AI Research News Feeds for August 22nd, 2025

AI RESEARCH PAPERS & ACADEMIC SOURCES

• Net zero needs AI — five actions to realize its promise
• GraphVelo allows for accurate inference of multimodal velocities and molecular mechanisms for single cells
• Variable selection for minimum-variance portfolios
• Sampling by averaging: A multiscale approach to score estimation
• A Unified Framework for Inference with General Missingness Patterns and Machine Learning Imputation
• Multiply Robust Conformal Risk Control with Coarsened Data
• On Prior Distributions for Orthogonal Function Sequences
• Boundary Detection Algorithm Inspired by Locally Linear Embedding
• CaLiV: LiDAR-to-Vehicle Calibration of Arbitrary Sensor Setups
• Handle-based Mesh Deformation Guided By Vision Language Model
• Bidirectional Temporal Information Propagation for Moving Infrared Small Target Detection
• A Curated Dataset and Deep Learning Approach for Minor Dent Detection in Vehicles
• Aligning Moments in Time using Video Queries
• Enhancing Novel View Synthesis from extremely sparse views with SfM-free 3D Gaussian Splatting Framework
• MExECON: Multi-view Extended Explicit Clothed humans Optimized via Normal integration
• Task-Generalized Adaptive Cross-Domain Learning for Multimodal Image Fusion
• ExtraGS: Geometric-Aware Trajectory Extrapolation with Uncertainty-Guided Generative Priors
• Multi-Object Sketch Animation with Grouping and Motion Trajectory Priors
• D3FNet: A Differential Attention Fusion Network for Fine-Grained Road Structure Extraction in Remote Perception Systems
• High-Frequency First: A Two-Stage Approach for Improving Image INR
• Fast globally optimal Truncated Least Squares point cloud registration with fixed rotation axis
• Multi-perspective monitoring of wildlife and human activities from camera traps and drones with deep learning models
• When and What: Diffusion-Grounded VideoLLM with Entity Aware Segmentation for Long Video Understanding
• Weakly-Supervised Learning for Tree Instances Segmentation in Airborne Lidar Point Clouds
• MapKD: Unlocking Prior Knowledge with Cross-Modal Distillation for Efficient Online HD Map Construction
• CM2LoD3: Reconstructing LoD3 Building Models Using Semantic Conflict Maps
• LLM-empowered Dynamic Prompt Routing for Vision-Language Models Tuning under Long-Tailed Distributions
• WorldWeaver: Generating Long-Horizon Video Worlds via Rich Perception
• Fine-grained Multi-class Nuclei Segmentation with Molecular-empowered All-in-SAM Model
• Waver: Wave Your Way to Lifelike Video Generation
• ATLAS: Decoupling Skeletal and Shape Parameters for Expressive Parametric Human Modeling
• Visual Autoregressive Modeling for Instruction-Guided Image Editing
• CineScale: Free Lunch in High-Resolution Cinematic Visual Generation
• Scalable FPGA Framework for Real-Time Denoising in High-Throughput Imaging: A DRAM-Optimized Pipeline using High-Level Synthesis
• \textit{adder-viz}: Real-Time Visualization Software for Transcoding Event Video
• Scalable Event-Based Video Streaming for Machines with MoQ
• Zero-shot Volumetric CT Super-Resolution using 3D Gaussian Splatting with Upsampled 2D X-ray Projection Priors
• Pathology-Informed Latent Diffusion Model for Anomaly Detection in Lymph Node Metastasis
• Lang2Lift: A Framework for Language-Guided Pallet Detection and Pose Estimation Integrated in Autonomous Outdoor Forklift Operation
• On the Effectiveness of Graph Reordering for Accelerating Approximate Nearest Neighbor Search on GPU
• DoSReMC: Domain Shift Resilient Mammography Classification using Batch Normalization Adaptation
• Self-supervised physics-informed generative networks for phase retrieval from a single X-ray hologram
• Deep Equilibrium Convolutional Sparse Coding for Hyperspectral Image Denoising
• Hessian-based lightweight neural network for brain vessel segmentation on a minimal training dataset
• Translating Images to Road Network: A Sequence-to-Sequence Perspective
• RESfM: Robust Deep Equivariant Structure from Motion
• Learning Motion Blur Robust Vision Transformers for Real-Time UAV Tracking
• Vulnerabilities in AI-generated Image Detection: The Challenge of Adversarial Attacks
• TripleMixer: A 3D Point Cloud Denoising Model for Adverse Weather
• 3DGS-LM: Faster Gaussian-Splatting Optimization with Levenberg-Marquardt
• Hadamard Attention Recurrent Transformer: A Strong Baseline for Stereo Matching Transformer
• Cross multiscale vision transformer for deep fake detection
• BannerAgency: Advertising Banner Design with Multimodal LLM Agents
• Neuro Symbolic Knowledge Reasoning for Procedural Video Question Answering
• TrackID3x3: A Dataset and Algorithm for Multi-Player Tracking with Identification and Pose Estimation in 3x3 Basketball Full-court Videos
• Understanding Co-speech Gestures in-the-wild
• Omni$^2$: Unifying Omnidirectional Image Generation and Editing in an Omni Model
• FastMap: Revisiting Structure from Motion through First-Order Optimization
• Creating a Historical Migration Dataset from Finnish Church Records, 1800-1920
• Referring Expression Instance Retrieval and A Strong End-to-End Baseline
• Omni-Video: Democratizing Unified Video Understanding and Generation
• Capturing Stable HDR Videos Using a Dual-Camera System
• Hybrid Autoregressive-Diffusion Model for Real-Time Streaming Sign Language Production
• Fast-DDPM: Fast Denoising Diffusion Probabilistic Models for Medical Image-to-Image Generation
• Parallel transport on matrix manifolds and Exponential Action
• NucleiMix: Realistic Data Augmentation for Nuclei Instance Segmentation
• Physics-Driven Autoregressive State Space Models for Medical Image Reconstruction
• A Study of Privacy-preserving Language Modeling Approaches
• M-HELP: Using Social Media Data to Detect Mental Health Help-Seeking Signals
• Principle Methods of Rendering Non-equivalent Words from Uzbek and Dari to Russian and English
• PyTOD: Programmable Task-Oriented Dialogue with Execution Feedback
• SLM4Offer: Personalized Marketing Offer Generation Using Contrastive Learning Based Fine-Tuning
• SLM-Bench: A Comprehensive Benchmark of Small Language Models on Environmental Impacts -- Extended Version
• HebID: Detecting Social Identities in Hebrew-language Political Text
• Dream 7B: Diffusion Large Language Models
• The Enemy from Within: A Study of Political Delegitimization Discourse in Israeli Political Speech
• SafetyFlow: An Agent-Flow System for Automated LLM Safety Benchmarking
• SDGO: Self-Discrimination-Guided Optimization for Consistent Safety in Large Language Models
• Position Bias Mitigates Position Bias:Mitigate Position Bias Through Inter-Position Knowledge Distillation
• Stemming -- The Evolution and Current State with a Focus on Bangla
• Robust Symbolic Reasoning for Visual Narratives via Hierarchical and Semantically Normalized Knowledge Graphs
• LLMs and Agentic AI in Insurance Decision-Making: Opportunities and Challenges For Africa
• Retrieval-Augmented Review Generation for Poisoning Recommender Systems
• AmbiSQL: Interactive Ambiguity Detection and Resolution for Text-to-SQL
• Adversarial Attacks against Neural Ranking Models via In-Context Learning
• On the Role of Entity and Event Level Conceptualization in Generalizable Reasoning: A Survey of Tasks, Methods, Applications, and Future Directions
• Everybody Likes to Sleep: A Computer-Assisted Comparison of Object Naming Data from 30 Languages
• Pub-Guard-LLM: Detecting Retracted Biomedical Articles with Reliable Explanations
• Robust Bias Detection in MLMs and its Application to Human Trait Ratings
• Advancing the Database of Cross-Linguistic Colexifications with New Workflows and Data
• Leveraging Large Language Models for Explainable Activity Recognition in Smart Homes: A Critical Evaluation
• WebEvolver: Enhancing Web Agent Self-Improvement with Coevolving World Model
• One-shot Entropy Minimization
• AraReasoner: Evaluating Reasoning-Based LLMs for Arabic NLP
• SAND: Boosting LLM Agents with Self-Taught Action Deliberation
• Versatile Framework for Song Generation with Prompt-based Control
• You Only Pose Once: A Minimalist's Detection Transformer for Monocular RGB Category-level 9D Multi-Object Pose Estimation
• Paired-Sampling Contrastive Framework for Joint Physical-Digital Face Attack Detection
• GasTwinFormer: A Hybrid Vision Transformer for Livestock Methane Emission Segmentation and Dietary Classification in Optical Gas Imaging
• CurveFlow: Curvature-Guided Flow Matching for Image Generation
• HiRQA: Hierarchical Ranking and Quality Alignment for Opinion-Unaware Image Quality Assessment
• Reliable Multi-view 3D Reconstruction for `Just-in-time' Edge Environments
• XDR-LVLM: An Explainable Vision-Language Large Model for Diabetic Retinopathy Diagnosis
• MeSS: City Mesh-Guided Outdoor Scene Generation with Cross-View Consistent Diffusion
• Adversarial Agent Behavior Learning in Autonomous Driving Using Deep Reinforcement Learning
• DyMorph-B2I: Dynamic and Morphology-Guided Binary-to-Instance Segmentation for Renal Pathology
• STAGNet: A Spatio-Temporal Graph and LSTM Framework for Accident Anticipation
• Collaborative Multi-Modal Coding for High-Quality 3D Generation
• Center-Oriented Prototype Contrastive Clustering
• AeroDuo: Aerial Duo for UAV-based Vision and Language Navigation
• Comp-X: On Defining an Interactive Learned Image Compression Paradigm With Expert-driven LLM Agent
• Normal and Abnormal Pathology Knowledge-Augmented Vision-Language Model for Anomaly Detection in Pathology Images
• RATopo: Improving Lane Topology Reasoning via Redundancy Assignment
• TPA: Temporal Prompt Alignment for Fetal Congenital Heart Defect Classification
• BasketLiDAR: The First LiDAR-Camera Multimodal Dataset for Professional Basketball MOT
• RCDINO: Enhancing Radar-Camera 3D Object Detection with DINOv2 Semantic Features
• An Empirical Study on How Video-LLMs Answer Video Questions
• Transfer learning optimization based on evolutionary selective fine tuning
• DriveSplat: Decoupled Driving Scene Reconstruction with Geometry-enhanced Partitioned Neural Gaussians
• DIO: Refining Mutual Information and Causal Chain to Enhance Machine Abstract Reasoning Ability
• Spiking Variational Graph Representation Inference for Video Summarization
• From Linearity to Non-Linearity: How Masked Autoencoders Capture Spatial Correlations
• Attribution, Citation, and Quotation: A Survey of Evidence-based Text Generation with Large Language Models
• Preliminary Ranking of WMT25 General Machine Translation Systems
• Bridging the Culture Gap: A Framework for LLM-Driven Socio-Cultural Localization of Math Word Problems in Low-Resource Languages
• Improving LLMs for Machine Translation Using Synthetic Preference Data
• Multilingual Datasets for Custom Input Extraction and Explanation Requests Parsing in Conversational XAI Systems
• Reward-Shifted Speculative Sampling Is An Efficient Test-Time Weak-to-Strong Aligner
• Identifying and Answering Questions with False Assumptions: An Interpretable Approach
• ContextualLVLM-Agent: A Holistic Framework for Multi-Turn Visually-Grounded Dialogue and Complex Instruction Following
• Fin-PRM: A Domain-Specialized Process Reward Model for Financial Reasoning in Large Language Models
• Select to Know: An Internal-External Knowledge Self-Selection Framework for Domain-Specific Question Answering
• Self-Guided Function Calling in Large Language Models via Stepwise Experience Recall
• Are Checklists Really Useful for Automatic Evaluation of Generative Tasks?
• WangchanThaiInstruct: An instruction-following Dataset for Culture-Aware, Multitask, and Multi-domain Evaluation in Thai
• UniCoM: A Universal Code-Switching Speech Generator
• EMNLP: Educator-role Moral and Normative Large Language Models Profiling
• TComQA: Extracting Temporal Commonsense from Text
• KG-EDAS: A Meta-Metric Framework for Evaluating Knowledge Graph Completion Models
• A Survey on Large Language Model Benchmarks
• Confidence-Modulated Speculative Decoding for Large Language Models
• Tree-like Pairwise Interaction Networks
• Effect Identification and Unit Categorization in the Multi-Score Regression Discontinuity Design with Application to LED Manufacturing
• End-to-End Analysis of Charge Stability Diagrams with Transformers
• Exploring the Landscape of Non-Equilibrium Memories with Neural Cellular Automata
• Scaling Group Inference for Diverse and High-Quality Generation
• Robust Sparse Mean Estimation via Incremental Learning
• A mathematical perspective on Transformers
• Contextual Bandits with Stage-wise Constraints
• Wasserstein Distributionally Robust Shallow Convex Neural Networks
• Scalable Time-Series Causal Discovery with Approximate Causal Ordering
• MATATA: Weakly Supervised End-to-End MAthematical Tool-Augmented Reasoning for Tabular Applications
• The Complexity Dynamics of Grokking
• Faster Convergence of Riemannian Stochastic Gradient Descent with Increasing Batch Size
• Deceptive Sequential Decision-Making via Regularized Policy Optimization
• MaskSDM with Shapley values to improve flexibility, robustness, and explainability in species distribution modeling
• Pairwise or Pointwise? Evaluating Feedback Protocols for Bias in LLM-Based Evaluation
• Bayes Error Rate Estimation in Difficult Situations
• Multi-Exit Kolmogorov-Arnold Networks: enhancing accuracy and parsimony
• Exploring Modularity of Agentic Systems for Drug Discovery
• Physics-Informed Neural Networks with Hard Nonlinear Equality and Inequality Constraints
• Causal Modelling of Cryptocurrency Price Movements Using Discretisation-Aware Bayesian Networks
• Neural reproducing kernel Banach spaces and representer theorems for deep networks
• ILeSiA: Interactive Learning of Robot Situational Awareness from Camera Input
• Adaptive Routing of Text-to-Image Generation Requests Between Large Cloud Model and Light-Weight Edge Model
• Rethinking Addressing in Language Models via Contexualized Equivariant Positional Encoding
• Inverse Problem Sampling in Latent Space Using Sequential Monte Carlo
• ABC: Achieving Better Control of Multimodal Embeddings using VLMs
• Online Convex Optimization and Integral Quadratic Constraints: An automated approach to regret analysis
• Improving Predictions of Convective Storm Wind Gusts through Statistical Post-Processing of Neural Weather Models
• Training neural control variates using correlated configurations
• Machine Learning Approaches to Vocal Register Classification in Contemporary Male Pop Music
• EgoDex: Learning Dexterous Manipulation from Large-Scale Egocentric Video
• Scalable Bayesian Monte Carlo: fast uncertainty estimation beyond deep ensembles
• Large Language Models Encode Semantics in Low-Dimensional Linear Subspaces
• A Robust BERT-Based Deep Learning Model for Automated Cancer Type Extraction from Unstructured Pathology Reports
• SafeLLM: Unlearning Harmful Outputs from Large Language Models against Jailbreak Attacks
• Revisiting Pre-processing Group Fairness: A Modular Benchmarking Framework
• Frequency-adaptive tensor neural networks for high-dimensional multi-scale problems
• SleepDIFFormer: Sleep Stage Classification via Multivariate Differential Transformer
• See Beyond a Single View: Multi-Attribution Learning Leads to Better Conversion Rate Prediction
• Learning ECG Representations via Poly-Window Contrastive Learning
• Deep Think with Confidence
• Evaluating Knowledge Graph Complexity via Semantic, Spectral, and Structural Metrics for Link Prediction
• Saving for the future: Enhancing generalization via partial logic regularization
• ExBigBang: A Dynamic Approach for Explainable Persona Classification through Contextualized Hybrid Transformer Analysis
• Enhancing Forecasting with a 2D Time Series Approach for Cohort-Based Data
• Fairness for the People, by the People: Minority Collective Action
• CITE: A Comprehensive Benchmark for Heterogeneous Text-Attributed Graphs on Catalytic Materials
• Federated Learning based on Self-Evolving Gaussian Clustering
• Measures of Overlapping Multivariate Gaussian Clusters in Unsupervised Online Learning
• Mini-Batch Robustness Verification of Deep Neural Networks
• Learning Protein-Ligand Binding in Hyperbolic Space
• Let's Grow an Unbiased Community: Guiding the Fairness of Graphs via New Links
• Jointly Computation- and Communication-Efficient Distributed Learning
• Stabilization of Perturbed Loss Function: Differential Privacy without Gradient Noise
• AI-Powered Machine Learning Approaches for Fault Diagnosis in Industrial Pumps
• Conformalized Exceptional Model Mining: Telling Where Your Model Performs (Not) Well
• Inductive Domain Transfer In Misspecified Simulation-Based Inference
• Continual Neural Topic Model
• Classification errors distort findings in automated speech processing: examples and solutions from child-development research
• Correct-By-Construction: Certified Individual Fairness through Neural Network Training
• Amortized In-Context Mixed Effect Transformer Models: A Zero-Shot Approach for Pharmacokinetics
• Tensorized Multi-Task Learning for Personalized Modeling of Heterogeneous Individuals with High-Dimensional Data
• An Efficient Open World Environment for Multi-Agent Social Learning
• Conditionally adaptive augmented Lagrangian method for physics-informed learning of forward and inverse problems using artificial neural networks
• Investigation of D-Wave quantum annealing for training Restricted Boltzmann Machines and mitigating catastrophic forgetting
• Communication Efficient LLM Pre-training with SparseLoCo
• Probability Density from Latent Diffusion Models for Out-of-Distribution Detection
• Intern-S1: A Scientific Multimodal Foundation Model
• Distributed Detection of Adversarial Attacks in Multi-Agent Reinforcement Learning with Continuous Action Space
• Computational Resolution of Hadamard Product Factorization for $4 \times 4$ Matrices
• Closing the Performance Gap in Generative Recommenders with Collaborative Tokenization and Efficient Modeling
• Personalized Recommendations via Active Utility-based Pairwise Sampling
• Denoising by neural network for muzzle blast detection
• Human Feedback Driven Dynamic Speech Emotion Recognition
• MCPTox: A Benchmark for Tool Poisoning Attack on Real-World MCP Servers
• AGP: A Novel Arabidopsis thaliana Genomics-Phenomics Dataset and its HyperGraph Baseline Benchmarking
• XAI-Driven Spectral Analysis of Cough Sounds for Respiratory Disease Characterization
• Potential and challenges of generative adversarial networks for super-resolution in 4D Flow MRI
• CUTE-MRI: Conformalized Uncertainty-based framework for Time-adaptivE MRI
• Generative AI models enable efficient and physically consistent sea-ice simulations
• A Vision-Based Shared-Control Teleoperation Scheme for Controlling the Robotic Arm of a Four-Legged Robot
• Kernel-based Equalized Odds: A Quantification of Accuracy-Fairness Trade-off in Fair Representation Learning
• Adaptive Anomaly Detection in Evolving Network Environments
• Integrated Sensing, Communication, and Computation for Over-the-Air Federated Edge Learning
• GEN2: A Generative Prediction-Correction Framework for Long-time Emulations of Spatially-Resolved Climate Extremes
• Pretrained Diffusion Models Are Inherently Skipped-Step Samplers
• MMQ: Multimodal Mixture-of-Quantization Tokenization for Semantic ID Generation and User Behavioral Adaptation
• CUPE: Contextless Universal Phoneme Encoder for Language-Agnostic Speech Processing
• Flow Matching at Scale: A Machine Learning Framework for Efficient Large-Size Sampling of Many-Body Systems
• An Enhanced Audio Feature Tailored for Anomalous Sound Detection Based on Pre-trained Models
• Bayesian Inference and Learning in Nonlinear Dynamical Systems: A Framework for Incorporating Explicit and Implicit Prior Knowledge
• Exploiting Vocabulary Frequency Imbalance in Language Model Pre-training
• Foundational Design Principles and Patterns for Building Robust and Adaptive GenAI-Native Systems
• JEDI-linear: Fast and Efficient Graph Neural Networks for Jet Tagging on FPGAs
• Influence-driven Curriculum Learning for Pre-training on Limited Data
• High-dimensional Asymptotics of Generalization Performance in Continual Ridge Regression
• BadFU: Backdoor Federated Learning through Adversarial Machine Unlearning
• HEAS: Hierarchical Evolutionary Agent Simulation Framework for Cross-Scale Modeling and Multi-Objective Search
• Backpropagation-Free Test-Time Adaptation via Probabilistic Gaussian Alignment
• Exploiting Policy Idling for Dexterous Manipulation
• Bayesian Optimization with Expected Improvement: No Regret and the Choice of Incumbent
• Towards Reliable and Generalizable Differentially Private Machine Learning (Extended Version)
• Continual Learning for Multimodal Data Fusion of a Soft Gripper
• Grounded-VideoLLM: Sharpening Fine-grained Temporal Grounding in Video Large Language Models
• Teuken-7B-Base & Teuken-7B-Instruct: Towards European LLMs
• Fine-tuning foundational models to code diagnoses from veterinary health records
• Knowledge-Guided Prompt Learning for Request Quality Assurance in Public Code Review
• Evaluation Agent: Efficient and Promptable Evaluation Framework for Visual Generative Models
• Large Language Models for Automated Literature Review: An Evaluation of Reference Generation, Abstract Writing, and Review Composition
• Modeling Discrimination with Causal Abstraction
• Learning to Generate Unit Tests for Automated Debugging
• Self-Supervised Prompt Optimization
• RefineCoder: Iterative Improving of Large Language Models via Adaptive Critique Refinement for Code Generation
• Ontology-Guided Reverse Thinking Makes Large Language Models Stronger on Knowledge Graph Question Answering
• Innamark: A Whitespace Replacement Information-Hiding Method
• Synthetic vs. Gold: The Role of LLM Generated Labels and Data in Cyberbullying Detection
• Pragmatic Inference Chain (PIC) Improving LLMs' Reasoning of Authentic Implicit Toxic Language
• A Case for Specialisation in Non-Human Entities
• Revisiting Out-of-Distribution Detection in Real-time Object Detection: From Benchmark Pitfalls to a New Mitigation Paradigm
• VerifiAgent: a Unified Verification Agent in Language Model Reasoning
• TextSplat: Text-Guided Semantic Fusion for Generalizable Gaussian Splatting
• MuSeD: A Multimodal Spanish Dataset for Sexism Detection in Social Media Videos
• Kuwain 1.5B: An Arabic SLM via Language Injection
• Cequel: Cost-Effective Querying of Large Language Models for Text Clustering
• On the Consistency of GNN Explanations for Malware Detection
• Annif at SemEval-2025 Task 5: Traditional XMTC augmented by LLMs
• Sadeed: Advancing Arabic Diacritization Through Small Language Model
• MMiC: Mitigating Modality Incompleteness in Clustered Federated Learning
• Mutarjim: Advancing Bidirectional Arabic-English Translation with a Small Language Model
• Versatile Cardiovascular Signal Generation with a Unified Diffusion Transformer
• Lossless Token Sequence Compression via Meta-Tokens
• LaMP-Cap: Personalized Figure Caption Generation With Multimodal Figure Profiles
• Deep regularization networks for inverse problems with noisy operators
• A Survey of Foundation Models for IoT: Taxonomy and Criteria-Based Analysis
• Empirical Evidence for Alignment Faking in a Small LLM and Prompt-Based Mitigation Techniques
• KEA Explain: Explanations of Hallucinations using Graph Kernel Analysis
• MCA-RG: Enhancing LLMs with Medical Concept Alignment for Radiology Report Generation
• Cross-Modality Masked Learning for Survival Prediction in ICI Treated NSCLC Patients
• Generation of structure-guided pMHC-I libraries using Diffusion Models
• Cohort-Aware Agents for Individualized Lung Cancer Risk Prediction Using a Retrieval-Augmented Model Selection Framework
• Structure-Aware Temporal Modeling for Chronic Disease Progression Prediction
• HHNAS-AM: Hierarchical Hybrid Neural Architecture Search using Adaptive Mutation Policies
• Linear Preference Optimization: Decoupled Gradient Control via Absolute Regularization
• Large Foundation Model for Ads Recommendation
• CuMoLoS-MAE: A Masked Autoencoder for Remote Sensing Data Reconstruction
• Aura-CAPTCHA: A Reinforcement Learning and GAN-Enhanced Multi-Modal CAPTCHA System
• Generative Neural Operators of Log-Complexity Can Simultaneously Solve Infinitely Many Convex Programs
• TOAST: Fast and scalable auto-partitioning based on principled static analysis
• Fragment-Wise Interpretability in Graph Neural Networks via Molecule Decomposition and Contribution Analysis
• Nonlinear Federated System Identification
• Rethinking the Potential of Layer Freezing for Efficient DNN Training
• Robust Estimation Under Heterogeneous Corruption Rates
• Enhancing Optimizer Stability: Momentum Adaptation of The NGN Step-size
• Evaluating Sparse Autoencoders for Monosemantic Representation
• Side Effects of Erasing Concepts from Diffusion Models
• Towards Source-Free Machine Unlearning
• SurgWound-Bench: A Benchmark for Surgical Wound Diagnosis
• SemToken: Semantic-Aware Tokenization for Efficient Long-Context Language Modeling
• Survey of Vision-Language-Action Models for Embodied Manipulation
• SparK: Query-Aware Unstructured Sparsity with Recoverable KV Cache Channel Pruning
• Locally Pareto-Optimal Interpretations for Black-Box Machine Learning Models
• GenTune: Toward Traceable Prompts to Improve Controllability of Image Refinement in Environment Design
• VocabTailor: Dynamic Vocabulary Selection for Downstream Tasks in Small Language Models
• Robust and Efficient Quantum Reservoir Computing with Discrete Time Crystal
• Explainable Knowledge Distillation for Efficient Medical Image Classification
• Conflict-Aware Soft Prompting for Retrieval-Augmented Generation
• M-$LLM^3$REC: A Motivation-Aware User-Item Interaction Framework for Enhancing Recommendation Accuracy with LLMs
• Way to Build Native AI-driven 6G Air Interface: Principles, Roadmap, and Outlook
• DesignCLIP: Multimodal Learning with CLIP for Design Patent Understanding
• IPIGuard: A Novel Tool Dependency Graph-Based Defense Against Indirect Prompt Injection in LLM Agents
• First RAG, Second SEG: A Training-Free Paradigm for Camouflaged Object Detection
• VideoEraser: Concept Erasure in Text-to-Video Diffusion Models
• Predicting Road Crossing Behaviour using Pose Detection and Sequence Modelling
• Unveiling Trust in Multimodal Large Language Models: Evaluation, Analysis, and Mitigation
• Image-Conditioned 3D Gaussian Splat Quantization
• EvoFormer: Learning Dynamic Graph-Level Representations with Structural and Temporal Bias Correction
• Bladder Cancer Diagnosis with Deep Learning: A Multi-Task Framework and Online Platform
• Hybrid Least Squares/Gradient Descent Methods for DeepONets
• When Audio and Text Disagree: Revealing Text Bias in Large Audio-Language Models
• Bridging Generalization and Personalization in Wearable Human Activity Recognition via On-Device Few-Shot Learning
• LLaSO: A Foundational Framework for Reproducible Research in Large Language and Speech Model
• An Empirical Study of Knowledge Distillation for Code Understanding Tasks
• Test-time Corpus Feedback: From Retrieval to RAG
• Mitigating Hallucinations in LM-Based TTS Models via Distribution Alignment Using GFlowNets
• Reliable Unlearning Harmful Information in LLMs with Metamorphosis Representation Projection
• A Solvable Molecular Switch Model for Stable Temporal Information Processing
• RadReason: Radiology Report Evaluation Metric with Reasons and Sub-Scores
• Subjective Behaviors and Preferences in LLM: Language of Browsing
• LGMSNet: Thinning a medical image segmentation model via dual-level multiscale fusion
• LLM-Driven Self-Refinement for Embodied Drone Task Planning
• LoUQAL: Low-fidelity informed Uncertainty Quantification for Active Learning in the chemical configuration space
• Are Virtual DES Images a Valid Alternative to the Real Ones?
• Trained Miniatures: Low cost, High Efficacy SLMs for Sales & Marketing
• GRASPED: Graph Anomaly Detection using Autoencoder with Spectral Encoder and Decoder (Full Version)
• Label Uncertainty for Ultrasound Segmentation
• Towards a 3D Transfer-based Black-box Attack via Critical Feature Guidance
• Benchmarking Computer Science Survey Generation
• Mind and Motion Aligned: A Joint Evaluation IsaacSim Benchmark for Task Planning and Low-Level Policies in Mobile Manipulation
• Row-Column Hybrid Grouping for Fault-Resilient Multi-Bit Weight Representation on IMC Arrays
• Foundation Models for Cross-Domain EEG Analysis Application: A Survey
• StreamMem: Query-Agnostic KV Cache Memory for Streaming Video Understanding
• Tutorial on the Probabilistic Unification of Estimation Theory, Machine Learning, and Generative AI
• EcomMMMU: Strategic Utilization of Visuals for Robust Multimodal E-Commerce Models
• Numerical models outperform AI weather forecasts of record-breaking extremes
• End-to-End Agentic RAG System Training for Traceable Diagnostic Reasoning
• "Does the cafe entrance look accessible? Where is the door?" Towards Geospatial AI Agents for Visual Inquiries
• Dissecting Tool-Integrated Reasoning: An Empirical Study and Analysis
• Neural Robot Dynamics
• LiveMCP-101: Stress Testing and Diagnosing MCP-enabled Agents on Challenging Queries
• Discovering Hidden Algebraic Structures via Transformers with Rank-Aware Beam GRPO
• SceneGen: Single-Image 3D Scene Generation in One Feedforward Pass
• CRISPR-GPT for Agentic Automation of Gene-editing Experiments
• Non-linear Welfare-Aware Strategic Learning
• Human-Object Interaction from Human-Level Instructions
• On Learning Action Costs from Input Plans
• Exploring the Effect of Explanation Content and Format on User Comprehension and Trust in Healthcare
• VLASCD: A Visual Language Action Model for Simultaneous Chatting and Decision Making
• CopyrightShield: Enhancing Diffusion Model Security against Copyright Infringement Attacks
• SycEval: Evaluating LLM Sycophancy
• PersonaBench: Evaluating AI Models on Understanding Personal Information through Accessing (Synthetic) Private User Data
• Automatic Curriculum Design for Zero-Shot Human-AI Coordination
• GATES: Cost-aware Dynamic Workflow Scheduling via Graph Attention Networks and Evolution Strategy
• Exploring Big Five Personality and AI Capability Effects in LLM-Simulated Negotiation Dialogues
• Using a cognitive architecture to consider antiBlackness in design and development of AI systems
• Unplug and Play Language Models: Decomposing Experts in Language Models at Inference Time
• Generating 3D Terrain with 2D Cellular Automata
• CREMA: A Contrastive Regularized Masked Autoencoder for Robust ECG Diagnostics across Clinical Domains
• OPDR: Order-Preserving Dimension Reduction for Semantic Embedding of Multimodal Scientific Data
• BoostTrack++: using tracklet information to detect more objects in multiple object tracking
• A Fully Spectral Neuro-Symbolic Reasoning Architecture with Graph Signal Processing as the Computational Backbone
• Goals and the Structure of Experience
• Collab-REC: An LLM-based Agentic Framework for Balancing Recommendations in Tourism
• Emergent Crowds Dynamics from Language-Driven Multi-Agent Interactions
• Don't Think Twice! Over-Reasoning Impairs Confidence Calibration
• Demonstrating Onboard Inference for Earth Science Applications with Spectral Analysis Algorithms and Deep Learning
• S3LoRA: Safe Spectral Sharpness-Guided Pruning in Adaptation of Agent Planner
• Argumentation for Explainable Workforce Optimisation (with Appendix)
• Open-Universe Assistance Games
• aiXiv: A Next-Generation Open Access Ecosystem for Scientific Discovery Generated by AI Scientists
• Mobile-Agent-v3: Foundamental Agents for GUI Automation
• PuzzleClone: An SMT-Powered Framework for Synthesizing Verifiable Data
• LLM4Sweat: A Trustworthy Large Language Model for Hyperhidrosis Support
• R-ConstraintBench: Evaluating LLMs on NP-Complete Scheduling
• See it. Say it. Sorted: Agentic System for Compositional Diagram Generation
• Computational Intelligence based Land-use Allocation Approaches for Mixed Use Areas
• Multiple Memory Systems for Enhancing the Long-term Memory of Agent
• Coarse-to-Fine Grounded Memory for LLM Agent Planning
• Search-Based Credit Assignment for Offline Preference-Based Reinforcement Learning
• RETAIL: Towards Real-world Travel Planning for Large Language Models
• DiagECG: An LLM-Driven Framework for Diagnostic Reasoning via Discretized ECG Tokenization
• Planning with Minimal Disruption
• GraSP: A Unified Graph-Based Framework for Scalable Generation, Quality Tagging, and Management of Synthetic Data for SFT and DPO
• From Bits to Boardrooms: A Cutting-Edge Multi-Agent LLM Framework for Business Excellence
• Think in Blocks: Adaptive Reasoning from Direct Response to Deep Reasoning
• Super-additive Cooperation in Language Model Agents
• DeepThink3D: Enhancing Large Language Models with Programmatic Reasoning in Complex 3D Situated Reasoning Tasks
• A Dynamical Systems Framework for Reinforcement Learning Safety and Robustness Verification
• Transduction is All You Need for Structured Data Workflows
• Adapting A Vector-Symbolic Memory for Lisp ACT-R
• Understanding Action Effects through Instrumental Empowerment in Multi-Agent Reinforcement Learning
• Futurity as Infrastructure: A Techno-Philosophical Interpretation of the AI Lifecycle
• GRAFT: GRaPH and Table Reasoning for Textual Alignment -- A Benchmark for Structured Instruction Following and Visual Reasoning
• NiceWebRL: a Python library for human subject experiments with reinforcement learning environments
• Measuring the environmental impact of delivering AI at Google Scale
• Response and Prompt Evaluation to Prevent Parasocial Relationships with Chatbots
• Language-Guided Tuning: Enhancing Numeric Optimization with Textual Feedback
• SVM/SVR Kernels as Quantum Propagators
• Accelerating GenAI Workloads by Enabling RISC-V Microkernel Support in IREE
• Efficient Switchable Safety Control in LLMs via Magic-Token-Guided Co-Training
• Privacy Preserving Inference of Personalized Content for Out of Matrix Users
• Collaborative Filtering using Variational Quantum Hopfield Associative Memory
• A Chinese Heart Failure Status Speech Database with Universal and Personalised Classification
• Transsion Multilingual Speech Recognition System for MLC-SLM 2025 Challenge
• Disentangling the Drivers of LLM Social Conformity: An Uncertainty-Moderated Dual-Process Mechanism
• Designing an Interdisciplinary Artificial Intelligence Curriculum for Engineering: Evaluation and Insights from Experts
• Fusing Structural Phenotypes with Functional Data for Early Prediction of Primary Angle Closure Glaucoma Progression
• A U-Statistic-based random forest approach for genetic interaction study
• Learning to Drive Ethically: Embedding Moral Reasoning into Autonomous Driving
• AI Testing Should Account for Sophisticated Strategic Behaviour
• Heatmap Regression without Soft-Argmax for Facial Landmark Detection
• TOM: An Open-Source Tongue Segmentation Method with Multi-Teacher Distillation and Task-Specific Data Augmentation
• Inference Time Debiasing Concepts in Diffusion Models
• Can synthetic data reproduce real-world findings in epidemiology? A replication study using tree-based generative AI
• Quantum Long Short-term Memory with Differentiable Architecture Search
• Fast Graph Neural Network for Image Classification
• Quantized Neural Networks for Microcontrollers: A Comprehensive Review of Methods, Platforms, and Applications
• Twin-Boot: Uncertainty-Aware Optimization via Online Two-Sample Bootstrapping
• TAIGen: Training-Free Adversarial Image Generation via Diffusion Models
• Reversible Unfolding Network for Concealed Visual Perception with Generative Refinement
• A Systematic Survey of Model Extraction Attacks and Defenses: State-of-the-Art and Perspectives
• MoEcho: Exploiting Side-Channel Attacks to Compromise User Privacy in Mixture-of-Experts LLMs
• Decentralized Vision-Based Autonomous Aerial Wildlife Monitoring
• From Basic Affordances to Symbolic Thought: A Computational Phylogenesis of Biological Intelligence
• LongRecall: A Structured Approach for Robust Recall Evaluation in Long-Form Text
• Wormhole Dynamics in Deep Neural Networks
• Mapping the Course for Prompt-based Structured Prediction
• Nemotron-CC-Math: A 133 Billion-Token-Scale High Quality Math Pretraining Dataset
• Hydra: A 1.6B-Parameter State-Space Language Model with Sparse Attention, Mixture-of-Experts, and Memory
• Equi-mRNA: Protein Translation Equivariant Encoding for mRNA Language Models
• Enhanced Predictive Modeling for Hazardous Near-Earth Object Detection: A Comparative Analysis of Advanced Resampling Strategies and Machine Learning Algorithms in Planetary Risk Assessment
• Universal Reinforcement Learning in Coalgebras: Asynchronous Stochastic Computation via Conduction