AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchUnlocking Pretrained Vision Transformers for Time Series Classification
arXiv:2506.08641v3 Announce Type: replace-cross Abstract: Adapting vision models for time series analysis is compelling, yet all existing approaches are falling short of dedicated time series foundation models (TSFMs) in…
- arXiv cs.AIresearchCalibrating Lightweight Sparse Autoencoder Feature Steering
arXiv:2506.12576v3 Announce Type: replace-cross Abstract: Sparse autoencoders (SAEs) can enable inference-time topic steering by modifying latent feature activations, but existing steering methods often fail when target-…
- arXiv cs.AIresearchDiscrete Tokenization for Multimodal LLMs: A Comprehensive Survey
arXiv:2507.22920v2 Announce Type: replace-cross Abstract: The rapid advancement of large language models (LLMs) has intensified the need for effective mechanisms to transform continuous multimodal data into discrete repr…
- arXiv cs.AIresearchSalQ-VLM: Fine-Grained Saliency-Guided Quantization for Vision-Language Models
arXiv:2508.03351v3 Announce Type: replace-cross Abstract: Large language models (LLMs) have demonstrated remarkable capabilities across diverse language tasks, motivating their extension to vision-language models (VLMs)…
- arXiv cs.AIresearchADSEL: Adaptive Dual Self-Expression Learning for EEG Feature Selection via Incomplete Multi-Dimensional Emotion Labels
arXiv:2508.05229v2 Announce Type: replace-cross Abstract: EEG based multi-dimension emotion recognition has attracted substantial research interest in affective computing. However, the high dimensionality of EEG features…
- arXiv cs.AIresearchBiased AI improves human performance but reduces perceived helpfulness
arXiv:2508.09297v4 Announce Type: replace-cross Abstract: Artificial intelligence (AI) increasingly shapes how people think, engage, and evaluate information. To minimize risk, most current systems are designed to presen…
- arXiv cs.AIresearchBridgeShield: Risk-Aware Graph Modeling for Cross-Chain Bridge Attack Detection
arXiv:2508.20517v2 Announce Type: replace-cross Abstract: Cross-chain bridges enable asset and state transfers across heterogeneous blockchains, but their complex cross-domain interactions introduce new attack surfaces t…
- arXiv cs.AIresearchRecidivism Prediction, Peer Effect Estimation, and Prediction-Powered Inference with LLM Text Measures
arXiv:2509.20634v3 Announce Type: replace-cross Abstract: We provide a new framework for estimating peer effects when outcomes are multivariate behavioral measures derived from written text using an LLM and the network f…
- arXiv cs.AIresearchIncorporating LLM Embeddings for Variation Across the Human Genome
arXiv:2509.20702v3 Announce Type: replace-cross Abstract: Recent advances in large language model (LLM) embeddings have enabled powerful representations for biological data, but most applications to date focus on gene-le…
- arXiv cs.AIresearchWhat Is The Political Content in LLMs' Pre- and Post-Training Data?
arXiv:2509.22367v3 Announce Type: replace-cross Abstract: Large language models (LLMs) reflect politically-slanted opinions in their generated text. Even though it is widely assumed that model behavior stem from training…
- arXiv cs.AIresearchVRWKV-Editor: Reducing quadratic complexity in transformer-based video editing
arXiv:2509.25998v4 Announce Type: replace-cross Abstract: In light of recent progress in video editing, deep learning models focusing on both spatial and temporal dependencies have emerged as the primary method. However,…
- arXiv cs.AIresearchSPID: Distilled Protein Backbone Generation
arXiv:2510.03095v4 Announce Type: replace-cross Abstract: Diffusion- and flow-based generative models have recently demonstrated strong performance in protein backbone generation tasks, offering unprecedented capabilitie…
- arXiv cs.AIresearchBimanual 3D Hand Motion and Articulation Forecasting in Everyday Images
arXiv:2510.06145v2 Announce Type: replace-cross Abstract: We tackle the problem of forecasting bimanual 3D hand motion and articulation from a single image in everyday settings. To address the lack of 3D hand annotations…
- arXiv cs.AIresearchAccelerated stochastic first-order method for convex optimization under heavy-tailed noise
arXiv:2510.11676v2 Announce Type: replace-cross Abstract: We study convex composite optimization problems, where the objective function is given by the sum of a prox-friendly function and a convex function whose subgradi…
- arXiv cs.AIresearchBreakFun: Jailbreaking LLMs via Object Instantiation under Simulated Code Execution
arXiv:2510.17904v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are widely used because they process structures, syntax and code well, but this same ability also makes them paradoxically vulnerable…
- arXiv cs.AIresearchHeterogeneous Robot Collaboration in Unstructured Environments with Grounded Generative Intelligence
arXiv:2510.26915v2 Announce Type: replace-cross Abstract: While heterogeneous teams have typically been designed for well-specified missions with known semantics, generative intelligence, i.e., large language models (LLM…
- arXiv cs.AIresearchGround-Truth Subgraphs for Better Training and Evaluation of Knowledge Graph Augmented LLMs
arXiv:2511.04473v4 Announce Type: replace-cross Abstract: Retrieval of information from graph-structured knowledge bases represents a promising direction for improving the factuality of LLMs. While various solutions have…
- arXiv cs.AIresearchBudgetMem: Training-Free Selective Memory for Cost-Efficient Long-Context Processing in Language Models
arXiv:2511.04919v3 Announce Type: replace-cross Abstract: Processing long documents with large language models (LLMs) is expensive: a single query over a 100K-token document can cost from tens of cents to over a dollar i…
- arXiv cs.AIresearchDiffusion Model in Latent Space for Medical Image Segmentation Task
arXiv:2512.01292v5 Announce Type: replace-cross Abstract: Medical image segmentation is crucial for clinical diagnosis and treatment planning. Traditional methods typically produce a single segmentation mask, failing to…
- arXiv cs.AIresearchTime Series Foundation Models for Process Model Forecasting
arXiv:2512.07624v3 Announce Type: replace-cross Abstract: Process Model Forecasting (PMF) aims to predict how the control-flow structure of a process evolves over time by modeling the temporal dynamics of directly-follow…
- arXiv cs.AIresearchSlot-ID: Identity-Preserving Video Generation from Reference Videos via Slot-Based Temporal Identity Encoding
arXiv:2601.01352v2 Announce Type: replace-cross Abstract: Human identity-preserving text-to-video generation remains challenging under large changes in viewpoint, facial expression, illumination, and motion. Existing met…
- arXiv cs.AIresearchDecoding Order Matters in Autoregressive Speech Synthesis
arXiv:2601.08450v2 Announce Type: replace-cross Abstract: Autoregressive speech synthesis often adopts a left-to-right order, yet generation order is a modelling choice. We investigate decoding order through masked diffu…
- arXiv cs.AIresearchILRR: Inference-Time Steering Method for Masked Diffusion Language Models
arXiv:2601.21647v2 Announce Type: replace-cross Abstract: Discrete Diffusion Language Models (DLMs) offer a promising non-autoregressive alternative for text generation, yet effective mechanisms for inference-time contro…
- arXiv cs.AIresearchHybridFlow: A 2-NFE Generative Policy for Real-Time Robotic Manipulation
arXiv:2602.13718v2 Announce Type: replace-cross Abstract: Generative policies for robotic manipulation must balance action accuracy with inference latency. We present HybridFlow, a three-stage policy inference procedure…
- arXiv cs.AIresearchSOTAlign: Semi-Supervised Alignment of Unimodal Vision and Language Models via Optimal Transport
arXiv:2602.23353v3 Announce Type: replace-cross Abstract: The Platonic Representation Hypothesis posits that neural networks trained on different modalities converge toward a shared statistical model of the world. Recent…
- arXiv cs.AIresearchScaling Laws of SignSGD in Linear Regression: When Does It Outperform SGD?
arXiv:2603.02069v2 Announce Type: replace-cross Abstract: We study scaling laws of signSGD under a power-law random features (PLRF) model that accounts for both feature and target decay. We analyze the population risk of…
- arXiv cs.AIresearchWhen Scaling Fails: Mitigating Audio Perception Decay of LALMs via Multi-Step Perception-Aware Reasoning
arXiv:2603.02266v2 Announce Type: replace-cross Abstract: Test-Time Scaling has shown notable efficacy in addressing complex problems through scaling inference compute. However, within Large Audio-Language Models (LALMs)…
- arXiv cs.AIresearchAn Interpretable, Controllable Time-Varying IIR Denoiser for On-Device Assistive Hearing
arXiv:2603.02794v4 Announce Type: replace-cross Abstract: We present TVBC (Time-Varying Biquad Cascade), an interpretable, low-latency speech enhancement model for real-time, on-device assistive hearing. A lightweight ne…
- arXiv cs.AIresearchDial: A Knowledge-Grounded Dialect-Specific NL2SQL System
arXiv:2603.07449v2 Announce Type: replace-cross Abstract: Enterprises commonly deploy heterogeneous database systems, each of which owns a distinct SQL dialect with different syntax rules, built-in functions, and executi…
- arXiv cs.AIresearchSAIL: Test-Time Scaling for In-Context Imitation Learning with VLM
arXiv:2603.08269v2 Announce Type: replace-cross Abstract: In-context imitation learning allows robots to acquire skills from demonstrations, yet one-shot trajectory generation remains fragile under environmental variatio…
- arXiv cs.AIresearchCCTU: A Benchmark for Tool Use under Complex Constraints
arXiv:2603.15309v2 Announce Type: replace-cross Abstract: Solving problems through tool use under explicit constraints constitutes a highly challenging yet unavoidable scenario for large language models (LLMs), requiring…
- arXiv cs.AIresearchFrom Noise to Signal: When Outliers Seed New Topics
arXiv:2603.18358v2 Announce Type: replace-cross Abstract: Outliers in dynamic topic modeling are typically treated as noise, yet we show that some can serve as early signals of emerging topics. We introduce a temporal ta…
- arXiv cs.AIresearchScaling Sim-to-Real VLA Reinforcement Learning with Generative 3D Worlds
arXiv:2603.18532v4 Announce Type: replace-cross Abstract: The strong performance of large vision-language models (VLMs) trained with reinforcement learning (RL) has motivated similar approaches for fine-tuning vision-lan…
- arXiv cs.AIresearchLearning Energy-Efficient Air--Ground Actuation for Hybrid Robots on Stair-Like Terrain
arXiv:2603.26687v2 Announce Type: replace-cross Abstract: Hybrid aerial--ground robots can use thrust to cross obstacles that impede wheel-driven motion, but deciding how much thrust to apply during contact remains chall…
- arXiv cs.AIresearchLG-HCC: Local Geometry-Aware Hierarchical Context Compression for 3D Gaussian Splatting
arXiv:2603.28431v4 Announce Type: replace-cross Abstract: Although 3D Gaussian Splatting (3DGS) enables high-fidelity real-time rendering, its prohibitive storage overhead severely hinders practical deployment. Recent an…
- arXiv cs.AIresearchWhat Makes Good Multilingual Reasoning? Disentangling Traces with Measurable Features
arXiv:2604.04720v2 Announce Type: replace-cross Abstract: Large Reasoning Models (LRMs) still exhibit large performance gaps between English and other languages, yet much current work assumes these gaps can be closed sim…
- arXiv cs.AIresearchNot All Forgetting Is Equal: Retention Dynamics in Fine-Tuned Image Classifiers
arXiv:2604.11508v3 Announce Type: replace-cross Abstract: Fine-tuning a pretrained classifier leaves some samples reliably learned and others cycling between correct and incorrect. Curriculum learning, data pruning and d…
- arXiv cs.AIresearchEnglish is Not All You Need: Systematically Exploring the Role of Multilinguality in LLM Post-Training
arXiv:2604.13286v2 Announce Type: replace-cross Abstract: Despite the widespread multilingual deployment of large language models, post-training pipelines remain predominantly English-centric, contributing to performance…
- arXiv cs.AIresearchLearning from Noisy Preferences: A Semi-Supervised Learning Approach to Direct Preference Optimization
arXiv:2604.24952v3 Announce Type: replace-cross Abstract: Human visual preferences are inherently multi-dimensional, encompassing aesthetics, detail fidelity, and semantic alignment. However, existing datasets provide on…
- arXiv cs.AIresearchUsing LLMs in Software Design: An Empirical Study of GitHub and A Practitioner Survey
arXiv:2605.01392v2 Announce Type: replace-cross Abstract: Recent advancements in Large Language Models (LLMs) have demonstrated significant potential across software engineering tasks, including software design, an area…
- arXiv cs.AIresearchProbe-Geometry Alignment: Erasing the Cross-Sequence Memorization Signature Below Chance
arXiv:2605.01699v4 Announce Type: replace-cross Abstract: Recent attacks show that behavioural unlearning of large language models leaves internal traces recoverable by adversarial probes. We characterise where this rete…
- arXiv cs.AIresearchRelational Retrieval: Leveraging Known-Novel Interactions for Generalized Category Discovery
arXiv:2605.09420v3 Announce Type: replace-cross Abstract: In this study, we tackle Generalized Category Discovery (GCD) via a Relational Retrieval perspective, explicitly coupling labeled and unlabeled data through bidir…
- arXiv cs.AIresearchTASTE: A Designer-Annotated Multi-Dimensional Preference Dataset for AI-Generated Graphic Design
arXiv:2605.20731v3 Announce Type: replace-cross Abstract: Text-to-image models now generate graphic design at production scale, yet their supervision still comes primarily from photo-style preference datasets with a sing…
- arXiv cs.AIresearchAction with Visual Primitives
arXiv:2605.22183v4 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) models have emerged as a promising paradigm for generalist robotic manipulation. A common design in current architectures maps langua…
- arXiv cs.AIresearchFinite-Particle Convergence Rates for Conservative and Non-Conservative Drifting Models
arXiv:2605.22795v3 Announce Type: replace-cross Abstract: We analyze finite-particle drifting models for one-step generative modeling. For a conservative velocity given by the difference of the kernel-smoothed data and m…
- arXiv cs.AIresearchReflex: Reinforcement Learning with Reflection Symmetry Exploitation in State-Based Continuous Control
arXiv:2605.23415v3 Announce Type: replace-cross Abstract: Reinforcement learning has long struggled with poor sample efficiency. One promising approach to mitigate this problem is leveraging group-invariant Markov Decisi…
- arXiv cs.AIresearchSiST-GNN: Simultaneous Spatial-Temporal Message Passing for Dynamic Graph Representation Learning
arXiv:2605.25548v2 Announce Type: replace-cross Abstract: Dynamic graph neural networks (DGNNs) that operate on snapshot sequences typically fall into one of two categories. \emph{Temporal-first} approaches build per-nod…
- arXiv cs.AIresearchDDGAD: Disagreement-Driven Graph Anomaly Detection via Adapt-Then-Combine
arXiv:2605.26446v3 Announce Type: replace-cross Abstract: Graph anomaly detection (GAD) commonly relies on message passing to jointly encode node attributes and neighborhood context. However, once the two are mixed, an a…
- arXiv cs.AIresearchSUPREME: A Multi-GPU Framework for Reproducible Image Unlearning Method Evaluation
arXiv:2606.00380v3 Announce Type: replace-cross Abstract: Machine unlearning removes the influence of specific training data from a trained model without retraining it from scratch. Evaluating an unlearning method requir…
- arXiv cs.AIresearchvla.cpp: A Unified Inference Runtime for Vision-Language-Action Models
arXiv:2606.08094v2 Announce Type: replace-cross Abstract: Deploying vision--language--action (VLA) models on robots requires adapting model-specific inference pipelines to heterogeneous processors and limited onboard mem…