AI Research News Feeds for August 25th, 2025

AI RESEARCH PAPERS & ACADEMIC SOURCES

• Bring Your Rear Cameras for Egocentric 3D Human Pose Estimation
• MambaIC: State Space Models for High-Performance Learned Image Compression
• Adaptive Multi-Order Graph Regularized NMF with Dual Sparsity for Hyperspectral Unmixing
• EHGCN: Hierarchical Euclidean-Hyperbolic Fusion via Motion-Aware GCN for Hybrid Event Stream Perception
• Highly Accurate and Diverse Traffic Data: The DeepScenario Open 3D Dataset
• VIBE: Video-to-Text Information Bottleneck Evaluation for TL;DR
• Not Only Consistency: Enhance Test-Time Adaptation with Spatio-temporal Inconsistency for Remote Physiological Measurement
• Improving U-Net Confidence on TEM Image Data with L2-Regularization, Transfer Learning, and Deep Fine-Tuning
• Mean-Field Generalisation Bounds for Learning Controls in Stochastic Environments
• Wavelet-Enhanced PaDiM for Industrial Anomaly Detection
• Expandable Residual Approximation for Knowledge Distillation
• Advances and Trends in the 3D Reconstruction of the Shape and Motion of Animals
• A Unified Voxel Diffusion Module for Point Cloud 3D Object Detection
• Ensemble learning of foundation models for precision oncology
• 4D Virtual Imaging Platform for Dynamic Joint Assessment via Uni-Plane X-ray and 2D-3D Registration
• High-Precision Mixed Feature Fusion Network Using Hypergraph Computation for Cervical Abnormal Cell Detection
• RAGSR: Regional Attention Guided Diffusion for Image Super-Resolution
• FTIO: Frequent Temporally Integrated Objects
• \textsc{T-Mask}: Temporal Masking for Probing Foundation Models across Camera Views in Driver Monitoring
• Forecast then Calibrate: Feature Caching as ODE for Efficient Diffusion Transformers
• MedOmni-45{\deg}: A Safety-Performance Benchmark for Reasoning-Oriented LLMs in Medicine
• PromptFlare: Prompt-Generalized Defense via Cross-Attention Decoy in Diffusion-Based Inpainting
• UniEM-3M: A Universal Electron Micrograph Dataset for Microstructural Segmentation and Generation
• IRSAMap:Towards Large-Scale, High-Resolution Land Cover Map Vectorization
• Robust Small Methane Plume Segmentation in Satellite Imagery
• EdgeDoc: Hybrid CNN-Transformer Model for Accurate Forgery Detection and Localization in ID Documents
• Learning Long-Range Action Representation by Two-Stream Mamba Pyramid Network for Figure Skating Assessment
• Enhanced Hybrid Technique for Efficient Digitization of Handwritten Marksheets
• Vision encoders should be image size agnostic and task driven
• Attention Mechanism in Randomized Time Warping
• SAMFusion: Sensor-Adaptive Multimodal Fusion for 3D Object Detection in Adverse Weather
• HAMSt3R: Human-Aware Multi-view Stereo 3D Reconstruction
• Arbitrary-Scale 3D Gaussian Super-Resolution
• Seeing Clearly, Forgetting Deeply: Revisiting Fine-Tuned Video Generators for Driving Simulation
• Harmonious Color Pairings: Insights from Human Preference and Natural Hue Statistics
• Robust Residual Finite Scalar Quantization for Neural Compression
• UnPose: Uncertainty-Guided Diffusion Priors for Zero-Shot Pose Estimation
• GUI Based Fuzzy Logic and Spatial Statistics for Unsupervised Microscopy Segmentation
• GelSLAM: A Real-time, High-Fidelity, and Robust 3D Tactile SLAM System
• Clinically-Informed Preprocessing Improves Stroke Segmentation in Low-Resource Settings
• Wavelet-Space Super-Resolution for Real-Time Rendering
• Prompting with Sign Parameters for Low-resource Sign Language Instruction Generation
• Lightweight and Fast Real-time Image Enhancement via Decomposition of the Spatial-aware Lookup Tables
• Self-Validated Learning for Particle Separation: A Correctness-Based Self-Training Framework Without Human Labels
• Towards Diagnostic Quality Flat-Panel Detector CT Imaging Using Diffusion Models
• NeuroKoop: Neural Koopman Fusion of Structural-Functional Connectomes for Identifying Prenatal Drug Exposure in Adolescents
• Decoding MGMT Methylation: A Step Towards Precision Medicine in Glioblastoma
• Explicit Correspondence Matching for Generalizable Neural Radiance Fields
• Geometric-Aware Low-Light Image and Video Enhancement via Depth Guidance
• Localized Gaussian Splatting Editing with Contextual Awareness
• A Novel Dataset for Video-Based Neurodivergent Classification Leveraging Extra-Stimulatory Behavior
• Zero-Shot Skeleton-based Action Recognition with Dual Visual-Text Alignment
• The unrealized potential of agroforestry for an emissions-intensive agricultural commodity
• LBONet: Supervised Spectral Descriptors for Shape Analysis
• Efficient Density Control for 3D Gaussian Splatting
• Cascaded Multi-Scale Attention for Enhanced Multi-Scale Feature Extraction and Interaction with Low-Resolution Images
• OccScene: Semantic Occupancy-based Cross-task Mutual Learning for 3D Scene Generation
• Continuous Knowledge-Preserving Decomposition with Adaptive Layer Selection for Few-Shot Class-Incremental Learning
• Review of Demographic Fairness in Face Recognition
• AutoSketch: VLM-assisted Style-Aware Vector Sketch Completion
• LBM: Latent Bridge Matching for Fast Image-to-Image Translation
• Ask Patients with Patience: Enabling LLMs for Human-Centric Medical Dialogue with Grounded Reasoning
• NitiBench: A Comprehensive Study of LLM Framework Capabilities for Thai Legal Question Answering
• Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models
• Is Small Language Model the Silver Bullet to Low-Resource Languages Machine Translation?
• MultiBLiMP 1.0: A Massively Multilingual Benchmark of Linguistic Minimal Pairs
• Exploration of Plan-Guided Summarization for Narrative Texts: the Case of Small Language Models
• DRP: Distilled Reasoning Pruning with Skill-aware Step Decomposition for Efficient Large Reasoning Models
• QA-LIGN: Aligning LLMs through Constitutionally Decomposed QA
• Enhancing Code-switched Text-to-Speech Synthesis Capability in Large Language Models with only Monolingual Corpora
• CAMA: Enhancing Multimodal In-Context Learning with Context-Aware Modulated Attention
• Evaluating Speech-to-Text x LLM x Text-to-Speech Combinations for AI Interview Systems
• Text-Driven 3D Hand Motion Generation from Sign Language Data
• VT-LVLM-AR: A Video-Temporal Large Vision-Language Model Adapter for Fine-Grained Action Recognition in Long-Term Videos
• Boosting Pathology Foundation Models via Few-shot Prompt-tuning for Rare Cancer Subtyping
• Semantic-Aware Ship Detection with Vision-Language Integration
• Automatic Retrieval of Specific Cows from Unlabeled Videos
• Investigating Different Geo Priors for Image Classification
• Glo-VLMs: Leveraging Vision-Language Models for Fine-Grained Diseased Glomerulus Classification
• Contributions to Label-Efficient Learning in Computer Vision and Remote Sensing
• Diverse Signer Avatars with Manual and Non-Manual Feature Modelling for Sign Language Production
• DRespNeT: A UAV Dataset and YOLOv8-DRN Model for Aerial Instance Segmentation of Building Access Points for Post-Earthquake Search-and-Rescue Missions
• NeuralMeshing: Complete Object Mesh Extraction from Casual Captures
• Counterspeech for Mitigating the Influence of Media Bias: Comparing Human and LLM-Generated Responses
• XFinBench: Benchmarking LLMs in Complex Financial Problem Solving and Reasoning
• Dancing with Deer: A Constructional Perspective on MWEs in the Era of LLMs
• Political Ideology Shifts in Large Language Models
• X-Troll: eXplainable Detection of State-Sponsored Information Operations Agents
• Ethical Considerations of Large Language Models in Game Playing
• Less Redundancy: Boosting Practicality of Vision Language Model in Walking Assistants
• Text Takes Over: A Study of Modality Bias in Multimodal Intent Detection
• XLQA: A Benchmark for Locale-Aware Multilingual Open-Domain Question Answering
• ParamBench: A Graduate-Level Benchmark for Evaluating LLM Understanding on Indic Subjects
• Seeing is Believing: Emotion-Aware Audio-Visual Language Modeling for Expressive Speech Generation
• ComicScene154: A Scene Dataset for Comic Analysis
• CMR-SPB: Cross-Modal Multi-Hop Reasoning over Text, Image, and Speech with Path Balance
• TULIP: Adapting Open-Source Large Language Models for Underrepresented Languages and Specialized Financial Tasks
• M3TQA: Massively Multilingual Multitask Table Question Answering
• LLMs that Understand Processes: Instruction-tuning for Semantics-Aware Process Mining
• JaParaPat: A Large-Scale Japanese-English Parallel Patent Application Corpus
• The Mediomatix Corpus: Parallel Data for Romansh Idioms via Comparable Schoolbooks
• ChatGPT-generated texts show authorship traits that identify them as non-human
• A Probabilistic Inference Scaling Theory for LLM Self-Correction
• What makes an entity salient in discourse?
• LLM-as-classifier: Semi-Supervised, Iterative Framework for Hierarchical Text Classification using Large Language Models
• HAMSA: Hijacking Aligned Compact Models via Stealthy Automation
• Unveiling Unicode's Unseen Underpinnings in Undermining Authorship Attribution
• Self-Disguise Attack: Induce the LLM to disguise itself for AIGT detection evasion
• Hardwired-Neurons Language Processing Units as General-Purpose Cognitive Substrates
• AetherCode: Evaluating LLMs' Ability to Win In Premier Programming Competitions
• Retrieval-Augmented Defense: Adaptive and Controllable Jailbreak Prevention for Large Language Models
• Prompting Techniques for Reducing Social Bias in LLMs through System 1 and System 2 Cognitive Processes
• Seamless Language Expansion: Enhancing Multilingual Mastery in Self-Supervised Models
• PublicHearingBR: A Brazilian Portuguese Dataset of Public Hearing Transcripts for Summarization of Long Documents
• Do LLMs write like humans? Variation in grammatical and rhetorical styles
• MINTQA: A Multi-Hop Question Answering Benchmark for Evaluating LLMs on New and Tail Knowledge
• Scalable Scientific Interest Profiling Using Large Language Models
• CyPortQA: Benchmarking Multimodal Large Language Models for Cyclone Preparedness in Port Operation
• MedCoT-RAG: Causal Chain-of-Thought RAG for Medical Question Answering
• DocHop-QA: Towards Multi-Hop Reasoning over Multimodal Document Collections
• QU-NLP at QIAS 2025 Shared Task: A Two-Phase LLM Fine-Tuning and Retrieval-Augmented Generation Approach for Islamic Inheritance Reasoning
• Plinius: Secure and Persistent Machine Learning Model Training
• LIB-KD: Teaching Inductive Bias for Efficient Vision Transformer Distillation and Compression
• A deformation-based framework for learning solution mappings of PDEs defined on varying domains
• Monolithic Hybrid Recommender System for Suggesting Relevant Movies
• Dynamic Optimization of Storage Systems Using Reinforcement Learning Techniques
• Rotary Offset Features in Large Language Models
• Fundamental Limits of Matrix Sensing: Exact Asymptotics, Universality, and Applications
• Contextualize-then-Aggregate: Circuits for In-Context Learning in Gemma-2 2B
• Expected Free Energy-based Planning as Variational Inference
• Representing spherical tensors with scalar-based machine-learning models
• Generative diffusion posterior sampling for informative likelihoods
• Towards Bridging the Reward-Generation Gap in Direct Alignment Algorithms
• General and Estimable Learning Bound Unifying Covariate and Concept Shifts
• A Malliavin calculus approach to score functions in diffusion generative models
• Asymptotic behavior of eigenvalues of large rank perturbations of large random matrices
• Bhav-Net: Knowledge Transfer for Cross-Lingual Antonym vs Synonym Distinction via Dual-Space Graph Transformers
• Format as a Prior: Quantifying and Analyzing Bias in LLMs for Heterogeneous Data
• Do Language Models Agree with Human Perceptions of Suspense in Stories?
• A Framework for Processing Textual Descriptions of Business Processes using a Constrained Language -- Technical Report
• Meet Your New Client: Writing Reports for AI -- Benchmarking Information Loss in Market Research Deliverables
• Avalia\c{c}\~ao de efici\^encia na leitura: uma abordagem baseada em PLN
• Enhancing Cryptocurrency Sentiment Analysis with Multimodal Features
• Embarrassed to observe: The effects of directive language in brand conversation
• A User Manual for cuHALLaR: A GPU Accelerated Low-Rank Semidefinite Programming Solver
• A simulation-based training framework for machine-learning applications in ARPES
• PickleBall: Secure Deserialization of Pickle-based Machine Learning Models
• Cross-Attention Multimodal Fusion for Breast Cancer Diagnosis: Integrating Mammography and Clinical Data with Explainability
• HePGA: A Heterogeneous Processing-in-Memory based GNN Training Accelerator
• FIRE-GNN: Force-informed, Relaxed Equivariance Graph Neural Network for Rapid and Accurate Prediction of Surface Properties
• Optimal Dynamic Regret by Transformers for Non-Stationary Reinforcement Learning
• Training a Foundation Model for Materials on a Budget
• CEQuest: Benchmarking Large Language Models for Construction Estimation
• From Indirect Object Identification to Syllogisms: Exploring Binary Mechanisms in Transformer Circuits
• Neural-Network Chemical Emulator for First-Star Formation: Robust Iterative Predictions over a Wide Density Range
• Domain Adaptation via Feature Refinement
• Deep learning-enabled virtual multiplexed immunostaining of label-free tissue for vascular invasion assessment
• Modeling User Preferences as Distributions for Optimal Transport-based Cross-domain Recommendation under Non-overlapping Settings
• Spike Agreement Dependent Plasticity: A scalable Bio-Inspired learning paradigm for Spiking Neural Networks
• Dac-Fake: A Divide and Conquer Framework for Detecting Fake News on Social Media
• Limit-Computable Grains of Truth for Arbitrary Computable Extensive-Form (Un)Known Games
• Structuring GUI Elements through Vision Language Models: Towards Action Space Generation
• A Sharp KL-Convergence Analysis for Diffusion Models under Minimal Assumptions
• Audio2Face-3D: Audio-driven Realistic Facial Animation For Digital Avatars
• LLM-GUARD: Large Language Model-Based Detection and Repair of Bugs and Security Vulnerabilities in C++ and Python
• Deep Intrinsic Coregionalization Multi-Output Gaussian Process Surrogate with Active Learning
• Integrated Noise and Safety Management in UAM via A Unified Reinforcement Learning Framework
• Beyond Interpretability: Exploring the Comprehensibility of Adaptive Video Streaming through Large Language Models
• Anti-establishment sentiment on TikTok: Implications for understanding influence(rs) and expertise on social media
• Reinforcement Learning-based Control via Y-wise Affine Neural Networks (YANNs)
• Underdamped Langevin MCMC with third order convergence
• Ensembles of Neural Surrogates for Parametric Sensitivity in Ocean Modeling
• ML-PWS: Estimating the Mutual Information Between Experimental Time Series Using Neural Networks
• Quality control in sublinear time: a case study via random graphs
• Parameter-Free Logit Distillation via Sorting Mechanism
• Machine Learning Time Propagators for Time-Dependent Density Functional Theory Simulations
• Transfer Learning via Lexical Relatedness: A Sarcasm and Hate Speech Case Study
• Fair and efficient contribution valuation for vertical federated learning
• Joint Optimization of Energy Consumption and Completion Time in Federated Learning
• Robust Graph Contrastive Learning with Information Restoration
• Implicit Regularization Makes Overparameterized Asymmetric Matrix Sensing Robust to Perturbations
• Reinforcement Learning for Jump-Diffusions, with Financial Applications
• Alignment of Diffusion Models: Fundamentals, Challenges, and Future
• Spiders Based on Anxiety: How Reinforcement Learning Can Deliver Desired User Experience in Virtual Reality Personalized Arachnophobia Treatment
• Decentralized Low-Rank Fine-Tuning of Large Language Models
• Analytics Modelling over Multiple Datasets using Vector Embeddings
• Validating LLM-as-a-Judge Systems under Rating Indeterminacy
• Partially Decentralized Multi-Agent Q-Learning via Digital Cousins for Wireless Networks
• Robustness of deep learning classification to adversarial input on GPUs: asynchronous parallel accumulation is a source of vulnerability
• Mirror, Mirror of the Flow: How Does Regularization Shape Implicit Bias?
• Imputation Not Required in Incremental Learning of Tabular Data with Missing Values
• CROP: Circuit Retrieval and Optimization with Parameter Guidance using LLMs
• Optimal Batch-Size Control for Low-Latency Federated Learning with Device Heterogeneity
• Low-dimensional embeddings of high-dimensional data
• An Efficient Hybridization of Graph Representation Learning and Metaheuristics for the Constrained Incremental Graph Drawing Problem
• Advancing rail safety: An onboard measurement system of rolling stock wheel flange wear based on dynamic machine learning algorithms
• Vector preference-based contextual bandits under distributional shifts
• Scalable Equilibrium Propagation via Intermediate Error Signals for Deep Convolutional CRNNs
• Quantum Federated Learning: A Comprehensive Survey
• Tessellation Groups, Harmonic Analysis on Non-compact Symmetric Spaces and the Heat Kernel in view of Cartan Convolutional Neural Networks
• A State-Space Approach to Nonstationary Discriminant Analysis
• Machine Learning for Medicine Must Be Interpretable, Shareable, Reproducible and Accountable by Design
• AgentFly: Fine-tuning LLM Agents without Fine-tuning LLMs
• SPL-LNS: Sampling-Enhanced Large Neighborhood Search for Solving Integer Linear Programs
• GEM: A Scale-Aware and Distribution-Sensitive Sparse Fine-Tuning Framework for Effective Downstream Adaptation
• UMATO: Bridging Local and Global Structures for Reliable Visual Analytics with Dimensionality Reduction
• PIANO: Physics Informed Autoregressive Network
• When Simpler Wins: Facebooks Prophet vs LSTM for Air Pollution Forecasting in Data-Constrained Northern Nigeria
• FEST: A Unified Framework for Evaluating Synthetic Tabular Data
• Chunked Data Shapley: A Scalable Dataset Quality Assessment for Machine Learning
• On the Evolution of Federated Post-Training Large Language Models: A Model Accessibility View
• OwkinZero: Accelerating Biological Discovery with AI
• Probabilistic Pretraining for Neural Regression
• RotaTouille: Rotation Equivariant Deep Learning for Contours
• Applications and Challenges of Fairness APIs in Machine Learning Software
• Sequential Cohort Selection
• Fast and Accurate RFIC Performance Prediction via Pin Level Graph Neural Networks and Probabilistic Flow
• Double Check My Desired Return: Transformer with Target Alignment for Offline Reinforcement Learning
• Boardwalk: Towards a Framework for Creating Board Games with LLMs
• NOSTRA: A noise-resilient and sparse data framework for trust region based multi objective Bayesian optimization
• Benchmarking the Robustness of Agentic Systems to Adversarially-Induced Harms
• MuST2-Learn: Multi-view Spatial-Temporal-Type Learning for Heterogeneous Municipal Service Time Estimation
• Escaping Saddle Points via Curvature-Calibrated Perturbations: A Complete Analysis with Explicit Constants and Empirical Validation
• Explainable AI in Deep Learning-Based Prediction of Solar Storms
• TinyML Towards Industry 4.0: Resource-Efficient Process Monitoring of a Milling Machine
• Closer to Reality: Practical Semi-Supervised Federated Learning for Foundation Model Adaptation
• Benchmarking Training Paradigms, Dataset Composition, and Model Scaling for Child ASR in ESPnet
• A deep reinforcement learning agent trained for interval timing exhibits similarities to biological systems
• A BERT-based Hierarchical Classification Model with Applications in Chinese Commodity Classification
• Better Together: Leveraging Multiple Digital Twins for Deployment Optimization of Airborne Base Stations
• SDEC: Semantic Deep Embedded Clustering
• Mining Mental Health Signals: A Comparative Study of Four Machine Learning Methods for Depression Detection from Social Media Posts in Sorani Kurdish
• A Review of Developmental Interpretability in Large Language Models
• Lexical Hints of Accuracy in LLM Reasoning Chains
• Mechanistic Exploration of Backdoored Large Language Model Attention Patterns
• Linkage Attacks Expose Identity Risks in Public ECG Data Sharing
• Correctness-Guaranteed Code Generation via Constrained Decoding
• Beyond Transcription: Mechanistic Interpretability in ASR
• CIGaRS I: Combined simulation-based inference from SNae Ia and host photometry
• Interpretable Kernels
• Continuous Determination of Respiratory Rate in Hospitalized Patients using Machine Learning Applied to Electrocardiogram Telemetry
• AI-Powered CPS-Enabled Urban Transportation Digital Twin: Methods and Applications
• Can Hallucinations Help? Boosting LLMs for Drug Discovery
• Ethical Concerns of Generative AI and Mitigation Strategies: A Systematic Mapping Study
• Towards Privacy-aware Mental Health AI Models: Advances, Challenges, and Opportunities
• Score as Action: Fine-Tuning Diffusion Generative Models by Continuous-time Reinforcement Learning
• SIGMA: Sheaf-Informed Geometric Multi-Agent Pathfinding
• Top-Theta Attention: Sparsifying Transformers by Compensated Thresholding
• One Example Shown, Many Concepts Known! Counterexample-Driven Conceptual Reasoning in Mathematical LLMs
• Soteria: Language-Specific Functional Parameter Steering for Multilingual Safety Alignment
• Collaborative Stance Detection via Small-Large Language Model Consistency Verification
• from Benign import Toxic: Jailbreaking the Language Model via Adversarial Metaphors
• Can Large Language Models Simulate Human Responses? A Case Study of Stated Preference Experiments in the Context of Heating-related Choices
• Text-to-3D Generation using Jensen-Shannon Score Distillation
• Comparative Explanations: Explanation Guided Decision Making for Human-in-the-Loop Preference Selection
• Leveraging LLMs for Utility-Focused Annotation: Reducing Manual Effort for Retrieval and RAG
• FedEFC: Federated Learning Using Enhanced Forward Correction Against Noisy Labels
• MSCCL++: Rethinking GPU Communication Abstractions for Cutting-edge AI Applications
• DIDS: Domain Impact-aware Data Sampling for Large Language Model Training
• Tripartite-GraphRAG via Plugin Ontologies
• Perceptual Implications of Automatic Anonymization in Pathological Speech
• MedArabiQ: Benchmarking Large Language Models on Arabic Medical Tasks
• Direct Image Classification from Fourier Ptychographic Microscopy Measurements without Reconstruction
• Enhancing and Scaling Search Query Datasets for Recommendation Systems
• CCD: Continual Consistency Diffusion for Lifelong Generative Modeling
• Hybrid Adaptive Modeling in Process Monitoring: Leveraging Sequence Encoders and Physics-Informed Neural Networks
• A Text-Based Recommender System that Leverages Explicit Affective State Preferences
• SpecExtend: A Drop-in Enhancement for Speculative Decoding of Long Sequences
• PoisonSwarm: Universal Harmful Information Synthesis via Model Crowdsourcing
• SPARE: Single-Pass Annotation with Reference-Guided Evaluation for Automatic Process Supervision and Reward Modelling
• GPU Kernel Scientist: An LLM-Driven Framework for Iterative Kernel Optimization
• Neural-Network solver of ideal MHD equilibria
• Generalized Tree Edit Distance (GTED): A Faithful Evaluation Metric for Statement Autoformalization
• A Simple "Try Again" Can Elicit Multi-Turn LLM Reasoning
• Z-Pruner: Post-Training Pruning of Large Language Models for Efficiency without Retraining
• PGF-Net: A Progressive Gated-Fusion Framework for Efficient Multimodal Sentiment Analysis
• Physics-Based Explainable AI for ECG Segmentation: A Lightweight Model
• Transforming Causality: Transformer-Based Temporal Causal Discovery with Prior Knowledge Integration
• Take That for Me: Multimodal Exophora Resolution with Interactive Questioning for Ambiguous Out-of-View Instructions
• On the Collapse Errors Induced by the Deterministic Sampler for Diffusion Models
• Beyond Human-prompting: Adaptive Prompt Tuning with Semantic Alignment for Anomaly Detection
• Through the Looking Glass: A Dual Perspective on Weakly-Supervised Few-Shot Segmentation
• STA-GANN: A Valid and Generalizable Spatio-Temporal Kriging Approach
• Towards Recommending Usability Improvements with Multimodal Large Language Models
• EGRA:Toward Enhanced Behavior Graphs and Representation Alignment for Multimodal Recommendation
• Motor Imagery EEG Signal Classification Using Minimally Random Convolutional Kernel Transform and Hybrid Deep Learning
• LLM-Assisted Semantic Alignment and Integration in Collaborative Model-Based Systems Engineering Using SysML v2
• A Relay-Chain-Powered Ciphertext-Policy Attribute-Based Encryption in Intelligent Transportation Systems
• Set Transformer Architectures and Synthetic Data Generation for Flow-Guided Nanoscale Localization
• SpecVLM: Enhancing Speculative Decoding of Video LLMs via Verifier-Guided Token Pruning
• OmniCache: A Trajectory-Oriented Global Perspective on Training-Free Cache Reuse for Diffusion Transformer Models
• An Investigation of Visual Foundation Models Robustness
• FlexMUSE: Multimodal Unification and Semantics Enhancement Framework with Flexible interaction for Creative Writing
• A XAI-based Framework for Frequency Subband Characterization of Cough Spectrograms in Chronic Respiratory Disease
• A Reduction of Input/Output Logics to SAT
• MCPVerse: An Expansive, Real-World Benchmark for Agentic Tool Use
• From Confidence to Collapse in LLM Factual Robustness
• Representation Learning of Auxiliary Concepts for Improved Student Modeling and Exercise Recommendation
• A Multimodal-Multitask Framework with Cross-modal Relation and Hierarchical Interactive Attention for Semantic Comprehension
• Exploiting Information Redundancy in Attention Maps for Extreme Quantization of Vision Transformers
• Retrieval Enhanced Feedback via In-context Neural Error-book
• Cyber Physical Awareness via Intent-Driven Threat Assessment: Enhanced Space Networks with Intershell Links
• LLMSymGuard: A Symbolic Safety Guardrail Framework Leveraging Interpretable Jailbreak Concepts
• Vevo2: Bridging Controllable Speech and Singing Voice Generation via Unified Prosody Learning
• Unsupervised Online Detection of Pipe Blockages and Leakages in Water Distribution Networks
• Uppaal Coshy: Automatic Synthesis of Compact Shields for Hybrid Systems
• Confusion is the Final Barrier: Rethinking Jailbreak Evaluation and Investigating the Real Misuse Threat of LLMs
• MizanQA: Benchmarking Large Language Models on Moroccan Legal Question Answering
• RoMedQA: The First Benchmark for Romanian Medical Question Answering
• Domain-aligned generative downscaling enhances projections of extreme climate events
• A Lightweight Group Multiscale Bidirectional Interactive Network for Real-Time Steel Surface Defect Detection
• Cetvel: A Unified Benchmark for Evaluating Language Understanding, Generation and Cultural Capacity of LLMs for Turkish
• OPERA: A Reinforcement Learning--Enhanced Orchestrated Planner-Executor Architecture for Reasoning-Oriented Multi-Hop Retrieval
• PediatricsMQA: a Multi-modal Pediatrics Question Answering Benchmark
• HOSt3R: Keypoint-free Hand-Object 3D Reconstruction from RGB images
• Disentangled Multi-modal Learning of Histology and Transcriptomics for Cancer Characterization
• FraPPE: Fast and Efficient Preference-based Pure Exploration
• SafeSpace: An Integrated Web Application for Digital Safety and Emotional Well-being
• Post Hoc Regression Refinement via Pairwise Rankings
• On Zero-Shot Reinforcement Learning
• FLAMES: Improving LLM Math Reasoning via a Fine-Grained Analysis of the Data Synthesis Pipeline
• Comparative Analysis of UAV Path Planning Algorithms for Efficient Navigation in Urban 3D Environments
• Guiding Diffusion Models with Reinforcement Learning for Stable Molecule Generation
• Towards Open World Detection: A Survey
• RL Is Neither a Panacea Nor a Mirage: Understanding Supervised vs. Reinforcement Learning Fine-Tuning for LLMs
• Enhanced NIRMAL Optimizer With Damped Nesterov Acceleration: A Comparative Analysis
• Time-Aware One Step Diffusion Network for Real-World Image Super-Resolution
• Sparse but Wrong: Incorrect L0 Leads to Incorrect Features in Sparse Autoencoders
• A Disease-Centric Vision-Language Foundation Model for Precision Oncology in Kidney Cancer
• Hierarchical Decision-Making for Autonomous Navigation: Integrating Deep Reinforcement Learning and Fuzzy Logic in Four-Wheel Independent Steering and Driving Systems
• MV-RAG: Retrieval Augmented Multiview Diffusion
• Overcoming classic challenges for artificial neural networks by providing incentives and practice
• Coarse-to-Fine Process Reward Modeling for Mathematical Reasoning
• VERUS-LM: a Versatile Framework for Combining LLMs with Symbolic Reasoning
• Efficient RL Training for Reasoning Models via Length-Aware Optimization
• HypER: Literature-grounded Hypothesis Generation and Distillation with Provenance
• Unsupervised Automata Learning via Discrete Optimization
• Towards Goal-oriented Intelligent Tutoring Systems in Online Education
• Explainable Bayesian Optimization
• A Curious Case of Remarkable Resilience to Gradient Attacks via Fully Convolutional and Differentiable Front End with a Skip Connection
• On the Challenges and Opportunities in Generative AI
• A Diffusion Model Framework for Unsupervised Neural Combinatorial Optimization
• Learning Image Priors through Patch-based Diffusion Models for Solving Inverse Problems
• Sentiment Reasoning for Healthcare
• Balancing Act: Prioritization Strategies for LLM-Designed Restless Bandit Rewards
• AutoVerus: Automated Proof Generation for Rust Code
• How Performance Pressure Influences AI-Assisted Decision Making
• Two pathways to resolve relational inconsistencies
• Establishing Task Scaling Laws via Compute-Efficient Model Ladders
• LearnLM: Improving Gemini for Learning
• Strategic Sample Selection for Improved Clean-Label Backdoor Attacks in Text Classification
• ASIC-Agent: An Autonomous Multi-Agent System for ASIC Design with Benchmark Evaluation
• Representation Learning with Adaptive Superpixel Coding
• Panoptic Segmentation of Environmental UAV Images : Litter Beach
• Automated Multi-label Classification of Eleven Retinal Diseases: A Benchmark of Modern Architectures and a Meta-Ensemble on a Large Synthetic Dataset
• Breaking Barriers in Software Testing: The Power of AI-Driven Automation
• CoVeRaP: Cooperative Vehicular Perception through mmWave FMCW Radars
• Time Series Based Network Intrusion Detection using MTF-Aided Transformer
• Pareto Actor-Critic for Communication and Computation Co-Optimization in Non-Cooperative Federated Learning Services
• Enhanced predictions of the Madden-Julian oscillation using the FuXi-S2S machine learning model: Insights into physical mechanisms
• OpenWHO: A Document-Level Parallel Corpus for Health Translation in Low-Resource Languages
• From Benchmark Data To Applicable Program Repair: An Experience Report
• Cooperative Design Optimization through Natural Language Interaction
• On Task Vectors and Gradients
• Two-flow Feedback Multi-scale Progressive Generative Adversarial Network
• GPLight+: A Genetic Programming Method for Learning Symmetric Traffic Signal Control Policy
• CYCLE-INSTRUCT: Fully Seed-Free Instruction Tuning via Dual Self-Training and Cycle Consistency
• ANSC: Probabilistic Capacity Health Scoring for Datacenter-Scale Reliability
• Spacetime-GR: A Spacetime-Aware Generative Model for Large Scale Online POI Recommendation
• The Fools are Certain; the Wise are Doubtful: Exploring LLM Confidence in Code Completion
• CommonKV: Compressing KV Cache with Cross-layer Parameter Sharing
• Machine Learning in Micromobility: A Systematic Review of Datasets, Techniques, and Applications
• Causal Beam Selection for Reliable Initial Access in AI-driven Beam Management
• GLARE: Agentic Reasoning for Legal Judgment Prediction
• Modular Embedding Recomposition for Incremental Learning
• Constraints-Guided Diffusion Reasoner for Neuro-Symbolic Learning
• LLM-Based Agents for Competitive Landscape Mapping in Drug Asset Due Diligence
• Learning in Focus: Detecting Behavioral and Collaborative Engagement Using Vision Transformers
• KG-o1: Enhancing Multi-hop Question Answering in Large Language Models via Knowledge Graph Integration
• InteChar: A Unified Oracle Bone Character List for Ancient Chinese Language Modeling
• Benchmarking the Legal Reasoning of LLMs in Arabic Islamic Inheritance Cases
• Benchmarking the Medical Understanding and Reasoning of Large Language Models in Arabic Healthcare Tasks
• Persuasiveness and Bias in LLM: Investigating the Impact of Persuasiveness and Reinforcement of Bias in Language Models
• LingVarBench: Benchmarking LLM for Automated Named Entity Recognition in Structured Synthetic Spoken Transcriptions
• MAC: A Live Benchmark for Multimodal Large Language Models in Scientific Understanding
• ReportBench: Evaluating Deep Research Agents via Academic Survey Tasks
• ALAS: Autonomous Learning Agent for Self-Updating Language Models
• SurfaceLogicKV: Surface and Logic Attention Behaviors are All You Need for Robust KV Cache Compression
• KL-based self-distillation for large language models
• Uplifted Attackers, Human Defenders: The Cyber Offense-Defense Balance for Trailing-Edge Organizations
• Chain-of-Query: Unleashing the Power of LLMs in SQL-Aided Table Understanding via Multi-Agent Collaboration
• Detecting Hope, Hate, and Emotion in Arabic Textual Speech and Multi-modal Memes Using Large Language Models
• From Clicks to Preference: A Multi-stage Alignment Framework for Generative Query Suggestion in Conversational System
• SCOPE: A Generative Approach for LLM Prompt Compression
• User-Assistant Bias in LLMs
• Research on intelligent generation of structural demolition suggestions based on multi-model collaboration
• Straggler-Resilient Federated Learning over A Hybrid Conventional and Pinching Antenna Network
• An Auditable Pipeline for Fuzzy Full-Text Screening in Systematic Reviews: Integrating Contrastive Semantic Highlighting and LLM Judgment
• Mini-Omni-Reasoner: Token-Level Thinking-in-Speaking in Large Speech Models
• DAIQ: Auditing Demographic Attribute Inference from Question in LLMs
• Who's Asking? Investigating Bias Through the Lens of Disability Framed Queries in LLMs
• A Functionality-Grounded Benchmark for Evaluating Web Agents in E-commerce Domains
• Alvorada-Bench: Can Language Models Solve Brazilian University Entrance Exams?
• MorphNAS: Differentiable Architecture Search for Morphologically-Aware Multilingual NER
• Statistical Comparative Analysis of Semantic Similarities and Model Transferability Across Datasets for Short Answer Grading
• CIA+TA Risk Assessment for AI Reasoning Vulnerabilities
• Coarse-to-Fine Personalized LLM Impressions for Streamlined Radiology Reports
• MGSC: A Multi-granularity Consistency Framework for Robust End-to-end Asr
• Building and Measuring Trust between Large Language Models
• Beyond Individuals: Collective Predictive Coding for Memory, Attention, and the Emergence of Language
• Securing Swarms: Cross-Domain Adaptation for ROS2-based CPS Anomaly Detection
• CARFT: Boosting LLM Reasoning via Contrastive Learning with Annotated Chain-of-Thought-based Reinforced Fine-Tuning
• Spatial Policy: Guiding Visuomotor Robotic Manipulation with Spatial-Aware Modeling and Reasoning
• NEAT: Concept driven Neuron Attribution in LLMs
• DeepMEL: A Multi-Agent Collaboration Framework for Multimodal Entity Linking
• Annif at the GermEval-2025 LLMs4Subjects Task: Traditional XMTC Augmented by Efficient LLMs
• Lean Meets Theoretical Computer Science: Scalable Synthesis of Theorem Proving Challenges in Formal-Informal Pairs
• TPLA: Tensor Parallel Latent Attention for Efficient Disaggregated Prefill \& Decode Inference
• Beyond Imaging: Vision Transformer Digital Twin Surrogates for 3D+T Biological Tissue Dynamics
• Jet-Nemotron: Efficient Language Model with Post Neural Architecture Search
• Evaluating Structured Decoding for Text-to-Table Generation: Evidence from Three Datasets
• Information Ecosystem Reengineering via Public Sector Knowledge Representation
• HyperFlexis: Joint Design of Algorithms and Systems for Multi-SLO Serving and Fast Scaling
• Probabilistic Forecasting Cryptocurrencies Volatility: From Point to Quantile Forecasts
• Noise, Adaptation, and Strategy: Assessing LLM Fidelity in Decision-Making
• T-ILR: a Neurosymbolic Integration for LTLf
• CoFE: A Framework Generating Counterfactual ECG for Explainable Cardiac AI-Diagnostics
• MMAPG: A Training-Free Framework for Multimodal Multi-hop Question Answering via Adaptive Planning Graphs
• Generative Foundation Model for Structured and Unstructured Electronic Health Records
• Urban Comfort Assessment in the Era of Digital Planning: A Multidimensional, Data-driven, and AI-assisted Framework
• Integrating Time Series into LLMs via Multi-layer Steerable Embedding Fusion for Enhanced Forecasting
• InMind: Evaluating LLMs in Capturing and Applying Individual Human Reasoning Styles
• IR-Agent: Expert-Inspired LLM Agents for Structure Elucidation from Infrared Spectra
• Extending FKG.in: Towards a Food Claim Traceability Network
• Bridging the Gap in Ophthalmic AI: MM-Retinal-Reason Dataset and OphthaReason Model toward Dynamic Multimodal Reasoning
• Graph RAG as Human Choice Model: Building a Data-Driven Mobility Agent with Preference Chain
• Competition and Attraction Improve Model Fusion
• The next question after Turing's question: Introducing the Grow-AI test
• AgentScope 1.0: A Developer-Centric Framework for Building Agentic Applications
• Do What? Teaching Vision-Language-Action Models to Reject the Impossible