validation of Croissant Metadata for ML Datasets
arXiv:2610.07132v1 Announce Type: cross Abstract: Croissant has emerged as a standard for machine-readable dataset metadata, yet populating its fields remains labor-intensive and requires careful reading of accompanying…
- arXiv cs.AIresearchCLM-as-a-Judge: Evaluating an Open Contrastive Decision Model on Public Judge Benchmarks
arXiv:2610.07177v1 Announce Type: cross Abstract: An open contrastive decision model is near chance as a judge on the hard public benchmarks: Contrastive-LM/CLM-v0.1-8B scores between 0.351 (best- of-four, chance 0.250)…
- arXiv cs.AIagentsAgentic Design Space Exploration for Joint Hardware Configuration Selection and Mapping of AI Inference Workloads on Heterogeneous Edge SoCs
arXiv:2610.07191v1 Announce Type: cross Abstract: Modern edge Systems-on-Chip (SoCs) integrate heterogeneous processing units (PUs) such as CPUs, GPUs, and NPUs, each with distinct performance and energy characteristics.…
- arXiv cs.AIagentsSPEAR: Five Principles for Interactive Human-Agent Alignment
arXiv:2610.07204v1 Announce Type: cross Abstract: Recent AI alignment work often frames alignment as a pre-deployment optimization problem: collect human feedback, learn preferences or principles, finetune the model, and…
- arXiv cs.AIresearchResponsible Institutional Analytics: Interpreting Bias with AI Support
arXiv:2610.07205v1 Announce Type: cross Abstract: Institutional Analytics (IA) dashboards inform decision-making in higher education, yet data limitations, constraints in analytical techniques, and missing contextual inf…
- arXiv cs.AIresearchDistributionally Robust Mixture-of-Experts Training
arXiv:2610.07207v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) transformers scale capacity by activating only a few experts per token, but this sparsity creates a hidden reliability problem: when routing is i…
- arXiv cs.AIresearchRoboCap: A New Platform for Egocentric Robot Learning
arXiv:2610.07217v2 Announce Type: replace-cross Abstract: Despite its promise for scaling robot learning, egocentric manipulation data is still scarce today. Collection at scale requires vertically integrating ergonomic…
- arXiv cs.AIresearchTIDE 2.0: an open, model-agnostic engine for keyed de-identification of clinical notes
arXiv:2610.07224v1 Announce Type: cross Abstract: Clinical notes capture most of what is documented about a patient's care, but they cannot be used for research until protected health information (PHI) is removed. De-ide…
- arXiv cs.AIresearchMinimal Witness Reinforcement Learning
arXiv:2610.07226v3 Announce Type: replace-cross Abstract: ``What are the irreducible conditions that are sufficient to produce an outcome?'' is one of the most common questions that recur across computation and science.…
- arXiv cs.AIagentsLineage-Aware Memory Governance: A Derivation-Gated Framework for Privacy-Preserving Column-Level Access Control in Enterprise AI Agents
arXiv:2610.07258v1 Announce Type: cross Abstract: Enterprise AI agents that share a memory store face two unaddressed risks: sensitive data can leak through legitimately computed results the requester could not derive, a…
- arXiv cs.AIresearchSAFESHIELD: A Decision-Organization Framework for Deployment-Time Safety of Small Language Models
arXiv:2610.07276v1 Announce Type: cross Abstract: Deployment-time safety of language models is commonly implemented through runtime guardrails such as input moderation, routing, retrieval verification, and output filteri…
- arXiv cs.AIagentsFlexiFlow: Bandit-based Model Switching in ML Workflows
arXiv:2610.07286v1 Announce Type: cross Abstract: Model optimizations help improve inference performance and accuracy of ML workflows. However, relying on a single model to perform inference across all data batches often…
- arXiv cs.AIagentsCatching Developers in the Flow: Low-Latency Agentic Program Repair at Google Scale
arXiv:2610.07289v1 Announce Type: cross Abstract: Manual repair of program failures is time-consuming and disruptive for software developers, particularly during the pre-submit phase where test failures occur within cont…
- arXiv cs.AIresearchPolar: LLM-Powered Synthesis of Real-World Cyber Evidence for Prioritization and Mitigation
arXiv:2610.07298v1 Announce Type: cross Abstract: Cyber threat analysis increasingly depends on evidence distributed across vendor advisories, vulnerability databases, and threat intelligence sources. Turning these fragm…
- arXiv cs.AIresearchFrom Sandbox to Enforcement: Confidence-Qualified Threat Intelligence for Critical Infrastructure
arXiv:2610.07310v1 Announce Type: cross Abstract: Security operations centres and national incident-response teams defending critical infrastructure collect abundant threat data yet struggle to turn it into actionable in…
- arXiv cs.AIresearchScale-Invariant Training for Time Series Foundation Models
arXiv:2610.07324v2 Announce Type: replace-cross Abstract: Time series foundation models (TSFMs) are trained on large collections of time series datasets that span various morphologies and domains. This setting exposes mo…
- arXiv cs.AIresearchMemory-Efficient Expert Routing for Distributed MoE Training
arXiv:2610.07333v1 Announce Type: cross Abstract: As Mixture-of-Experts (MoE) models scale toward hundreds of experts and higher top-$k$ routing, memory efficiency in distributed training becomes a critical bottleneck. P…
- arXiv cs.AIagentsSelective Critique for Cost-Aware LLM Agents in Long-Horizon Decision Making
arXiv:2610.07335v1 Announce Type: cross Abstract: Improving the reliability of large language model (LLM) agents in long-horizon decision-making remains a key challenge. When deployed as autonomous agents interacting wit…
- arXiv cs.AIresearchLogbook: Extremely Long-form Audio Event Understanding
arXiv:2610.07338v1 Announce Type: cross Abstract: Audio benchmarks are built around short, pre-segmented clips, limiting model design to brief inputs or fixed vocabularies. To close this gap, we introduce Logbook, a benc…
- arXiv cs.AIresearchA doctrine-grounded visual question answering dataset for Tactical Combat Casualty Care
arXiv:2610.07339v1 Announce Type: cross Abstract: Tactical Combat Casualty Care (TC3) requires responders to connect visual observations of injuries and interventions with established clinical guidance. Developing vision…
- arXiv cs.AIresearchCausalBind: Causal Modeling and Learning for Protein-Molecule Virtual Screening
arXiv:2610.07340v1 Announce Type: cross Abstract: Protein-molecule virtual screening is increasingly cast as a problem of representation learning in a shared embedding space. Existing methods rely on dense holistic align…
- arXiv cs.AIresearchStepped MoE: Segment-Level Routing with Configurable Inference Complexity
arXiv:2610.07348v1 Announce Type: cross Abstract: Training large language models (LLMs) is resource-intensive, and adapting them for diverse deployment scenarios with varying computational constraints remains challenging…
- arXiv cs.AIagentsRELACE: retrospective likelihood-based action credit estimation for long-horizon language agents
arXiv:2610.07349v1 Announce Type: cross Abstract: Group Relative Policy Optimization (GRPO) avoids a separate critic by estimating advantages from rollout groups. For multi-turn agents, however, trajectory-level supervis…
- arXiv cs.AIresearchA Validated Dataset and Benchmark for Coherent Multi-Diagram SysML Models
arXiv:2610.07356v1 Announce Type: cross Abstract: Systems engineers use several diagrams to describe the structure and behavior of systems. Engineers create these diagrams together to make sure that they use the same ele…
- arXiv cs.AIresearchWildMatch: Weakly Supervised Image Matcher Adaptation for Wildlife Re-Identification
arXiv:2610.07384v1 Announce Type: cross Abstract: Individual animal re-identification from camera-trap imagery is an instance retrieval problem central to non-invasive wildlife monitoring: a query image must retrieve the…
- arXiv cs.AIresearchDeepAJM: Deep Association Joint Model for Irregularly Sampled data
arXiv:2610.07388v2 Announce Type: replace-cross Abstract: Joint Models simultaneously model longitudinal and survival outcomes, leveraging patterns in patients' longitudinal trajectory to improve the prediction of surviv…
- arXiv cs.AIresearchInference and learning in sparse autoencoders as natural gradient flow
arXiv:2610.07389v1 Announce Type: cross Abstract: Sparse autoencoders are widely used to uncover interpretable features in neural networks, yet reliable recovery remains difficult when features overlap or activate infreq…
- arXiv cs.AIresearchActive Feature Acquisition for Cost-Efficient Temporal Prediction with Reduced Participant Burden
arXiv:2610.07452v1 Announce Type: cross Abstract: Accurate forecasting of pathological outcomes is a central problem in psychology. To do so, psychologists often collect intensive longitudinal data. However, in such stud…
- arXiv cs.AIresearchAlignQuant: Tile-Aligned Mixed-Precision Quantization for Efficient LLM Generation
arXiv:2610.07457v1 Announce Type: cross Abstract: Fine-grained mixed-precision quantization promises efficient large language model inference, but local precision choices can conflict with regular GPU storage and computa…