AI Research News Feeds for September 23rd, 2025

AI RESEARCH PAPERS & ACADEMIC SOURCES

• Multi-needle Localization for Pelvic Seed Implant Brachytherapy based on Tip-handle Detection and Matching
• Can multimodal representation learning by alignment preserve modality-specific information?
• Enhancing Semantic Segmentation with Continual Self-Supervised Pre-training
• ContextFlow: Training-Free Video Object Editing via Adaptive Context Enrichment
• Semantic and Visual Crop-Guided Diffusion Models for Heterogeneous Tissue Synthesis in Histopathology
• ProDyG: Progressive Dynamic Scene Reconstruction via Gaussian Splatting from Monocular Videos
• Trainee Action Recognition through Interaction Analysis in CCATT Mixed-Reality Training
• Does Audio Matter for Modern Video-LLMs and Their Benchmarks?
• SmaRT: Style-Modulated Robust Test-Time Adaptation for Cross-Domain Brain Tumor Segmentation in MRI
• Accurate and Efficient Low-Rank Model Merging in Core Space
• From Restoration to Reconstruction: Rethinking 3D Gaussian Splatting for Underwater Scenes
• Degradation-Aware All-in-One Image Restoration via Latent Prior Encoding
• TS-P$^2$CL: Plug-and-Play Dual Contrastive Learning for Vision-Guided Medical Time Series Classification
• Selecting Optimal Camera Views for Gait Analysis: A Multi-Metric Assessment of 2D Projections
• FROQ: Observing Face Recognition Models for Efficient Quality Assessment
• Depth Edge Alignment Loss: DEALing with Depth in Weakly Supervised Semantic Segmentation
• Neurodynamics-Driven Coupled Neural P Systems for Multi-Focus Image Fusion
• Automatic Intermodal Loading Unit Identification using Computer Vision: A Scoping Review
• RCTDistill: Cross-Modal Knowledge Distillation Framework for Radar-Camera 3D Object Detection with Temporal Fusion
• Automated Labeling of Intracranial Arteries with Uncertainty Quantification Using Deep Learning
• Adaptive Fast-and-Slow Visual Program Reasoning for Long-Form VideoQA
• Dual-View Alignment Learning with Hierarchical-Prompt for Class-Imbalance Multi-Label Classification
• Multi-Agent Amodal Completion: Direct Synthesis with Fine-Grained Semantic Guidance
• Neural-MMGS: Multi-modal Neural Gaussian Splats for Large-Scale Scene Reconstruction
• Incorporating the Refractory Period into Spiking Neural Networks through Spike-Triggered Threshold Dynamics
• I2VWM: Robust Watermarking for Image to Video Generation
• OmniInsert: Mask-Free Video Insertion of Any Reference via Diffusion Transformer Models
• A$^2$M$^2$-Net: Adaptively Aligned Multi-Scale Moment for Few-Shot Action Recognition
• VideoArtGS: Building Digital Twins of Articulated Objects from Monocular Video
• Evict3R: Training-Free Token Eviction for Memory-Bounded Streaming Visual Geometry Transformers
• SISMA: Semantic Face Image Synthesis with Mamba
• Clothing agnostic Pre-inpainting Virtual Try-ON
• Development and validation of an AI foundation model for endoscopic diagnosis of esophagogastric junction adenocarcinoma: a cohort and deep learning study
• SD-VLM: Spatial Measuring and Understanding with Depth-Encoded Vision-Language Models
• Tailored Transformation Invariance for Industrial Anomaly Detection
• DINOv3-Diffusion Policy: Self-Supervised Large Visual Model for Visuomotor Diffusion Policy Learning
• Predicting Depth Maps from Single RGB Images and Addressing Missing Information in Depth Estimation
• SouLLMate: An Application Enhancing Diverse Mental Health Support with Adaptive LLMs, Prompt Engineering, and RAG Techniques
• Visual Instruction Pretraining for Domain-Specific Foundation Models
• MRN: Harnessing 2D Vision Foundation Models for Diagnosing Parkinson's Disease with Limited 3D MR Data
• PRNU-Bench: A Novel Benchmark and Model for PRNU-Based Camera Identification
• Interpreting Attention Heads for Image-to-Text Information Flow in Large Vision-Language Models
• Domain Adaptive Object Detection for Space Applications with Real-Time Constraints
• COLA: Context-aware Language-driven Test-time Adaptation
• From Benchmarks to Reality: Advancing Visual Anomaly Detection by the VAND 3.0 Challenge
• Tensor-Based Self-Calibration of Cameras via the TrifocalCalib Method
• Overview of PlantCLEF 2023: Image-based Plant Identification at Global Scale
• From Documents to Database: Failure Modes for Industrial Assets
• Through the Lens of Human-Human Collaboration: A Configurable Research Platform for Exploring Human-Agent Collaboration
• OnePiece: Bringing Context Engineering and Reasoning to Industrial Cascade Ranking System
• MetaEmbed: Scaling Multimodal Retrieval at Test-Time with Flexible Late Interaction
• Scaling Efficient LLMs
• MindRef: Mimicking Human Memory for Hierarchical Reference Retrieval with Fine-Grained Location Awareness
• Temporal Scaling Law for Large Language Models
• LLaSA: A Sensor-Aware LLM for Natural Language Reasoning of Human Activity from IMU Data
• Evaluating Fairness in Large Vision-Language Models Across Diverse Demographic Attributes and Prompts
• On the Low-Rank Parametrization of Reward Models for Controlled Language Generation
• Rethinking Backdoor Detection Evaluation for Language Models
• GALLa: Graph Aligned Large Language Models for Improved Source Code Understanding
• Autiverse: Eliciting Autistic Adolescents' Daily Narratives through AI-guided Multimodal Journaling
• LingoQ: Bridging the Gap between ESL Learning and Work through AI-Generated Work-Related Quizzes
• AutiHero: Leveraging Generative AI in Social Narratives to Engage Parents in Story-Driven Behavioral Guidance for Autistic Children
• WISE: Weak-Supervision-Guided Step-by-Step Explanations for Multimodal LLMs in Image Classification
• Deep Learning Inductive Biases for fMRI Time Series Classification during Resting-state and Movie-watching
• From Prediction to Understanding: Will AI Foundation Models Transform Brain Science?
• Explainability matters: The effect of liability rules on the healthcare sector
• Barwise Section Boundary Detection in Symbolic Music Using Convolutional Neural Networks
• Bayesian Ego-graph inference for Networked Multi-Agent Reinforcement Learning
• Safe Guaranteed Dynamics Exploration with Probabilistic Models
• Knowledge Distillation for Variational Quantum Convolutional Neural Networks on Heterogeneous Data
• Increase Alpha: Performance and Risk of an AI-Driven Trading Framework
• A Study on Stabilizer R\'enyi Entropy Estimation using Machine Learning
• Exploring AI Capabilities in Participatory Budgeting within Smart Cities: The Case of Sao Paulo
• Comparing RAG and GraphRAG for Page-Level Retrieval Question Answering on Math Textbook
• AdaptiveGuard: Towards Adaptive Runtime Safety for LLM-Powered Software
• Cross-Attention with Confidence Weighting for Multi-Channel Audio Alignment
• Leveraging Multiple Speech Enhancers for Non-Intrusive Intelligibility Prediction for Hearing-Impaired Listeners
• A Chain-of-thought Reasoning Breast Ultrasound Dataset Covering All Histopathology Categories
• $\texttt{DiffSyn}$: A Generative Diffusion Approach to Materials Synthesis Planning
• Prompt-with-Me: in-IDE Structured Prompt Management for LLM-Driven Software Engineering
• MaskVCT: Masked Voice Codec Transformer for Zero-Shot Voice Conversion With Increased Controllability via Multiple Guidances
• Agentic AI for Multi-Stage Physics Experiments at a Large-Scale User Facility Particle Accelerator
• Generalizability of Large Language Model-Based Agents: A Comprehensive Survey
• Significativity Indices for Agreement Values
• Towards Quantifying the Hessian Structure of Neural Networks
• Wasserstein Convergence of Score-based Generative Models under Semiconvexity and Discontinuous Gradients
• Minimax Adaptive Online Nonparametric Regression over Besov Spaces
• Multi-scale clustering and source separation of InSight mission seismic data
• Addressing the Inconsistency in Bayesian Deep Learning via Generalized Laplace Approximation
• Theoretical Insights into CycleGAN: Analyzing Approximation and Estimation Errors in Unpaired Data Generation
• Robust Reinforcement Learning with Dynamic Distortion Risk Measures
• Measure-to-measure interpolation using Transformers
• Are Deep Learning Methods Suitable for Downscaling Global Climate Projections? An Intercomparison for Temperature and Precipitation over Spain
• Flatten Graphs as Sequences: Transformers are Scalable Graph Generators
• Unsupervised Structural-Counterfactual Generation under Domain Shift
• Bayesian Algorithms for Adversarial Online Learning: from Finite to Infinite Action Spaces
• Cover Learning for Large-Scale Topology Representation
• Locally minimax optimal confidence sets for the best model
• On Quantification of Borrowing of Information in Hierarchical Bayesian Models
• Bayesian Semi-supervised Inference via a Debiased Modeling Approach
• Synth-MIA: A Testbed for Auditing Privacy Leakage in Tabular Data Synthesis
• Core-elements Subsampling for Alternating Least Squares
• Deep Learning as the Disciplined Construction of Tame Objects
• Validation-Free Sparse Learning: A Phase Transition Approach to Feature Selection
• Tight PAC-Bayesian Risk Certificates for Contrastive Learning
• Learning Massive-scale Partial Correlation Networks in Clinical Multi-omics Studies with HP-ACCORD
• An AI-powered Bayesian generative modeling approach for causal inference in observational studies
• Conformal Prediction with Upper and Lower Bound Models
• AICO: Feature Significance Tests for Supervised Learning
• Robust Mixture Models for Algorithmic Fairness Under Latent Heterogeneity
• Bilateral Distribution Compression: Reducing Both Data Size and Dimensionality
• Whitening Spherical Gaussian Mixtures in the Large-Dimensional Regime
• Robust, Online, and Adaptive Decentralized Gaussian Processes
• Fr\'echet Geodesic Boosting
• Kernel K-means clustering of distributional data
• Functional effects models: Accounting for preference heterogeneity in panel data with machine learning
• Learning Centre Partitions from Summaries
• Overfitting in Adaptive Robust Optimization
• Hessian-guided Perturbed Wasserstein Gradient Flows for Escaping Saddle Points
• Data-efficient Kernel Methods for Learning Hamiltonian Systems
• Chat-CBM: Towards Interactive Concept Bottleneck Models with Frozen Large Language Models
• SimToken: A Simple Baseline for Referring Audio-Visual Segmentation
• An Empirical Study on the Robustness of YOLO Models for Underwater Object Detection
• Low-Rank Adaptation of Evolutionary Deep Neural Networks for Efficient Learning of Time-Dependent PDEs
• Conditional Multidimensional Scaling with Incomplete Conditioning Data
• System-Level Uncertainty Quantification with Multiple Machine Learning Models: A Theoretical Framework
• DoubleGen: Debiased Generative Modeling of Counterfactuals
• Risk Comparisons in Linear Regression: Implicit Regularization Dominates Explicit Regularization
• Bias-variance Tradeoff in Tensor Estimation
• CARINOX: Inference-time Scaling with Category-Aware Reward-based Initial Noise Optimization and Exploration
• CSDformer: A Conversion Method for Fully Spike-Driven Transformer
• MAESTRO: Task-Relevant Optimization via Adaptive Feature Enhancement and Suppression for Multi-task 3D Perception
• Stable Video-Driven Portraits
• ChartHal: A Fine-grained Framework Evaluating Hallucination of Large Vision Language Models in Chart Understanding
• Multimodal Medical Image Classification via Synergistic Learning Pre-training
• Vision-Based Driver Drowsiness Monitoring: Comparative Analysis of YOLOv5-v11 Models
• SAMSON: 3rd Place Solution of LSVOS 2025 VOS Challenge
• 4D-MoDe: Towards Editable and Scalable Volumetric Streaming via Motion-Decoupled 4D Gaussian Compression
• 4DGCPro: Efficient Hierarchical 4D Gaussian Compression for Progressive Volumetric Video Streaming
• Unified Multimodal Coherent Field: Synchronous Semantic-Spatial-Vision Fusion for Brain Tumor Segmentation
• Pre-Trained CNN Architecture for Transformer-Based Image Caption Generation Model
• Revisiting Vision Language Foundations for No-Reference Image Quality Assessment
• Diff-GNSS: Diffusion-based Pseudorange Error Estimation
• Interpreting vision transformers via residual replacement model
• Real-Time Fish Detection in Indonesian Marine Ecosystems Using Lightweight YOLOv10-nano Architecture
• Single-Image Depth from Defocus with Coded Aperture and Diffusion Posterior Sampling
• Multi-scale Temporal Prediction via Incremental Generation and Multi-agent Collaboration
• EmbodiedSplat: Personalized Real-to-Sim-to-Real Navigation with Gaussian Splats from a Mobile Device
• Emergent 3D Correspondence from Neural Shape Representation
• Training-Free Label Space Alignment for Universal Domain Adaptation
• Explainable AI for Analyzing Person-Specific Patterns in Facial Recognition Tasks
• Echo-Path: Pathology-Conditioned Echo Video Generation
• Guided and Unguided Conditional Diffusion Mechanisms for Structured and Semantically-Aware 3D Point Cloud Generation
• Point-RTD: Replaced Token Denoising for Pretraining Transformer Models on Point Clouds
• MirrorSAM2: Segment Mirror in Videos with Depth Perception
• DT-NeRF: A Diffusion and Transformer-Based Optimization Approach for Neural Radiance Fields in 3D Reconstruction
• SPFSplatV2: Efficient Self-Supervised Pose-Free 3D Gaussian Splatting from Sparse Views
• Optimized Learned Image Compression for Facial Expression Recognition
• Task-Oriented Communications for 3D Scene Representation: Balancing Timeliness and Fidelity
• Automated Facility Enumeration for Building Compliance Checking using Door Detection and Large Language Models
• DepTR-MOT: Unveiling the Potential of Depth-Informed Trajectory Refinement for Multi-Object Tracking
• UIPro: Unleashing Superior Interaction Capability For GUI Agents
• SmokeSeer: 3D Gaussian Splatting for Smoke Removal and Scene Reconstruction
• MoCLIP-Lite: Efficient Video Recognition by Fusing CLIP with Motion Vectors
• SFN-YOLO: Towards Free-Range Poultry Detection via Scale-aware Fusion Networks
• AlignedGen: Aligning Style Across Generated Images
• Uncertainty-Supervised Interpretable and Robust Evidential Segmentation
• The SAGES Critical View of Safety Challenge: A Global Benchmark for AI-Assisted Surgical Quality Assessment
• CoBEVMoE: Heterogeneity-aware Feature Fusion with Dynamic Mixture-of-Experts for Collaborative Perception
• Stencil: Subject-Driven Generation with Context Guidance
• SAEC: Scene-Aware Enhanced Edge-Cloud Collaborative Industrial Vision Inspection with Multimodal LLM
• SynergyNet: Fusing Generative Priors and State-Space Models for Facial Beauty Prediction
• Ambiguous Medical Image Segmentation Using Diffusion Schr\"{o}dinger Bridge
• Efficient 3D Scene Reconstruction and Simulation from Sparse Endoscopic Views
• From Easy to Hard: The MIR Benchmark for Progressive Interleaved Multi-Image Reasoning
• Towards Generalized Synapse Detection Across Invertebrate Species
• AgriDoctor: A Multimodal Intelligent Assistant for Agriculture
• Learning Attribute-Aware Hash Codes for Fine-Grained Image Retrieval via Query Optimization
• Geodesic Prototype Matching via Diffusion Maps for Interpretable Fine-Grained Recognition
• CardiacCLIP: Video-based CLIP Adaptation for LVEF Prediction in a Few-shot Manner
• Informative Text-Image Alignment for Visual Affordance Learning with Foundation Models
• Enhanced Detection of Tiny Objects in Aerial Images
• A Dual-Modulation Framework for RGB-T Crowd Counting via Spatially Modulated Attention and Adaptive Fusion
• HyRF: Hybrid Radiance Fields for Memory-efficient and High-quality Novel View Synthesis
• VidCLearn: A Continual Learning Approach for Text-to-Video Generation
• MO R-CNN: Multispectral Oriented R-CNN for Object Detection in Remote Sensing Image
• Penalizing Boundary Activation for Object Completeness in Diffusion Models
• LLM-Assisted Semantic Guidance for Sparsely Annotated Remote Sensing Object Detection
• The 1st Solution for 7th LSVOS RVOS Track: SaSaSa2VA
• Optimal Transport for Handwritten Text Recognition in a Low-Resource Regime
• VCE: Safe Autoregressive Image Generation via Visual Contrast Exploitation
• A Cross-Hierarchical Multi-Feature Fusion Network Based on Multiscale Encoder-Decoder for Hyperspectral Change Detection
• DocIQ: A Benchmark Dataset and Feature Fusion Network for Document Image Quality Assessment
• When Color-Space Decoupling Meets Diffusion for Adverse-Weather Image Restoration
• $\mathtt{M^3VIR}$: A Large-Scale Multi-Modality Multi-View Synthesized Benchmark Dataset for Image Restoration and Content Creation
• SAM-DCE: Addressing Token Uniformity and Semantic Over-Smoothing in Medical Segmentation
• Rethinking Evaluation of Infrared Small Target Detection
• Learning from Gene Names, Expression Values and Images: Contrastive Masked Text-Image Pretraining for Spatial Transcriptomics Representation Learning
• PRISM: Precision-Recall Informed Data-Free Knowledge Distillation via Generative Diffusion
• ME-Mamba: Multi-Expert Mamba with Efficient Knowledge Capture and Fusion for Multimodal Survival Analysis
• SLAM-Former: Putting SLAM into One Transformer
• Parameter-efficient fine-tuning (PEFT) of Vision Foundation Models for Atypical Mitotic Figure Classification
• Prototype-Based Pseudo-Label Denoising for Source-Free Domain Adaptation in Remote Sensing Semantic Segmentation
• Catching the Details: Self-Distilled RoI Predictors for Fine-Grained MLLM Perception
• Leveraging RGB Images for Pre-Training of Event-Based Hand Pose Estimation
• Min: Mixture of Noise for Pre-Trained Model-Based Class-Incremental Learning
• CAMBench-QR : A Structure-Aware Benchmark for Post-Hoc Explanations with QR Understanding
• HyPlaneHead: Rethinking Tri-plane-like Representations in Full-Head Image Synthesis
• DiffEye: Diffusion-Based Continuous Eye-Tracking Data Generation Conditioned on Natural Images
• MMPart: Harnessing Multi-Modal Large Language Models for Part-Aware 3D Generation
• Artificial Satellite Trails Detection Using U-Net Deep Neural Network and Line Segment Detector Algorithm
• Benchmarking and Mitigating MCQA Selection Bias of Large Vision-Language Models
• MedGS: Gaussian Splatting for Multi-Modal 3D Medical Imaging
• Looking in the mirror: A faithful counterfactual explanation method for interpreting deep image classification models
• L2M-Reg: Building-level Uncertainty-aware Registration of Outdoor LiDAR Point Clouds and Semantic 3D City Models
• ISCS: Parameter-Guided Channel Ordering and Grouping for Learned Image Compression
• ConfidentSplat: Confidence-Weighted Depth Fusion for Accurate 3D Gaussian Splatting SLAM
• FitPro: A Zero-Shot Framework for Interactive Text-based Pedestrian Retrieval in Open World
• Segment-to-Act: Label-Noise-Robust Action-Prompted Video Segmentation Towards Embodied Intelligence
• IPF-RDA: An Information-Preserving Framework for Robust Data Augmentation
• Active View Selection for Scene-level Multi-view Crowd Counting and Localization with Limited Labels
• Towards a Transparent and Interpretable AI Model for Medical Image Classifications
• Spectral Compressive Imaging via Chromaticity-Intensity Decomposition
• InstanceAssemble: Layout-Aware Image Generation via Instance Assembling Attention
• Animalbooth: multimodal feature enhancement for animal subject personalization
• When Confidence Fails: Revisiting Pseudo-Label Selection in Semi-supervised Semantic Segmentation
• Text-Scene: A Scene-to-Language Parsing Framework for 3D Scene Understanding
• Pain in 3D: Generating Controllable Synthetic Faces for Automated Pain Assessment
• A Novel Metric for Detecting Memorization in Generative Models for Brain MRI Synthesis
• Describe-to-Score: Text-Guided Efficient Image Complexity Assessment
• CGTGait: Collaborative Graph and Transformer for Gait Emotion Recognition
• Enhancing Scientific Visual Question Answering via Vision-Caption aware Supervised Fine-Tuning
• Follow-Your-Emoji-Faster: Towards Efficient, Fine-Controllable, and Expressive Freestyle Portrait Animation
• DA-Font: Few-Shot Font Generation via Dual-Attention Hybrid Integration
• Towards Anytime Retrieval: A Benchmark for Anytime Person Re-Identification
• Unlocking Hidden Potential in Point Cloud Networks with Attention-Guided Grouping-Feature Coordination
• ADVEDM:Fine-grained Adversarial Attack against VLM-based Embodied Agents
• Are VLMs Ready for Lane Topology Awareness in Autonomous Driving?
• MedCutMix: A Data-Centric Approach to Improve Radiology Vision-Language Pre-training with Disease Awareness
• Eye Gaze Tells You Where to Compute: Gaze-Driven Efficient VLMs
• Octree Latent Diffusion for Semantic 3D Scene Generation and Completion
• RLGF: Reinforcement Learning with Geometric Feedback for Autonomous Driving Video Generation
• CommonForms: A Large, Diverse Dataset for Form Field Detection
• OS-DiffVSR: Towards One-step Latent Diffusion Model for High-detailed Real-world Video Super-Resolution
• SlowFast-SCI: Slow-Fast Deep Unfolding Learning for Spectral Compressive Imaging
• FG-Attn: Leveraging Fine-Grained Sparsity In Diffusion Transformers
• PM25Vision: A Large-Scale Benchmark Dataset for Visual Estimation of Air Quality
• Efficient Rectified Flow for Image Fusion
• ST-GS: Vision-Based 3D Semantic Occupancy Prediction with Spatial-Temporal Gaussian Splatting
• Person Identification from Egocentric Human-Object Interactions using 3D Hand Pose
• Captioning for Text-Video Retrieval via Dual-Group Direct Preference Optimization
• OpenGVL - Benchmarking Visual Temporal Progress for Data Curation
• Mano Report
• Introducing Resizable Region Packing Problem in Image Generation, with a Heuristic Solution
• Accurate Thyroid Cancer Classification using a Novel Binary Pattern Driven Local Discrete Cosine Transform Descriptor
• StereoAdapter: Adapting Stereo Depth Estimation to Underwater Scenes
• 3D Gaussian Flats: Hybrid 2D/3D Photometric Scene Reconstruction
• TractoTransformer: Diffusion MRI Streamline Tractography using CNN and Transformer Networks
• Improved mmFormer for Liver Fibrosis Staging via Missing-Modality Compensation
• Explainable Gait Abnormality Detection Using Dual-Dataset CNN-LSTM Models
• Cross-Corpus and Cross-domain Handwriting Assessment of NeuroDegenerative Diseases via Time-Series-to-Image Conversion
• VaseVQA: Multimodal Agent and Benchmark for Ancient Greek Pottery
• AutoArabic: A Three-Stage Framework for Localizing Video-Text Retrieval Benchmarks
• Evaluating the Effectiveness and Scalability of LLM-Based Data Augmentation for Retrieval
• Purely Semantic Indexing for LLM-based Generative Recommendation and Retrieval
• Advancing Reference-free Evaluation of Video Captions with Factual Analysis
• Long document summarization using page specific target text alignment and distilling page importance
• The Role of Vocabularies in Learning Sparse Representations for Ranking
• Idiosyncratic Versus Normative Modeling of Atypical Speech Recognition: Dysarthric Case Studies
• SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?
• Localizing Malicious Outputs from CodeLLM
• SVeritas: Benchmark for Robust Speaker Verification under Diverse Conditions
• Bringing Pedagogy into Focus: Evaluating Virtual Teaching Assistants' Question-Answering in Asynchronous Learning Environments
• ReDepress: A Cognitive Framework for Detecting Depression Relapse from Social Media
• Variation in Verification: Understanding Verification Dynamics in Large Language Models
• WenetSpeech-Chuan: A Large-Scale Sichuanese Corpus with Rich Annotation for Dialectal Speech Processing
• Cross-Attention is Half Explanation in Speech-to-Text Models
• RadEval: A framework for radiology text evaluation
• The PIMMUR Principles: Ensuring Validity in Collective Behavior of LLM Societies
• TMD-TTS: A Unified Tibetan Multi-Dialect Text-to-Speech Synthesis for \"U-Tsang, Amdo and Kham Speech Dataset Generation
• ARK-V1: An LLM-Agent for Knowledge Graph Question Answering Requiring Commonsense Reasoning
• SEQR: Secure and Efficient QR-based LoRA Routing
• Longitudinal and Multimodal Recording System to Capture Real-World Patient-Clinician Conversations for AI and Encounter Research: Protocol
• Hierarchical Retrieval: The Geometry and a Pretrain-Finetune Recipe
• Make Every Letter Count: Building Dialect Variation Dictionaries from Monolingual Corpora
• CorPipe at CRAC 2025: Evaluating Multilingual Encoders for Multilingual Coreference Resolution
• Unsupervised Learning and Representation of Mandarin Tonal Categories by a Generative CNN
• How Persuasive is Your Context?
• SiDiaC: Sinhala Diachronic Corpus
• Improving Zero-shot Sentence Decontextualisation with Content Selection and Planning
• Transformer-Encoder Trees for Efficient Multilingual Machine Translation and Speech Translation
• Training-free Truthfulness Detection via Value Vectors in LLMs
• D-REX: A Benchmark for Detecting Deceptive Reasoning in Large Language Models
• HICode: Hierarchical Inductive Coding with LLMs
• Dorabella Cipher as Musical Inspiration
• Breaking Token Into Concepts: Exploring Extreme Compression in Token Representation Via Compositional Shared Semantics
• Qwen3-Omni Technical Report
• A State-Update Prompting Strategy for Efficient and Robust Multi-turn Dialogue
• DIVERS-Bench: Evaluating Language Identification Across Domain Shifts and Code-Switching
• One Agent to Serve All: a Lite-Adaptive Stylized AI Assistant for Millions of Multi-Style Official Accounts
• Learning to vary: Teaching LMs to reproduce human linguistic variability in next-word prediction
• Findings of the Fourth Shared Task on Multilingual Coreference Resolution: Can LLMs Dethrone Traditional Approaches?
• Everyday Physics in Korean Contexts: A Culturally Grounded Physical Reasoning Benchmark
• Towards Adaptive Context Management for Intelligent Conversational Question Answering
• Fine-Grained Detection of AI-Generated Text Using Sentence-Level Segmentation
• Trust Me, I Can Convince You: The Contextualized Argument Appraisal Framework
• Specification-Aware Machine Translation and Evaluation for Purpose Alignment
• Asking a Language Model for Diverse Responses
• MSCoRe: A Benchmark for Multi-Stage Collaborative Reasoning in LLM Agents
• AuditoryBench++: Can Language Models Understand Auditory Knowledge without Hearing?
• Crosslingual Optimized Metric for Translation Assessment of Indian Languages
• PG-CE: A Progressive Generation Dataset with Constraint Enhancement for Controllable Text Generation
• Turk-LettuceDetect: A Hallucination Detection Models for Turkish RAG Applications
• When TableQA Meets Noise: A Dual Denoising Framework for Complex Questions and Large-scale Tables
• TASO: Task-Aligned Sparse Optimization for Parameter-Efficient Model Adaptation
• Evaluating LLM-Generated Versus Human-Authored Responses in Role-Play Dialogues
• Investigating Bias: A Multilingual Pipeline for Generating, Solving, and Evaluating Math Problems with LLMs
• Filling in the Clinical Gaps in Benchmark: Case for HealthBench for the Japanese medical system
• Semantic Reformulation Entropy for Robust Hallucination Detection in QA Tasks
• SLAyiNG: Towards Queer Language Processing
• Codifying Natural Langauge Tasks
• PRINCIPLES: Synthetic Strategy Memory for Proactive Dialogue Agents
• Diagnosing Model Editing via Knowledge Spectrum
• AttnComp: Attention-Guided Adaptive Context Compression for Retrieval-Augmented Generation
• MapCoder-Lite: Squeezing Multi-Agent Coding into a Single Small LLM
• Enhancing Cross-Lingual Transfer through Reversible Transliteration: A Huffman-Based Approach for Low-Resource Languages
• CorefInst: Leveraging LLMs for Multilingual Coreference Resolution
• Leveraging Audio-Visual Data to Reduce the Multilingual Gap in Self-Supervised Speech Models
• Can LLMs Reason Over Non-Text Modalities in a Training-Free Manner? A Case Study with In-Context Representation Learning
• Better Late Than Never: Evaluation of Latency Metrics for Simultaneous Speech-to-Text Translation
• Scale-free Characteristics of Multilingual Legal Texts and the Limitations of LLMs
• Robustness of Neurosymbolic Reasoners on First-Order Logic Problems
• FinDebate: Multi-Agent Collaborative Intelligence for Financial Analysis
• EpiCache: Episodic KV Cache Management for Long Conversational Question Answering
• DIWALI - Diversity and Inclusivity aWare cuLture specific Items for India: Dataset and Assessment of LLMs for Cultural Text Adaptation in Indian Context
• Vision Language Models Are Not (Yet) Spelling Correctors
• RealBench: A Chinese Multi-image Understanding Benchmark Close to Real-world Scenarios
• QWHA: Quantization-Aware Walsh-Hadamard Adaptation for Parameter-Efficient Fine-Tuning on Large Language Models
• MedFact: A Large-scale Chinese Dataset for Evidence-based Medical Fact-checking of LLM Responses
• GeoPQA: Bridging the Visual Perception Gap in MLLMs for Geometric Reasoning
• SFT-TA: Supervised Fine-Tuned Agents in Multi-Agent LLMs for Automated Inductive Thematic Analysis
• FlagEval Findings Report: A Preliminary Evaluation of Large Reasoning Models on Automatically Verifiable Textual and Visual Questions
• Attention Consistency for LLMs Explanation
• LifeAlign: Lifelong Alignment for Large Language Models with Memory-Augmented Focalized Preference Optimization
• Evolution of Concepts in Language Model Pre-Training
• Prompt-Based Simplification for Plain Language using Spanish Language Models
• Extending Automatic Machine Translation Evaluation to Book-Length Documents
• Probabilistic Token Alignment for Large Language Model Fusion
• Automated Knowledge Graph Construction using Large Language Models and Sentence Complexity Modelling
• Multi-View Attention Multiple-Instance Learning Enhanced by LLM Reasoning for Cognitive Distortion Detection
• Scaling, Simplification, and Adaptation: Lessons from Pretraining on Machine-Translated Text
• AIMMerging: Adaptive Iterative Model Merging Using Training Trajectories for Language Model Continual Learning
• Can GRPO Boost Complex Multimodal Table Understanding?
• CLaC at DISRPT 2025: Hierarchical Adapters for Cross-Framework Multi-lingual Discourse Relation Classification
• CUTE: A Multilingual Dataset for Enhancing Cross-Lingual Knowledge Transfer in Low-Resource Languages
• K-DeCore: Facilitating Knowledge Transfer in Continual Structured Knowledge Reasoning via Knowledge Decoupling
• AirQA: A Comprehensive QA Dataset for AI Research with Instance-Level Evaluation
• Preference Distillation via Value based Reinforcement Learning
• Advancing Speech Understanding in Speech-Aware Language Models with GRPO
• The Transfer Neurons Hypothesis: An Underlying Mechanism for Language Latent Space Transitions in Multilingual LLMs
• Modeling Bottom-up Information Quality during Language Processing
• TactfulToM: Do LLMs Have the Theory of Mind Ability to Understand White Lies?
• OPEN-THEATRE: An Open-Source Toolkit for LLM-based Interactive Drama
• Semi-Supervised Synthetic Data Generation with Fine-Grained Relevance Control for Short Video Search Relevance Modeling
• Time to Revist Exact Match
• A Multi-Level Benchmark for Causal Language Understanding in Social Media Discourse
• Angular Dispersion Accelerates $k$-Nearest Neighbors Machine Translation
• The Sound of Syntax: Finetuning and Comprehensive Evaluation of Language Models for Speech Pathology
• MoRoVoc: A Large Dataset for Geographical Variation Identification of the Spoken Romanian Language
• Domain-Adaptive Pre-Training for Arabic Aspect-Based Sentiment Analysis: A Comparative Study of Domain Adaptation and Fine-Tuning Strategies
• KuBERT: Central Kurdish BERT Model and Its Application for Sentiment Analysis
• Cognitive Linguistic Identity Fusion Score (CLIFS): A Scalable Cognition-Informed Approach to Quantifying Identity Fusion from Text
• Semantic-Driven Topic Modeling for Analyzing Creativity in Virtual Brainstorming
• Multi-task Pretraining for Enhancing Interpretable L2 Pronunciation Assessment
• ChemOrch: Empowering LLMs with Chemical Intelligence via Synthetic Instructions
• MPCG: Multi-Round Persona-Conditioned Generation for Modeling the Evolution of Misinformation with LLMs
• From Uniform to Heterogeneous: Tailoring Policy Optimization to Every Token's Nature
• MCP: A Control-Theoretic Orchestration Framework for Synergistic Efficiency and Interpretability in Multimodal Large Language Models
• Computational-Assisted Systematic Review and Meta-Analysis (CASMA): Effect of a Subclass of GnRH-a on Endometriosis Recurrence
• Redefining Experts: Interpretable Decomposition of Language Models for Toxicity Mitigation
• Robust Native Language Identification through Agentic Decomposition
• Reinforcement Learning Meets Large Language Models: A Survey of Advancements and Applications Across the LLM Lifecycle
• EG-MLA: Embedding-Gated Multi-head Latent Attention for Scalable and Efficient LLMs
• Decoding Uncertainty: The Impact of Decoding Strategies for Uncertainty Estimation in Large Language Models
• HARE: an entity and relation centric evaluation framework for histopathology reports
• RephQA: Evaluating Readability of Large Language Models in Public Health Question Answering
• Whisper-UT: A Unified Translation Framework for Speech and Text
• 'Rich Dad, Poor Lad': How do Large Language Models Contextualize Socioeconomic Factors in College Admission ?
• Evaluating CxG Generalisation in LLMs via Construction-Based NLI Fine Tuning
• Intrinsic Meets Extrinsic Fairness: Assessing the Downstream Impact of Bias Mitigation in Large Language Models
• Computational Analysis of Conversation Dynamics through Participant Responsivity
• Leveraging Multilingual Training for Authorship Representation: Enhancing Generalization across Languages and Domains
• Challenging the Evaluator: LLM Sycophancy Under User Rebuttal
• Mental Multi-class Classification on Social Media: Benchmarking Transformer Architectures against LSTM Models
• Evaluation of Ensemble Learning Techniques for handwritten OCR Improvement
• Predicting First Year Dropout from Pre Enrolment Motivation Statements Using Text Mining
• Machine Learning for Quantum Noise Reduction
• How Can Quantum Deep Learning Improve Large Language Models?
• Motional representation; the ability to predict odor characters using molecular vibrations
• GraphMend: Code Transformations for Fixing Graph Breaks in PyTorch 2
• Vibrational Fingerprints of Strained Polymers: A Spectroscopic Pathway to Mechanical State Prediction
• Language Modeling with Learned Meta-Tokens
• Test-Time Learning and Inference-Time Deliberation for Efficiency-First Offline Reinforcement Learning in Care Coordination and Population Health Management
• TF-DWGNet: A Directed Weighted Graph Neural Network with Tensor Fusion for Multi-Omics Cancer Subtype Classification
• Neural Atlas Graphs for Dynamic Scene Decomposition and Editing
• Similarity-Guided Diffusion for Long-Gap Music Inpainting
• Equilibrium flow: From Snapshots to Dynamics
• Adaptive Kernel Design for Bayesian Optimization Is a Piece of CAKE with LLMs
• Unveiling m-Sharpness Through the Structure of Stochastic Gradient Noise
• Control Disturbance Rejection in Neural ODEs
• Reinforced Generation of Combinatorial Structures: Applications to Complexity Theory
• Strategic Dishonesty Can Undermine AI Safety Evaluations of Frontier LLM
• Learning to Rank with Top-$K$ Fairness
• Learning functions, operators and dynamical systems with kernels
• Spiffy: Multiplying Diffusion LLM Acceleration via Lossless Speculative Decoding
• Deep Reinforcement Learning in Factor Investment
• On the Detection of Internal Defects in Structured Media
• Understanding Post-Training Structural Changes in Large Language Models
• Improving After-sales Service: Deep Reinforcement Learning for Dynamic Time Slot Assignment with Commitments and Customer Preferences
• Deep Hierarchical Learning with Nested Subspace Networks
• Confidence-gated training for efficient early-exit neural networks
• GaussianPSL: A novel framework based on Gaussian Splatting for exploring the Pareto frontier in multi-criteria optimization
• Optimizing Inference in Transformer-Based Models: A Multi-Method Benchmark
• SingLEM: Single-Channel Large EEG Model
• Medical priority fusion: achieving dual optimization of sensitivity and interpretability in nipt anomaly detection
• StefaLand: An Efficient Geoscience Foundation Model That Improves Dynamic Land-Surface Predictions
• Joint Optimization of Memory Frequency, Computing Frequency, Transmission Power and Task Offloading for Energy-efficient DNN Inference
• Intra-Cluster Mixup: An Effective Data Augmentation Technique for Complementary-Label Learning
• Budgeted Adversarial Attack against Graph-Based Anomaly Detection in Sensor Networks
• An AutoML Framework using AutoGluonTS for Forecasting Seasonal Extreme Temperatures
• Flatness is Necessary, Neural Collapse is Not: Rethinking Generalization via Grokking
• GEM-T: Generative Tabular Data via Fitting Moments
• Learning Neural Antiderivatives
• Revealing Multimodal Causality with Large Language Models
• Elucidating the Design Space of FP4 training
• Remote Sensing-Oriented World Model
• MTM: A Multi-Scale Token Mixing Transformer for Irregular Multivariate Time Series Classification
• MSGAT-GRU: A Multi-Scale Graph Attention and Recurrent Model for Spatiotemporal Road Accident Prediction
• Global Optimization via Softmin Energy Minimization
• Conv-like Scale-Fusion Time Series Transformer: A Multi-Scale Representation for Variable-Length Long Time Series
• BiLCNet : BiLSTM-Conformer Network for Encrypted Traffic Classification with 5G SA Physical Channel Records
• Achilles' Heel of Mamba: Essential difficulties of the Mamba architecture demonstrated by synthetic data
• An Unlearning Framework for Continual Learning
• SeqBattNet: A Discrete-State Physics-Informed Neural Network with Aging Adaptation for Battery Modeling
• Comparing Data Assimilation and Likelihood-Based Inference on Latent State Estimation in Agent-Based Models
• Mechanistic Interpretability with SAEs: Probing Religion, Violence, and Geography in Large Language Models
• Fast, Accurate and Interpretable Graph Classification with Topological Kernels
• Cluster Workload Allocation: A Predictive Approach Leveraging Machine Learning Efficiency
• A non-smooth regularization framework for learning over multitask graphs
• A Generative Conditional Distribution Equality Testing Framework and Its Minimax Analysis
• ConfClip: Confidence-Weighted and Clipped Reward for Reinforcement Learning in LLMs
• Training the next generation of physicians for artificial intelligence-assisted clinical neuroradiology: ASNR MICCAI Brain Tumor Segmentation (BraTS) 2025 Lighthouse Challenge education platform
• GraphWeave: Interpretable and Robust Graph Generation via Random Walk Trajectories
• Physics-Informed Operator Learning for Hemodynamic Modeling
• SPRINT: Stochastic Performative Prediction With Variance Reduction
• VQEzy: An Open-Source Dataset for Parameter Initialize in Variational Quantum Eigensolvers
• Generalizable End-to-End Tool-Use RL with Synthetic CodeGym
• Robust Anomaly Detection Under Normality Distribution Shift in Dynamic Graphs
• Efficient Sliced Wasserstein Distance Computation via Adaptive Bayesian Optimization
• Distributionally Robust Safety Verification of Neural Networks via Worst-Case CVaR
• MVCL-DAF++: Enhancing Multimodal Intent Recognition via Prototype-Aware Contrastive Alignment and Coarse-to-Fine Dynamic Attention Fusion
• Periodic Graph-Enhanced Multivariate Time Series Anomaly Detector
• Path-Weighted Integrated Gradients for Interpretable Dementia Classification
• A Comprehensive Performance Comparison of Traditional and Ensemble Machine Learning Models for Online Fraud Detection
• Regularizing Extrapolation in Causal Inference
• PMRT: A Training Recipe for Fast, 3D High-Resolution Aerodynamic Prediction
• Dendritic Resonate-and-Fire Neuron for Effective and Efficient Long Sequence Modeling
• SignalLLM: A General-Purpose LLM Agent Framework for Automated Signal Processing
• Conditional Policy Generator for Dynamic Constraint Satisfaction and Optimization
• Active Learning for Machine Learning Driven Molecular Dynamics
• Causal Representation Learning from Multimodal Clinical Records under Non-Random Modality Missingness
• Prospective Multi-Graph Cohesion for Multivariate Time Series Anomaly Detection
• TraceHiding: Scalable Machine Unlearning for Mobility Data
• Graph Signal Generative Diffusion Models
• Enhancing Performance and Calibration in Quantile Hyperparameter Optimization
• TSGym: Design Choices for Deep Multivariate Time-Series Forecasting
• On the Limits of Tabular Hardness Metrics for Deep RL: A Study with the Pharos Benchmark
• Ultra-short-term solar power forecasting by deep learning and data reconstruction
• GRPOformer: Advancing Hyperparameter Optimization via Group Relative Policy Optimization
• ScenGAN: Attention-Intensive Generative Model for Uncertainty-Aware Renewable Scenario Forecasting
• On the Simplification of Neural Network Architectures for Predictive Process Monitoring
• Flow-Induced Diagonal Gaussian Processes
• Unrolled Graph Neural Networks for Constrained Optimization
• Time Series Forecasting Using a Hybrid Deep Learning Method: A Bi-LSTM Embedding Denoising Auto Encoder Transformer
• Detecting Urban PM$_{2.5}$ Hotspots with Mobile Sensing and Gaussian Process Regression
• Dynamic Expert Specialization: Towards Catastrophic Forgetting-Free Multi-Domain MoE Adaptation
• DRES: Fake news detection by dynamic representation and ensemble selection
• The Complexity of Finding Local Optima in Contrastive Learning
• FedEL: Federated Elastic Learning for Heterogeneous Devices
• Auditability and the Landscape of Distance to Multicalibration
• Adaptive Graph Convolution and Semantic-Guided Attention for Multimodal Risk Detection in Social Networks
• Gradient Interference-Aware Graph Coloring for Multitask Learning
• PTQTP: Post-Training Quantization to Trit-Planes for Large Language Models
• Persistence Spheres: Bi-continuous Representations of Persistence Diagrams
• Adaptive Overclocking: Dynamic Control of Thinking Path Length via Real-Time Reasoning Signals
• Long-Tailed Out-of-Distribution Detection with Refined Separate Class Learning
• $\boldsymbol{\lambda}$-Orthogonality Regularization for Compatible Representation Learning
• HypeMARL: Multi-Agent Reinforcement Learning For High-Dimensional, Parametric, and Distributed Systems
• A Hybrid PCA-PR-Seq2Seq-Adam-LSTM Framework for Time-Series Power Outage Prediction
• Interpretable Clinical Classification with Kolgomorov-Arnold Networks
• Discrete Diffusion Models: Novel Analysis and New Sampler Guarantees
• Geometric Mixture Classifier (GMC): A Discriminative Per-Class Mixture of Hyperplanes
• DISCO: Disentangled Communication Steering for Large Language Models
• KANO: Kolmogorov-Arnold Neural Operator
• SOLAR: Switchable Output Layer for Accuracy and Robustness in Once-for-All Training
• LVADNet3D: A Deep Autoencoder for Reconstructing 3D Intraventricular Flow from Sparse Hemodynamic Data
• Towards Interpretable and Efficient Attention: Compressing All by Contracting a Few
• A Closer Look at Model Collapse: From a Generalization-to-Memorization Perspective
• GRIL: Knowledge Graph Retrieval-Integrated Learning with Large Language Models
• Federated Learning with Ad-hoc Adapter Insertions: The Case of Soft-Embeddings for Training Classifier-as-Retriever
• LLM-Guided Co-Training for Text Classification
• mmExpert: Integrating Large Language Models for Comprehensive mmWave Data Synthesis and Understanding
• SCAN: Self-Denoising Monte Carlo Annotation for Robust Process Reward Learning
• ViTCAE: ViT-based Class-conditioned Autoencoder
• Learned Digital Codes for Over-the-Air Federated Learning
• Near-Optimal Sample Complexity Bounds for Constrained Average-Reward MDPs
• Self-Supervised Learning of Graph Representations for Network Intrusion Detection
• Causality-Induced Positional Encoding for Transformer-Based Representation Learning of Non-Sequential Features
• FairTune: A Bias-Aware Fine-Tuning Framework Towards Fair Heart Rate Prediction from PPG
• Comparison of Deterministic and Probabilistic Machine Learning Algorithms for Precise Dimensional Control and Uncertainty Quantification in Additive Manufacturing
• Architectural change in neural networks using fuzzy vertex pooling
• ROOT: Rethinking Offline Optimization as Distributional Translation via Probabilistic Bridge
• Auto-bidding under Return-on-Spend Constraints with Uncertainty Quantification
• Improving Deep Tabular Learning
• Guided Sequence-Structure Generative Modeling for Iterative Antibody Optimization
• EMPEROR: Efficient Moment-Preserving Representation of Distributions
• Federated Learning for Financial Forecasting
• Local Mechanisms of Compositional Generalization in Conditional Diffusion
• Towards Universal Debiasing for Language Models-based Tabular Data Generation
• Revisiting Broken Windows Theory
• Detection and Simulation of Urban Heat Islands Using a Fine-Tuned Geospatial Foundation Model
• Surgical-MambaLLM: Mamba2-enhanced Multimodal Large Language Model for VQLA in Robotic Surgery
• Audio-Conditioned Diffusion LLMs for ASR and Deliberation Processing
• When Big Models Train Small Ones: Label-Free Model Parity Alignment for Efficient Visual Question Answering using Small VLMs
• KungfuBot2: Learning Versatile Motion Skills for Humanoid Whole-Body Control
• AISTAT lab system for DCASE2025 Task6: Language-based audio retrieval
• On the de-duplication of the Lakh MIDI dataset
• Governed By Agents: A Survey On The Role Of Agentic AI In Future Computing Environments
• ProtoVQA: An Adaptable Prototypical Framework for Explainable Fine-Grained Visual Question Answering
• Design and Development of an Intelligent LLM-based LDAP Honeypot
• InteGround: On the Evaluation of Verification and Retrieval Planning in Integrative Grounding
• Train to Defend: First Defense Against Cryptanalytic Neural Network Parameter Extraction Attacks
• TranTac: Leveraging Transient Tactile Signals for Contact-Rich Robotic Manipulation
• Rethinking the Role of Text Complexity in Language Model Pretraining
• V-CECE: Visual Counterfactual Explanations via Conceptual Edits
• From Scores to Steps: Diagnosing and Improving LLM Performance in Evidence-Based Medical Calculations
• SQS: Enhancing Sparse Perception Models via Query-based Splatting in Autonomous Driving
• Benchmarking Contextual and Paralinguistic Reasoning in Speech-LLMs: A Case Study with In-the-Wild Data
• Analyzing the Effects of Supervised Fine-Tuning on Model Knowledge from Token and Parameter Levels
• PruneCD: Contrasting Pruned Self Model to Improve Decoding Factuality
• FakeChain: Exposing Shallow Cues in Multi-Step Deepfake Detection
• Implicit Behavioral Alignment of Language Agents in High-Stakes Crowd Simulations
• Entropic Causal Inference: Graph Identifiability
• Thermal Imaging-based Real-time Fall Detection using Motion Flow and Attention-enhanced Convolutional Recurrent Architecture
• The Oracle Has Spoken: A Multi-Aspect Evaluation of Dialogue in Pythia
• Can an Individual Manipulate the Collective Decisions of Multi-Agents?
• Synergies between Federated Foundation Models and Smart Power Grids
• Seeing Culture: A Benchmark for Visual Reasoning and Grounding
• Causal Fuzzing for Verifying Machine Unlearning
• Lattice Boltzmann Model for Learning Real-World Pixel Dynamicity
• AIPsychoBench: Understanding the Psychometric Differences between LLMs and Humans
• No Need for Real 3D: Fusing 2D Vision with Pseudo 3D Representations for Robotic Manipulation Learning
• Enhancing Financial RAG with Agentic AI and Multi-HyDE: A Novel Approach to Knowledge Retrieval and Hallucination Reduction
• CoUn: Empowering Machine Unlearning via Contrastive Learning
• Evaluating Behavioral Alignment in Conflict Dialogue: A Multi-Dimensional Comparison of LLM Agents and Humans
• GRID: Graph-based Reasoning for Intervention and Discovery in Built Environments
• Pico: A Modular Framework for Hypothesis-Driven Small Language Model Research
• LenslessMic: Audio Encryption and Authentication via Lensless Computational Imaging
• AHA -- Predicting What Matters Next: Online Highlight Detection Without Looking Ahead
• SENSE-7: Taxonomy and Dataset for Measuring User Perceptions of Empathy in Sustained Human-AI Conversations
• LightCode: Compiling LLM Inference for Photonic-Electronic Systems
• PersonaMatrix: A Recipe for Persona-Aware Evaluation of Legal Summarization
• KRAST: Knowledge-Augmented Robotic Action Recognition with Structured Text for Vision-Language Models
• A Generative AI System for Biomedical Data Discovery with Grammar-Based Visualizations
• Energy Equity, Infrastructure and Demographic Analysis with XAI Methods
• Robust LLM Training Infrastructure at ByteDance
• Patterns in the Transition From Founder-Leadership to Community Governance of Open Source
• How Large Language Models are Designed to Hallucinate
• Overhearing LLM Agents: A Survey, Taxonomy, and Roadmap
• Highly Imbalanced Regression with Tabular Data in SEP and Other Applications
• Agentic Reasoning for Robust Vision Systems via Increased Test-Time Compute
• Estimating Clinical Lab Test Result Trajectories from PPG using Physiological Foundation Model and Patient-Aware State Space Model -- a UNIPHY+ Approach
• From Canopy to Ground via ForestGen3D: Learning Cross-Domain Generation of 3D Forest Structure from Aerial-to-Terrestrial LiDAR
• QUINTA: Reflexive Sensibility For Responsible AI Research and Data-Driven Processes
• Secure Confidential Business Information When Sharing Machine Learning Models
• A study on Deep Convolutional Neural Networks, transfer learning, and Mnet model for Cervical Cancer Detection
• R-Net: A Reliable and Resource-Efficient CNN for Colorectal Cancer Detection with XAI Integration
• Imaging Modalities-Based Classification for Lung Cancer Detection
• HausaMovieReview: A Benchmark Dataset for Sentiment Analysis in Low-Resource African Language
• Socratic Mind: Impact of a Novel GenAI-Powered Assessment Tool on Student Learning and Higher-Order Thinking
• Gender and Political Bias in Large Language Models: A Demonstration Platform
• Digging Into the Internal: Causality-Based Analysis of LLM Function Calling
• SubDyve: Subgraph-Driven Dynamic Propagation for Virtual Screening Enhancement Controlling False Positive
• SecureFixAgent: A Hybrid LLM Agent for Automated Python Static Vulnerability Repair
• Comparative Analysis of STEM and non-STEM Teachers' Needs for Integrating AI into Educational Environments
• Stabilizing Information Flow Entropy: Regularization for Safe and Interpretable Autonomous Driving Perception
• "I think this is fair'': Uncovering the Complexities of Stakeholder Decision-Making in AI Fairness Assessment
• On the Variational Costs of Changing Our Minds
• The STAR-XAI Protocol: An Interactive Framework for Inducing Second-Order Agency in AI Agents
• Improving Large Language Models Function Calling and Interpretability via Guided-Structured Templates
• Reasoning Core: A Scalable RL Environment for LLM Symbolic Reasoning
• Breast Cancer Classification Using Gradient Boosting Algorithms Focusing on Reducing the False Negative and SHAP for Explainability
• EPIC: Generative AI Platform for Accelerating HPC Operational Data Analytics
• DarwinWafer: A Wafer-Scale Neuromorphic Chip
• Discovering Software Parallelization Points Using Deep Neural Networks
• On LLM-Based Scientific Inductive Reasoning Beyond Equations
• REAMS: Reasoning Enhanced Algorithm for Maths Solving
• Is It Certainly a Deepfake? Reliability Analysis in Detection & Generation Ecosystem
• MontePrep: Monte-Carlo-Driven Automatic Data Preparation without Target Data Instances
• LIMI: Less is More for Agency
• Table2LaTeX-RL: High-Fidelity LaTeX Code Generation from Table Images via Reinforced Multimodal Language Models
• EngiBench: A Benchmark for Evaluating Large Language Models on Engineering Problem Solving
• Virtual Arc Consistency for Linear Constraints inCost Function Networks
• DA-Mamba: Dialogue-aware selective state-space model for multimodal engagement estimation
• Efficient & Correct Predictive Equivalence for Decision Trees
• Mitigating Strategy-Selection Bias in Reasoning for More Effective Test-Time Scaling
• MEF: A Systematic Evaluation Framework for Text-to-Image Models
• Orcust: Stepwise-Feedback Reinforcement Learning for GUI Agent
• Can Agents Judge Systematic Reviews Like Humans? Evaluating SLRs with LLM-based Multi-Agent System
• Mind the Gap: Comparing Model- vs Agentic-Level Red Teaming with Action-Graph Observability on GPT-OSS-20B
• CogAtom: From Cognitive Atoms to Olympiad-level Mathematical Reasoning in Large Language Models
• LLaVul: A Multimodal LLM for Interpretable Vulnerability Reasoning about Source Code
• Medical AI Consensus: A Multi-Agent Framework for Radiology Report Generation and Evaluation
• Multi-Scenario Highway Lane-Change Intention Prediction: A Physics-Informed AI Framework for Three-Class Classification
• Correlation or Causation: Analyzing the Causal Structures of LLM and LRM Reasoning Process
• Program Synthesis via Test-Time Transduction
• Evaluating Multimodal Large Language Models with Daily Composite Tasks in Home Environments
• SPICED: A Synaptic Homeostasis-Inspired Framework for Unsupervised Continual EEG Decoding
• AI Pangaea: Unifying Intelligence Islands for Adapting Myriad Tasks
• A Multimodal Conversational Assistant for the Characterization of Agricultural Plots from Geospatial Open Data
• Audio-Guided Dynamic Modality Fusion with Stereo-Aware Attention for Audio-Visual Navigation
• Quantum Abduction: A New Paradigm for Reasoning under Uncertainty
• KAHAN: Knowledge-Augmented Hierarchical Analysis and Narration for Financial Data Narration
• From domain-landmark graph learning to problem-landmark graph generation
• RALLM-POI: Retrieval-Augmented LLM for Zero-shot Next POI Recommendation with Geographical Reranking
• Intention-aware Hierarchical Diffusion Model for Long-term Trajectory Anomaly Detection
• Governing Automated Strategic Intelligence
• MCTS-EP: Empowering Embodied Planning with Online Preference Optimization
• ARE: Scaling Up Agent Environments and Evaluations
• Shall We Play a Game? Language Models for Open-ended Wargames
• MoEs Are Stronger than You Think: Hyper-Parallel Inference Scaling with RoE
• Question Answering with LLMs and Learning from Answer Sets
• FESTA: Functionally Equivalent Sampling for Trust Assessment of Multimodal LLMs
• NUMINA: A Natural Understanding Benchmark for Multi-dimensional Intelligence and Numerical Reasoning Abilities
• Sycophancy Mitigation Through Reinforcement Learning with Uncertainty-Aware Adaptive Reasoning Trajectories
• Automated Procedural Analysis via Video-Language Models for AI-assisted Nursing Skills Assessment
• Prompt-Driven Agentic Video Editing System: Autonomous Comprehension of Long-Form, Story-Driven Media
• Roundtable Policy: Improving Scientific Reasoning and Narratives through Confidence-Weighted Consensus of LLMs
• The Principles of Human-like Conscious Machine
• Large Language Models as End-to-end Combinatorial Optimization Solvers
• seqBench: A Tunable Benchmark to Quantify Sequential Reasoning Limits of LLMs
• LLMs as Layout Designers: A Spatial Reasoning Perspective
• On the Non-Uniqueness of Representation of $(U,N)$-Implications
• Psychometric Personality Shaping Modulates Capabilities and Safety in Language Models
• A Unified AI Approach for Continuous Monitoring of Human Health and Diseases from Intensive Care Unit to Home with Physiological Foundation Models (UNIPHY+)
• Evaluation of Causal Reasoning for Large Language Models in Contextualized Clinical Scenarios of Laboratory Test Interpretation
• VORTEX: Aligning Task Utility and Human Preferences through LLM-Guided Reward Shaping
• Proactive Statistical Process Control Using AI: A Time Series Forecasting Approach for Semiconductor Manufacturing
• Domain-Specific Constitutional AI: Enhancing Safety in LLM-Powered Mental Health Chatbots
• GPO: Learning from Critical Steps to Improve LLM Reasoning
• Checking extracted rules in Neural Networks
• SalaMAnder: Shapley-based Mathematical Expression Attribution and Metric for Chain-of-Thought Reasoning
• Zero-Shot Human Mobility Forecasting via Large Language Model with Hierarchical Reasoning
• Identifying Critical Pathways in Coronary Heart Disease via Fuzzy Subgraph Connectivity
• A global view of diverse construction methods of fuzzy implication functions rooted on F-chains