AI news archive.
19,861 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchLocalizing Persona Representations in LLMs
arXiv:2505.24539v5 Announce Type: replace-cross Abstract: We present a study on how and where personas -- defined by distinct sets of human characteristics, values, and beliefs -- are encoded in the representation space…
- arXiv cs.AIresearchBenchmarks Are Not Atomic: Composition-Aware LLM Evaluation using BenchHub
arXiv:2506.00482v2 Announce Type: replace-cross Abstract: LLM benchmarks are often treated as coherent measurement units, yet they are heterogeneous collections of instances spanning diverse domains, skills, formats, and…
- arXiv cs.AIresearchSimulating Student-Tutor Interactions with LLMs for Multilingual Math Feedback
arXiv:2506.04920v2 Announce Type: replace-cross Abstract: Large language models (LLMs) have demonstrated the ability to generate formative feedback and instructional hints in English, making them increasingly relevant fo…
- arXiv cs.AIresearchGUARD: Guided Unlearning and Retention via Data Attribution for Large Language Models
arXiv:2506.10946v3 Announce Type: replace-cross Abstract: Unlearning large language models (LLMs) is important due to copyright and privacy concerns. A key challenge is unintended forgetting, where removing data inadvert…
- arXiv cs.AIresearchFrom Empirical Evaluation to Context-Aware Enhancement: Repairing Regression Errors with LLMs
arXiv:2506.13182v4 Announce Type: replace-cross Abstract: [...] Since then, various APR approaches, especially those leveraging the power of large language models (LLMs), have been rapidly developed to fix general softwa…
- arXiv cs.AIagentsCo-Creative Learning via Metropolis-Hastings Interaction between Humans and AI
arXiv:2506.15468v2 Announce Type: replace-cross Abstract: We propose co-creative learning as a novel paradigm where humans and AI, i.e., biological and artificial agents, mutually integrate their partial perceptual infor…
- arXiv cs.AIresearchRCStat: A Statistical Framework for using Relative Contextualization in Transformers
arXiv:2506.19549v2 Announce Type: replace-cross Abstract: Identifying which tokens and activations truly influence model predictions is critical for both the efficiency and interpretability of large auto-regressive langu…
- arXiv cs.AIagentsCurriculum-Adapted Robust Reinforcement Learning for UAV Deconfliction in Adversarial Environments
arXiv:2506.21129v3 Announce Type: replace-cross Abstract: Autonomous unmanned aerial vehicles (UAVs) increasingly rely on reinforcement learning (RL) for navigation, yet global navigation satellite system (GNSS) spoofing…
- arXiv cs.AIresearchFrom Individuals to Interactions: Benchmarking Gender Bias in Multimodal Large Language Models from the Lens of Social Relationship
arXiv:2506.23101v2 Announce Type: replace-cross Abstract: Ensuring the trustworthiness of Multimodal Large Language Models (MLLMs) requires rigorous evaluation of gender bias, particularly in socially sensitive domains.…
- arXiv cs.AIresearchPhysics-Informed Graph Neural Networks to Reconstruct Local Fields Considering Finite Strain Hyperelasticity
arXiv:2507.05291v2 Announce Type: replace-cross Abstract: We propose a physics-informed machine learning framework called P-DivGNN to reconstruct local stress fields at the micro-scale, in the context of multi-scale simu…
- arXiv cs.AIresearchUnequal Voices: How LLMs Construct Constrained Queer Narratives
arXiv:2507.15585v2 Announce Type: replace-cross Abstract: One way social groups are marginalized in discourse is that the narratives told about them often default to a narrow, stereotyped range of topics. In contrast, de…
- arXiv cs.AIagentsWeb3 x AI Agents: Landscape, Integrations, and Foundational Challenges
arXiv:2508.02773v4 Announce Type: replace-cross Abstract: The convergence of Web3 technologies and AI agents is a rapidly evolving frontier that is reshaping decentralized ecosystems. This paper presents the first compre…
- arXiv cs.AIresearchThere are Levels to It: Red Teaming Dialogue With Hierarchical Reinforcement Learning
arXiv:2508.04451v2 Announce Type: replace-cross Abstract: Red teaming is essential for securing Large Language Models, yet current automated methods remain limited by templates and single-turn attacks. To simulate the co…
- arXiv cs.AIagentsRCR-Router: Efficient Role-Aware Context Routing for Multi-Agent LLM Systems with Structured Memory
arXiv:2508.04903v4 Announce Type: replace-cross Abstract: Large language model (LLM) agents are increasingly organized into multi-agent systems for complex reasoning and decision-making tasks. However, scaling such syste…
- arXiv cs.AIresearchOne Size Does Not Fit All: Distribution-Guided Tensor-Wise Adaptive Sparsification for Model Merging
arXiv:2508.06163v2 Announce Type: replace-cross Abstract: Merging multiple task-specific fine-tuned checkpoints into a single multi-task model provides an attractive alternative to joint multi-task training. To mitigate…
- arXiv cs.AIresearchSciRerankBench: Benchmarking Rerankers Towards Scientific Retrieval-Augmented Generation with LLMs
arXiv:2508.08742v3 Announce Type: replace-cross Abstract: Scientific literature question answering is a pivotal step towards new scientific discoveries. Recently, \textit{two-stage} retrieval-augmented generated large la…
- arXiv cs.AIresearchPPGFlowECG: Latent Rectified Flow with Cross-Modal Encoding for PPG-Guided ECG Generation and Cardiovascular Disease Detection
arXiv:2509.19774v3 Announce Type: replace-cross Abstract: Electrocardiography (ECG) is the clinical gold standard for cardiovascular disease (CVD) assessment, yet continuous monitoring is constrained by the need for dedi…
- arXiv cs.AIresearchTENET: One Step Toward Test-Driven Development for Repository-Level Code Generation
arXiv:2509.24148v5 Announce Type: replace-cross Abstract: Test-Driven Development (TDD) is a widely adopted practice that requires developers to create and execute tests alongside implementation. With recent advances in…
- arXiv cs.AIresearchCognifying Education: Mapping AI's transformative role in emotional, creative, and collaborative learning
arXiv:2509.25266v2 Announce Type: replace-cross Abstract: Artificial intelligence (AI) is rapidly reshaping educational practice, challenging long held assumptions about teaching and learning. This article integrates con…
- arXiv cs.AIresearchFrequency-Aware Model Parameter Explorer: A new attribution method for improving explainability
arXiv:2510.03245v3 Announce Type: replace-cross Abstract: State-of-the-art attribution methods such as AttEXplore rely on adversarial sample generation that applies an all-pass filter across the frequency spectrum: they…
- arXiv cs.AIresearchRethinking Langevin Thompson Sampling from A Stochastic Approximation Perspective
arXiv:2510.05023v2 Announce Type: replace-cross Abstract: Langevin-based Thompson Sampling (TS) methods for multi-armed bandits use likelihood gradients to approximate posterior samples when exact sampling is intractable…
- arXiv cs.AIresearchBatch Bayesian Active Learning with Partial Batch Label Sampling
arXiv:2510.09877v4 Announce Type: replace-cross Abstract: Over the past couple of decades, many active learning acquisition functions have been proposed, leaving practitioners with an unclear choice of which to use. Baye…
- arXiv cs.AIresearchA Function-Centric Perspective on Flat and Sharp Minima
arXiv:2510.12451v3 Announce Type: replace-cross Abstract: Flat minima are strongly associated with improved generalisation in deep neural networks. However, this connection has proven nuanced in recent studies, with both…
- arXiv cs.AIresearchSoK: Systematizing LLM Prompt Security: Taxonomies, Datasets, and Unified Evaluation of Attacks and Defenses
arXiv:2510.15476v5 Announce Type: replace-cross Abstract: Prompt-security claims for large language models (LLMs) are often compared as properties of individual attacks or defenses, although measured harmfulness depends…
- arXiv cs.AIresearchIterative Critique-Refine Framework for Enhancing LLM Personalization
arXiv:2510.24469v2 Announce Type: replace-cross Abstract: Personalized text generation requires models not only to produce coherent text but also to align with a target user's style, tone, and topical focus. Existing ret…
- arXiv cs.AIresearchTraining LLMs Beyond Next Token Prediction - Filling the Mutual Information Gap
arXiv:2511.00198v2 Announce Type: replace-cross Abstract: Autoregressive language models are usually trained with next-token prediction under a fixed left-to-right target order. This paper studies target-token organizati…
- arXiv cs.AIresearchEnergy-Efficient Deep Learning Without Backpropagation: A Rigorous Hardware-Validated Benchmarking Study of Forward-Only Algorithms
arXiv:2511.01061v2 Announce Type: replace-cross Abstract: Backpropagation (BP) powers modern deep learning, but it must hold every activation in memory until a global backward sweep consumes it, and that sweep locks each…
- arXiv cs.AIresearchExploring the Underwater World Segmentation without Extra Training
arXiv:2511.07923v3 Announce Type: replace-cross Abstract: Accurate segmentation of marine organisms is vital for biodiversity monitoring and ecological assessment, yet existing datasets and models remain largely limited…
- arXiv cs.AIresearchNaga: Vedic Encoding for Deep State Space Models
arXiv:2511.13510v2 Announce Type: replace-cross Abstract: This paper presents Naga, a deep State Space Model (SSM) encoding approach inspired by structural concepts from Vedic mathematics. The proposed method introduces…
- arXiv cs.AIresearchUnSAMv2: Self-Supervised Learning Enables Segment Anything at Any Granularity
arXiv:2511.13714v2 Announce Type: replace-cross Abstract: The Segment Anything Model (SAM) family has become a widely adopted vision foundation model, but its ability to control segmentation granularity remains limited.…
- arXiv cs.AIresearchNeural Evidence and Behavioural Welfare Economics: When Can Neuroeconomics Inform Policy?
arXiv:2511.19548v2 Announce Type: replace-cross Abstract: Neuroeconomics appears to offer welfare economics an evidential supplement, since neural data may help distinguish choices that express genuine preferences from c…
- arXiv cs.AIresearchExplainable Visual Anomaly Detection via Concept Bottleneck Models
arXiv:2511.20088v3 Announce Type: replace-cross Abstract: In recent years, VAD (Visual Anomaly Detection) has attracted significant attention due to its high interpretability, as it provides visual explanations by highli…
- arXiv cs.AIresearchSemantics as a Shield: Label Disguise Defense (LDD) against Prompt Injection in LLM Sentiment Classification
arXiv:2511.21752v3 Announce Type: replace-cross Abstract: Prompt injection can cause large language models (LLMs) to deviate from their intended tasks by following instructions embedded in untrusted input. We investigate…
- arXiv cs.AIresearchMeasuring all the noises of LLM Evals
arXiv:2512.21326v3 Announce Type: replace-cross Abstract: Applying well-established statistical methods effectively to LLM evals requires consideration of model correlations and prediction noises. Leading evals often ign…
- arXiv cs.AIresearchPhyAVBench: A Challenging Audio Physics-Sensitivity Benchmark for Physically Grounded Text-to-Audio-Video Generation
arXiv:2512.23994v5 Announce Type: replace-cross Abstract: Text-to-audio-video (T2AV) generation is central to applications such as filmmaking and world modeling. However, current models often fail to produce physically p…
- arXiv cs.AIresearchMechanistic Knobs in LLMs: Retrieving and Steering High-Order Semantic Features via Sparse Autoencoders
arXiv:2601.02978v3 Announce Type: replace-cross Abstract: Recent work in Mechanistic Interpretability (MI) has enabled the identification and intervention of internal features in Large Language Models (LLMs). However, a…
- arXiv cs.AIresearchHERO: Hierarchical Evidential Reasoning Optimization for Radiology Report Generation via Reason-then-Summarize
arXiv:2601.03321v5 Announce Type: replace-cross Abstract: Multimodal Large Language Models (MLLMs) have substantially advanced Radiology Report Generation (RRG), yet aligning them through reinforcement learning (RL) rema…
- arXiv cs.AIresearchECLIPTICA -- A Framework for Switchable LLM Alignment via CITA - Contrastive Instruction-Tuned Alignment
arXiv:2601.06157v2 Announce Type: replace-cross Abstract: Alignment in large language models (LLMs) is still largely static: after training, the policy is frozen. DPO, GRPO methods typically imprint one behavior into the…
- arXiv cs.AIresearchGeoMotionGPT: Geometry-Aligned Motion Understanding with Large Language Models
arXiv:2601.07632v5 Announce Type: replace-cross Abstract: Discrete motion tokenization has recently enabled Large Language Models (LLMs) to serve as versatile backbones for motion understanding and motion-language reason…
- arXiv cs.AIresearchHiking in the Wild: A Scalable Perceptive Parkour Framework for Humanoids
arXiv:2601.07718v2 Announce Type: replace-cross Abstract: Achieving robust humanoid locomotion in complex, unstructured environments requires transitioning from reactive proprioception to proactive perception. However, i…
- arXiv cs.AIresearchRepresentational Similarity Alignment of Figurative Language in Humans and Large Language Models
arXiv:2601.09041v2 Announce Type: replace-cross Abstract: Large language models (LLMs) often produce judgments similar to those of humans, but their interpretation of figurative language remains unclear. We compared huma…
- arXiv cs.AIresearchArchitecture Before the Formula: Training-Dependent Computational Semantics in Transformers
arXiv:2601.11618v4 Announce Type: replace-cross Abstract: An operation's input-output function does not, by itself, determine its computational role within a represented process. That role also depends on the processes p…
- arXiv cs.AIresearchLess is More: Label-Guided Efficient Summarization of Procedural and Instructional Videos
arXiv:2601.12243v3 Announce Type: replace-cross Abstract: Video summarization transforms long videos into concise representations that are easier to document and analyze, especially in high-stakes domains such as surgica…
- arXiv cs.AIresearchNeuroVLA: A Brain-inspired Embodied Intelligence for Fluid and Fast Reflexive Robotics Control
arXiv:2601.14628v2 Announce Type: replace-cross Abstract: Vision-language-action (VLA) models inherit powerful semantic representations from large language models, but carrying over their design priorities can leave the…
- arXiv cs.AIagentsNeuroFilter: Activation-Based Guardrails for Privacy-Conscious LLM Agents
arXiv:2601.14660v3 Announce Type: replace-cross Abstract: Agentic Large Language Models (LLMs) are models able to reason, plan, and execute tools over unstructured data. These abilities enable transformative applications…
- arXiv cs.AIresearchWhere is the multimodal goal post? On the Ability of Foundation Models to Recognize Highlights in Videos
arXiv:2601.16333v3 Announce Type: replace-cross Abstract: Foundation models are used for many real-world applications involving language understanding and generation from temporally-ordered multimodal events. In this wor…
- arXiv cs.AIresearchRetell, Reward, Repeat: Reinforcement Learning for Narrative Theory-Informed Story Retelling
arXiv:2601.17226v3 Announce Type: replace-cross Abstract: Counterfactual storytelling exposes LLM shortcomings in causal reasoning, where few-shot prompting and SFT fail to produce logical and rational narratives. This c…
- arXiv cs.AIresearchUni-RS: A Spatially Faithful Unified Understanding and Generation Model for Remote Sensing
arXiv:2601.17673v2 Announce Type: replace-cross Abstract: Unified remote sensing multimodal models exhibit a pronounced spatial reversal curse: Although they can accurately recognize and describe object locations in imag…
- arXiv cs.AIresearchPixSearch: Region-Grounded Retrieval for Knowledgeable Large Multimodal Models
arXiv:2601.19060v2 Announce Type: replace-cross Abstract: Multimodal Retrieval-Augmented Generation (MM-RAG) augments visual question answering (VQA) with external knowledge, but many systems rely on modular pipelines th…
- arXiv cs.AIresearchThe pretraining domain outweighs the training objective in setting the privacy-utility trade-off of differentially private medical image analysis
arXiv:2601.19618v3 Announce Type: replace-cross Abstract: Differential privacy limits how much each training image can influence a medical imaging model but lowers its diagnostic accuracy. Pretrained weights reduce this…