AI news archive.
10,659 articles filed under Research, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchWatermarking Diffusion Language Models
arXiv:2509.24368v3 Announce Type: replace-cross Abstract: We introduce the first watermark tailored for diffusion language models (DLMs), an emergent LLM paradigm able to generate tokens in arbitrary order, in contrast t…
- arXiv cs.AIresearchSGM: A Statistical Godel Machine for Risk-Controlled Recursive Self-Modification
arXiv:2510.10232v2 Announce Type: replace-cross Abstract: Recursive self-modification is increasingly central in AutoML, neural architecture search, and adaptive optimization, yet no existing framework ensures that such…
- arXiv cs.AIresearchLearn2Drive: A neural network-based framework for socially compliant automated vehicle control
arXiv:2510.21736v2 Announce Type: replace-cross Abstract: This study introduces a novel control framework for adaptive cruise control (ACC) in automated driving, leveraging neural networks and physics-informed constraint…
- arXiv cs.AIresearchPre-train to Gain: Robust Learning Without Clean Labels
arXiv:2511.20844v2 Announce Type: replace-cross Abstract: Training deep networks with noisy labels leads to poor generalization and degraded accuracy due to overfitting to label noise. Existing approaches for learning wi…
- arXiv cs.AIresearchEffective and Efficient Threat Hunting with Small Language Models
arXiv:2512.06660v3 Announce Type: replace-cross Abstract: Analysts in Security Operations Centers query massive telemetry streams using Kusto Query Language (KQL), but writing correct KQL demands specialized expertise th…
- arXiv cs.AIresearchFedVideoMAE: Efficient Federated Video Moderation with Differential Privacy and Secure Aggregation
arXiv:2512.18809v3 Announce Type: replace-cross Abstract: Short-form video moderation is increasingly pushed toward edge and privacy-sensitive settings, where users may intend videos for a limited audience, such as frien…
- arXiv cs.AIresearchKinship Data Benchmark for Multi-hop Reasoning
arXiv:2601.07794v3 Announce Type: replace-cross Abstract: Multi-hop kinship reasoning is a natural testbed for LLM compositionality, but existing benchmarks (notably CLUTRR) cover only the descriptive Eskimo system. We i…
- arXiv cs.AIresearchCoMa: Contextual Massing Generation with Vision-Language Models
arXiv:2601.08464v2 Announce Type: replace-cross Abstract: Context-aware building massing is an important early-stage design task: given a site for buildings, a generated massing should not only fit the target parcel, but…
- arXiv cs.AIresearchAmbient Dataloops: Generative Models for Dataset Refinement
arXiv:2601.15417v2 Announce Type: replace-cross Abstract: We propose Ambient Dataloops, an iterative framework for refining datasets that makes it easier for diffusion models to learn the underlying data distribution. Mo…
- arXiv cs.AIresearchSim-and-Human Co-training for Data-Efficient and Scene-Generalizable Bimanual Manipulation
arXiv:2601.19406v2 Announce Type: replace-cross Abstract: Real-robot demonstrations are prohibitively expensive, while simulation data and real-world human demonstrations are both scalable but each leaves a distinct gap:…
- arXiv cs.AIresearchL2R: Low-Rank and Lipschitz-Controlled Routing for Mixture-of-Experts
arXiv:2601.21349v3 Announce Type: replace-cross Abstract: Mixture-of-Experts (MoE) models scale neural networks by conditionally activating a small subset of experts, where the router plays a central role in determining…
- arXiv cs.AIresearchWhy $\beta_1 = \beta_2$ Is Dynamically Special in Adam
arXiv:2601.21739v3 Announce Type: replace-cross Abstract: Adam has been at the core of large-scale training for almost a decade, yet the role of its two momentum parameters remains poorly understood. Recent work shows th…
- arXiv cs.AIresearchRethinking the Design Space of Reinforcement Learning for Diffusion Models: On the Importance of Likelihood Estimation Beyond Loss Design
arXiv:2602.04663v3 Announce Type: replace-cross Abstract: Reinforcement learning has been widely applied to diffusion and flow models for visual tasks such as text-to-image generation. However, these tasks remain challen…
- arXiv cs.AIresearchPerturbing the Phase: Analyzing Adversarial Robustness of Complex-Valued Neural Networks
arXiv:2602.06577v2 Announce Type: replace-cross Abstract: Complex-valued neural networks (CVNNs) are rising in popularity for all kinds of applications. To safely use CVNNs in practice, analyzing their robustness against…
- arXiv cs.AIresearchExploring Sparsity and Smoothness of Arbitrary Lp Norms in Adversarial Attacks
arXiv:2602.06578v2 Announce Type: replace-cross Abstract: Adversarial attacks against deep neural networks are commonly constructed under $\ell_p$ norm constraints, most often using $p=1$, $p=2$ or $p=\infty$, and potent…
- arXiv cs.AIresearchCross-Sectional Asset Retrieval via Future-Aligned Soft Contrastive Learning
arXiv:2602.10711v2 Announce Type: replace-cross Abstract: Asset retrieval (finding similar assets in a financial universe) is central to quantitative investment decision-making. Existing approaches define similarity thro…
- arXiv cs.AIresearchHigh-Resolution Range Profile Classifiers Require Aspect-Angle Awareness
arXiv:2603.00087v2 Announce Type: replace-cross Abstract: We revisit High-Resolution Range Profile (HRRP) classification with aspect-angle conditioning. While prior work often assumes that aspect-angle information is inc…
- arXiv cs.AIresearchTTSR: Test-Time Self-Evolving via Reflection
arXiv:2603.03297v3 Announce Type: replace-cross Abstract: Test-time training (TTT) adapts large language models (LLMs) during inference using only unlabeled test inputs. Existing methods, however, face two major bottlene…
- arXiv cs.AIresearchDrag-Aware Aerodynamic Manipulability for Torque-Limited Redundant Multirotors: Aerodynamic Promptness based on the Symmetric Acceleration Capacity
arXiv:2603.07998v2 Announce Type: replace-cross Abstract: Aerodynamic promptness quantifies how rotor-speed variations generate multirotor wrench variations, but its Euclidean formulation assigns the same local cost to a…
- arXiv cs.AIresearchMAPLE: Metadata Augmented Private Language Evolution
arXiv:2603.19258v3 Announce Type: replace-cross Abstract: Differentially private (DP) fine-tuning of large language models (LLMs) requires massive compute and full model access, which rules out state-of-the-art proprieta…
- arXiv cs.AIresearchDomain Elastic Transform: Bayesian Function Registration for High-Dimensional Scientific Data
arXiv:2603.21235v3 Announce Type: replace-cross Abstract: Nonrigid registration is conventionally divided into point set registration, which aligns sparse geometries, and image registration, which aligns continuous inten…
- arXiv cs.AIresearchWhen Consistency Becomes Bias: Interviewer Effects in Semi-Structured Clinical Interviews
arXiv:2603.24651v2 Announce Type: replace-cross Abstract: Automatic depression detection from doctor-patient conversations has gained momentum thanks to the availability of public corpora and advances in language modelin…
- arXiv cs.AIresearchWhen Perplexity Lies: Generation-Focused Distillation of Hybrid Sequence Models
arXiv:2603.26556v3 Announce Type: replace-cross Abstract: Converting a pretrained Transformer into a more efficient hybrid model through distillation offers a promising approach to reducing inference costs. However, achi…
- arXiv cs.AIresearchGreen-ELM: Efficient Analytic Learning via High-Dimensional Random Projections
arXiv:2604.15613v4 Announce Type: replace-cross Abstract: We present Green-ELM, a non-iterative neural architecture that replaces gradient-based optimization of the output layer with a closed-form analytic solution over…
- arXiv cs.AIresearchA Two-Stage Multi-Modal MRI Framework for Lifespan Brain Age Prediction
arXiv:2604.16655v2 Announce Type: replace-cross Abstract: The accurate quantification of brain age from MRI has emerged as an important biomarker of brain health. However, existing approaches are often restricted to narr…
- arXiv cs.AIresearchLarge language models eroding science understanding: an empirical study of malignment
arXiv:2604.25639v2 Announce Type: replace-cross Abstract: This paper is accepted and in press for AI and Ethics. This paper includes the supplementary data file at the end of the manuscript. This study examines whether l…
- arXiv cs.AIresearchLearning to Theorize the World from Observation
arXiv:2605.03413v4 Announce Type: replace-cross Abstract: What does it mean to understand the world? Contemporary world models often operationalize understanding as accurate future prediction in latent or observation spa…
- arXiv cs.AIresearchClin-JEPA: A Multi-Phase Co-Training Framework for Joint-Embedding Predictive Pretraining on EHR Patient Trajectories
arXiv:2605.10840v5 Announce Type: replace-cross Abstract: Joint-embedding predictive architectures (JEPA) learn representations by predicting in latent space, as in computer vision; retaining the action-conditioned predi…
- arXiv cs.AIresearchReinforcement Learning for Graph Generation under a Hard Assortativity Constraint
arXiv:2605.23285v2 Announce Type: replace-cross Abstract: Generating graph ensembles with precisely controlled structural properties is central to investigating how network structure shapes function. Canonical ensembles…
- arXiv cs.AIresearchBatch Normalization Amplifies Memorization and Privacy Risks
arXiv:2605.24420v2 Announce Type: replace-cross Abstract: Batch Normalization (BN) is widely adopted to enable faster convergence and more stable training of deep neural networks. However, its impact on privacy and memor…
- arXiv cs.AIresearchBy Their Fruits You Will Know Them: Comparing Formalizations of Law by the Decisions They Encode
arXiv:2605.25186v2 Announce Type: replace-cross Abstract: Formalizing legal provisions promises machine-accessible law and automated legal reasoning, and recent LLMs make it tempting to generate such formalizations direc…
- arXiv cs.AIresearchMitigating Stethoscope-Induced Shortcuts in Respiratory Sound Classification under Federated Domain Generalization with Causality-Inspired Interventions
arXiv:2605.29862v2 Announce Type: replace-cross Abstract: AI-driven respiratory sound classification (RSC) is promising for automated pulmonary disease detection, yet multi-site deployment is hindered by inter-stethoscop…
- arXiv cs.AIresearchAnyAudio-Judge: A Dynamic Rubric-Based Benchmark and Evaluator for Audio Instruction Following
arXiv:2606.03116v2 Announce Type: replace-cross Abstract: The rapid advancement of instruction-guided audio generation has highlighted the critical need for robust alignment evaluation. Current automated evaluation metho…
- arXiv cs.AIresearchEssentialGIN: a new approach for gene essentiality prediction based on graph isomorphism neural networks
arXiv:2606.07700v2 Announce Type: replace-cross Abstract: Background: Prediction of essential genes (proteins), is a basic and challenging problem but at the same time very costly and time-consuming in wet-lab experiment…
- arXiv cs.AIresearchExploring a Layer-Wise Design Space for KV Cache Eviction
arXiv:2606.15157v2 Announce Type: replace-cross Abstract: KV cache eviction methods typically use a single retention-rule family throughout a model, making eviction-method identity a model-level design choice. Yet Transf…
- arXiv cs.AIresearchRedact or Keep? A Fully Local AI Cascade for Educational Dialogue De-Identification
arXiv:2606.18372v2 Announce Type: replace-cross Abstract: Educational dialogue is a valuable but sensitive resource for research: the same transcripts that capture authentic learning often capture personally identifiable…
- arXiv cs.AIresearchScaling Audio Models Efficiently: Joint Optimization of Scale, Resolution, Adaptation, Precision, and Sparsity
arXiv:2606.22790v5 Announce Type: replace-cross Abstract: Large automatic speech recognition (ASR) models such as Whisper must be deployed across hardware with widely varying memory and inference-speed constraints. We pr…
- arXiv cs.AIresearchAccelerating Q-learning through Efficient Value-Sharing across Actions
arXiv:2606.29806v3 Announce Type: replace-cross Abstract: Action values are foundational to many control algorithms such as Q-learning. Therefore, efficient action-value learning is central to reinforcement learning (RL)…
- arXiv cs.AIresearchWorldRoamBench: An Open-World Benchmark for Long-Horizon Stability of Interactive World Models
arXiv:2606.31672v5 Announce Type: replace-cross Abstract: Despite rapid progress in interactive world models (IWMs), short-horizon performance does not establish sustained action following, visual stability, physical pla…
- arXiv cs.AIresearchPrompt-Driven Exploration: Language as an Exploration Space for VLA Reinforcement Learning
arXiv:2607.08837v4 Announce Type: replace-cross Abstract: Exploration is essential to RL since a policy cannot improve by repeatedly sampling the behaviors it already prefers. Standard methods inject stochasticity in the…
- arXiv cs.AIresearchWhen Data Imbalance Helps: Robust Generalization Through Shortcut Saturation
arXiv:2607.10116v2 Announce Type: replace-cross Abstract: We study robust generalization under spurious correlations: tasks where a shortcut feature is correlated with the true label in training but anti-correlated in an…
- arXiv cs.AIresearchLimits of Reliability and Scaling in Language Models
arXiv:2607.14112v3 Announce Type: replace-cross Abstract: Large language models (LLMs) are trained and evaluated as though perfect reliability is achievable for any task given sufficient scale. We show that this assumpti…
- arXiv cs.AIresearchMulti-Axis Max@K Reinforcement Learning for Representative Diversity in Text-to-Image Generation
arXiv:2607.14962v2 Announce Type: replace-cross Abstract: Text-to-image (T2I) models can synthesize realistic, prompt-aligned images, yet samples generated for the same prompt often cover only a small subset of visually…
- arXiv cs.AIresearchBuilding a Neural Network from Scratch: Implementation, Evaluation, and Optimization
arXiv:2607.16682v2 Announce Type: replace-cross Abstract: The widespread adoption of high-level deep learning libraries, while accelerating model development, has increasingly abstracted away the internal mechanics of ne…
- arXiv cs.AIresearchEvaluating Large Language Models for Symbolic Security Protocol Analysis
arXiv:2607.20712v2 Announce Type: replace-cross Abstract: Security protocols verification relies on formal tools such as ProVerif and OFMC. This study evaluates whether large language models (LLMs) can perform comparable…
- arXiv cs.AIresearchMyMentorLLM: A psychotherapy GenAI environment with multimodal voice/text patients, trainees and experts for deliberate practice
arXiv:2607.25667v2 Announce Type: replace-cross Abstract: Psychotherapists need repeated training and supervision; however, scalability is problematic. We present MyMentorLLM, a multimodal voice- and text-based deliberat…
- arXiv cs.AIresearchKeep It Simple: Multi-Key Episodic Memory Retrieval for Ultra-Long Video Understanding
arXiv:2608.07663v2 Announce Type: replace-cross Abstract: When videos extend from hours to days, directly processing them end-to-end becomes impractical for current Multi-modal Large Language Models (MLLMs). This ultra-l…
- arXiv cs.AIresearchGigaBrain-WBC-0.5: A Behavior World Model for Robust Humanoid Whole-Body Tracking with Environment Interaction
arXiv:2608.18234v4 Announce Type: replace-cross Abstract: General-purpose motion trackers enable humanoid robots to follow diverse whole-body motions while maintaining balance, but are trained only on flat ground, failin…
- arXiv cs.AIresearchModeling Human Behavior with Type Vectors Using AI
arXiv:2608.18265v4 Announce Type: replace-cross Abstract: We introduce a general, easy-to-implement AI-based modeling technique for analyzing human behavior. A key feature of this approach, which contrasts with existing…
- arXiv cs.AIresearchPonderPounce: A Pretrained MLLM as an Episode Context Engine for Robot Control
arXiv:2608.24115v2 Announce Type: replace-cross Abstract: Multimodal large language models (MLLMs) can integrate long visual histories and infer behavior from a few examples, yet vision-language-action models rarely use…