AI Research News Feeds for October 24th, 2025

AI RESEARCH PAPERS & ACADEMIC SOURCES

• GenLit: Reformulating Single-Image Relighting as Video Generation
• Towards Physical Understanding in Video Generation: A 3D Point Regularization Approach
• BevSplat: Resolving Height Ambiguity via Feature-Based Gaussian Primitives for Weakly-Supervised Cross-View Localization
• 8-Calves Image dataset
• Panoptic-CUDAL: Rural Australia Point Cloud Dataset in Rainy Conditions
• ControlFusion: A Controllable Image Fusion Framework with Language-Vision Degradation Prompts
• FreeGraftor: Training-Free Cross-Image Feature Grafting for Subject-Driven Text-to-Image Generation
• Learning Dense Hand Contact Estimation from Imbalanced Data
• Rebalancing Contrastive Alignment with Bottlenecked Semantic Increments in Text-Video Retrieval
• Comprehensive Evaluation and Analysis for NSFW Concept Erasure in Text-to-Image Diffusion Models
• Mesh-RFT: Enhancing Mesh Generation via Fine-grained Reinforcement Fine-Tuning
• REOBench: Benchmarking Robustness of Earth Observation Foundation Models
• MODEM: A Morton-Order Degradation Estimation Mechanism for Adverse Weather Image Recovery
• SeG-SR: Integrating Semantic Knowledge into Remote Sensing Image Super-Resolution via Vision-Language Model
• PreFM: Online Audio-Visual Event Parsing via Predictive Future Modeling
• Revisiting End-to-End Learning with Slide-level Supervision in Computational Pathology
• PlantSegNeRF: A few-shot, cross-species method for plant 3D instance point cloud reconstruction via joint-channel NeRF with multi-view image instance matching
• OpenWorldSAM: Extending SAM2 for Universal Image Segmentation with Language Prompts
• SnapMoGen: Human Motion Generation from Expressive Texts
• Novel Class Discovery for Point Cloud Segmentation via Joint Learning of Causal Representation and Reasoning
• Spatial-DISE: A Unified Benchmark for Evaluating Spatial Reasoning in Vision-Language Models
• X-Reflect: Cross-Reflection Prompting for Multimodal Recommendation
• A primal-dual algorithm for image reconstruction with input-convex neural network regularizers
• Generative diffusion model surrogates for mechanistic agent-based biological models
• A novel attention mechanism for noise-adaptive and robust segmentation of microtubules in microscopy images
• MoORE: SVD-based Model MoE-ization for Conflict- and Oblivion-Resistant Multi-Task Adaptation
• DeepCausalMMM: A Deep Learning Framework for Marketing Mix Modeling with Causal Inference
• Near optimal sample complexity for matrix and tensor normal models via geodesic convexity
• Field theory for optimal signal propagation in ResNets
• Causal Post-Processing of Predictive Models
• ROTI-GCV: Generalized Cross-Validation for right-ROTationally Invariant Data
• Tex-ViT: A Generalizable, Robust, Texture-based dual-branch cross-attention deepfake detector
• Unity is Power: Semi-Asynchronous Collaborative Training of Large-Scale Models with Structured Pruning in Resource-Limited Clients
• JAMUN: Bridging Smoothed Molecular Dynamics and Score-Based Learning for Conformational Ensembles
• Stochastic gradient descent in high dimensions for multi-spiked tensor PCA
• Iterative Self-Tuning LLMs for Enhanced Jailbreaking Capabilities
• Prognostic Framework for Robotic Manipulators Operating Under Dynamic Task Severities
• Deep Continuous-Time State-Space Models for Marked Event Sequences
• Sampling from multi-modal distributions with polynomial query complexity in fixed dimension via reverse diffusion
• Statistical Inference for Generative Model Comparison
• CoCoA Is ADMM: Unifying Two Paradigms in Distributed Optimization
• IRIS: An Immersive Robot Interaction System
• Quantum speedup of non-linear Monte Carlo problems
• Sample-efficient Learning of Concepts with Theoretical Guarantees: from Data to Concepts without Interventions
• SMRS: advocating a unified reporting standard for surrogate models in the artificial intelligence era
• Multifidelity Simulation-based Inference for Computationally Expensive Simulators
• OpenMIBOOD: Open Medical Imaging Benchmarks for Out-Of-Distribution Detection
• Sharp Gaussian approximations for Decentralized Federated Learning
• Transfer Faster, Price Smarter: Minimax Dynamic Pricing under Cross-Market Preference Shift
• On the Emergence of Linear Analogies in Word Embeddings
• A decomposition-based robust training of physics-informed neural networks for nearly incompressible linear elasticity
• Sherlock: Self-Correcting Reasoning in Vision-Language Models
• BioCLIP 2: Emergent Properties from Scaling Hierarchical Contrastive Learning
• Proxy Target: Bridging the Gap Between Discrete Spiking Neural Networks and Continuous Control
• Quantitative LLM Judges
• MVP-Shapley: Feature-based Modeling for Evaluating the Most Valuable Player in Basketball
• SafeDiver: Cooperative AUV-USV Assisted Diver Communication via Multi-agent Reinforcement Learning Approach
• Behavioral Biometrics for Automatic Detection of User Familiarity in VR
• Fourier-Based GAN Fingerprint Detection using ResNet50
• Transformed Multi-view 3D Shape Features with Contrastive Learning
• FutrTrack: A Camera-LiDAR Fusion Transformer for 3D Multiple Object Tracking
• A Unified Detection Pipeline for Robust Object Detection in Fisheye-Based Traffic Surveillance
• Extreme Views: 3DGS Filter for Novel View Synthesis from Out-of-Distribution Camera Poses
• BrainPuzzle: Hybrid Physics and Data-Driven Reconstruction for Transcranial Ultrasound Tomography
• Exposing Blindspots: Cultural Bias Evaluation in Generative Image Models
• Filter-Based Reconstruction of Images from Events
• Data-Adaptive Transformed Bilateral Tensor Low-Rank Representation for Clustering
• Endoshare: A Source Available Solution to De-Identify and Manage Surgical Videos
• Attentive Convolution: Unifying the Expressivity of Self-Attention with Convolutional Efficiency
• Physics-Guided Fusion for Robust 3D Tracking of Fast Moving Small Objects
• Inverse Image-Based Rendering for Light Field Generation from Single Images
• Revisiting Logit Distributions for Reliable Out-of-Distribution Detection
• PartNeXt: A Next-Generation Dataset for Fine-Grained and Hierarchical 3D Part Understanding
• Monocular Visual 8D Pose Estimation for Articulated Bicycles and Cyclists
• TOMCAT: Test-time Comprehensive Knowledge Accumulation for Compositional Zero-Shot Learning
• Evaluating Video Models as Simulators of Multi-Person Pedestrian Trajectories
• SPAN: Continuous Modeling of Suspicion Progression for Temporal Intention Localization
• A Structured Review and Quantitative Profiling of Public Brain MRI Datasets for Foundation Model Development
• RAPO++: Cross-Stage Prompt Optimization for Text-to-Video Generation via Data Alignment and Test-Time Scaling
• FlowCycle: Pursuing Cycle-Consistent Flows for Text-based Editing
• Towards Objective Obstetric Ultrasound Assessment: Contrastive Representation Learning for Fetal Movement Detection
• EditInfinity: Image Editing with Binary-Quantized Generative Models
• COS3D: Collaborative Open-Vocabulary 3D Segmentation
• Seeing the Unseen: Mask-Driven Positional Encoding and Strip-Convolution Context Modeling for Cross-View Object Geo-Localization
• Real-Time Currency Detection and Voice Feedback for Visually Impaired Individuals
• GMFVAD: Using Grained Multi-modal Feature to Improve Video Anomaly Detection
• Causal Debiasing for Visual Commonsense Reasoning
• Knowledge-Informed Neural Network for Complex-Valued SAR Image Recognition
• DMC$^3$: Dual-Modal Counterfactual Contrastive Construction for Egocentric Video Question Answering
• HyperET: Efficient Training in Hyperbolic Space for Multi-modal Large Language Models
• AnyPcc: Compressing Any Point Cloud with a Single Universal Model
• AccuQuant: Simulating Multiple Denoising Steps for Quantizing Diffusion Models
• Positional Encoding Field
• Mitigating Cross-modal Representation Bias for Multicultural Image-to-Recipe Retrieval
• Conan: Progressive Learning to Reason Like a Detective over Multi-Scale Visual Evidence
• Reliable and Reproducible Demographic Inference for Fairness in Face Analysis
• EchoDistill: Bidirectional Concept Distillation for One-Step Diffusion Personalization
• Deep Learning-Powered Visual SLAM Aimed at Assisting Visually Impaired Navigation
• From Cheap to Pro: A Learning-based Adaptive Camera Parameter Network for Professional-Style Imaging
• From Far and Near: Perceptual Evaluation of Crowd Representations Across Levels of Detail
• EmbodiedBrain: Expanding Performance Boundaries of Task Planning for Embodied Intelligence
• GenColorBench: A Color Evaluation Benchmark for Text-to-Image Generation Models
• SeViCES: Unifying Semantic-Visual Evidence Consensus for Long Video Understanding
• Better Tokens for Better 3D: Advancing Vision-Language Modeling in 3D Medical Imaging
• UltraHR-100K: Enhancing UHR Image Synthesis with A Large-Scale High-Quality Dataset
• HybridSOMSpikeNet: A Deep Model with Differentiable Soft Self-Organizing Maps and Spiking Dynamics for Waste Classification
• Diagnosing Visual Reasoning: Challenges, Insights, and a Path Forward
• Mixing Importance with Diversity: Joint Optimization for KV Cache Compression in Large Vision-Language Models
• ALICE-LRI: A General Method for Lossless Range Image Generation for Spinning LiDAR Sensors without Calibration Metadata
• AutoScape: Geometry-Consistent Long-Horizon Scene Generation
• ACS-SegNet: An Attention-Based CNN-SegFormer Segmentation Network for Tissue Segmentation in Histopathology
• DyPE: Dynamic Position Extrapolation for Ultra High Resolution Diffusion
• CUPID: Pose-Grounded Generative 3D Reconstruction from a Single Image
• Radar-Camera Fused Multi-Object Tracking: Online Calibration and Common Feature
• ARGenSeg: Image Segmentation with Autoregressive Image Generation Model
• SpectraMorph: Structured Latent Learning for Self-Supervised Hyperspectral Super-Resolution
• LayerComposer: Interactive Personalized T2I via Spatially-Aware Layered Canvas
• HoloCine: Holistic Generation of Cinematic Multi-Shot Long Video Narratives
• Automating Iconclass: LLMs and RAG for Large-Scale Classification of Religious Woodcuts
• AI Pose Analysis and Kinematic Profiling of Range-of-Motion Variations in Resistance Training
• Kinaema: a recurrent sequence model for memory and pose in motion
• GUSL-Dehaze: A Green U-Shaped Learning Approach to Image Dehazing
• Dino-Diffusion Modular Designs Bridge the Cross-Domain Gap in Autonomous Parking
• Frequency Cam: Imaging Periodic Signals in Real-Time
• MS-BART: Unified Modeling of Mass Spectra and Molecules for Structure Elucidation
• On Optimal Hyperparameters for Differentially Private Deep Transfer Learning
• H-SPLID: HSIC-based Saliency Preserving Latent Information Decomposition
• Large Multimodal Models-Empowered Task-Oriented Autonomous Communications: Design Methodology and Implementation Challenges
• Attention Enhanced Entity Recommendation for Intelligent Monitoring in Cloud Systems
• Connecting Jensen-Shannon and Kullback-Leibler Divergences: A New Bound for Representation Learning
• xTime: Extreme Event Prediction with Hierarchical Knowledge Distillation and Expert Fusion
• Bayesian Jammer Localization with a Hybrid CNN and Path-Loss Mixture of Experts
• From Masks to Worlds: A Hitchhiker's Guide to World Models
• Separating the what and how of compositional computation to enable reuse and continual learning
• Optimizing Clinical Fall Risk Prediction: A Data-Driven Integration of EHR Variables with the Johns Hopkins Fall Risk Assessment Tool
• No-Regret Thompson Sampling for Finite-Horizon Markov Decision Processes with Gaussian Processes
• Amplifying Prominent Representations in Multimodal Learning via Variational Dirichlet Process
• MEIcoder: Decoding Visual Stimuli from Neural Activity by Leveraging Most Exciting Inputs
• Out-of-distribution Tests Reveal Compositionality in Chess Transformers
• BadGraph: A Backdoor Attack Against Latent Diffusion Model for Text-Guided Graph Generation
• KL-Regularized Reinforcement Learning is Designed to Mode Collapse
• Spectral Thresholds in Correlated Spiked Models and Fundamental Limits of Partial Least Squares
• Neurotremor: A wearable Supportive Device for Supporting Upper Limb Muscle Function
• Low-Latency Neural Inference on an Edge Device for Real-Time Handwriting Recognition from EEG Signals
• Multi-Resolution Analysis of the Convective Structure of Tropical Cyclones for Short-Term Intensity Guidance
• SODBench: A Large Language Model Approach to Documenting Spreadsheet Operations
• Artificial Intelligence Powered Identification of Potential Antidiabetic Compounds in Ficus religiosa
• Transforming Multi-Omics Integration with GANs: Applications in Alzheimer's and Cancer
• Compressing Biology: Evaluating the Stable Diffusion VAE for Phenotypic Drug Discovery
• Deep Sequence-to-Sequence Models for GNSS Spoofing Detection
• Guiding diffusion models to reconstruct flow fields from sparse data
• SecureInfer: Heterogeneous TEE-GPU Architecture for Privacy-Critical Tensors for Large Language Model Deployment
• Enhanced Cyclic Coordinate Descent Methods for Elastic Net Penalized Linear Models
• Improving Predictive Confidence in Medical Imaging via Online Label Smoothing
• Simultaneously Solving Infinitely Many LQ Mean Field Games In Hilbert Spaces: The Power of Neural Operators
• On Encoding Matrices using Quantum Circuits
• Throwing Vines at the Wall: Structure Learning via Random Search
• From Facts to Folklore: Evaluating Large Language Models on Bengali Cultural Knowledge
• Endogenous Aggregation of Multiple Data Envelopment Analysis Scores for Large Data Sets
• BIOCAP: Exploiting Synthetic Captions Beyond Labels in Biological Foundation Models
• Extending machine learning model for implicit solvation to free energy calculations
• AsyncHZP: Hierarchical ZeRO Parallelism with Asynchronous Scheduling for Scalable LLM Training
• Compositional Generation for Long-Horizon Coupled PDEs
• Multimedia-Aware Question Answering: A Review of Retrieval and Cross-Modal Reasoning Architectures
• Empower Words: DualGround for Structured Phrase and Sentence-Level Temporal Grounding
• Calibrating Multimodal Consensus for Emotion Recognition
• Capability of using the normalizing flows for extraction rare gamma events in the TAIGA experiment
• Neural Networks for Censored Expectile Regression Based on Data Augmentation
• ComProScanner: A multi-agent based framework for composition-property structured data extraction from scientific literature
• A Transformer Inspired AI-based MIMO receiver
• Testing Most Influential Sets
• PointMapPolicy: Structured Point Cloud Processing for Multi-Modal Imitation Learning
• Learning Coupled Earth System Dynamics with GraphDOP
• Partial Optimality in Cubic Correlation Clustering for General Graphs
• Learning Decentralized Routing Policies via Graph Attention-based Multi-Agent Reinforcement Learning in Lunar Delay-Tolerant Networks
• Concentration and excess risk bounds for imbalanced classification with synthetic oversampling
• Decoding the Ear: A Framework for Objectifying Expressiveness from Human Preference Through Efficient Alignment
• Adversary-Aware Private Inference over Wireless Channels
• Blur2seq: Blind Deblurring and Camera Trajectory Estimation from a Single Camera Motion-blurred Image
• Diffusion Autoencoders with Perceivers for Long, Irregular and Multimodal Astronomical Sequences
• Strategic Costs of Perceived Bias in Fair Selection
• Efficient Multi-bit Quantization Network Training via Weight Bias Correction and Bit-wise Coreset Sampling
• Learning to Triage Taint Flows Reported by Dynamic Program Analysis in Node.js Packages
• CSU-PCAST: A Dual-Branch Transformer Framework for medium-range ensemble Precipitation Forecasting
• AlphaFlow: Understanding and Improving MeanFlow Models
• Alleviating Forgetfulness of Linear Attention by Hybrid Sparse Attention and Contextualized Learnable Token Eviction
• Video Prediction of Dynamic Physical Simulations With Pixel-Space Spatiotemporal Transformers
• The Faiss library
• Multi Task Inverse Reinforcement Learning for Common Sense Reward
• Log Neural Controlled Differential Equations: The Lie Brackets Make a Difference
• Channel Balance Interpolation in the Lightning Network via Machine Learning
• Assessing the Probabilistic Fit of Neural Regressors via Conditional Congruence
• Solving 0-1 Integer Programs with Unknown Knapsack Constraints Using Membership Oracles
• Pretraining Decision Transformers with Reward Prediction for In-Context Multi-task Structured Bandit Learning
• Optimizing Time Series Forecasting Architectures: A Hierarchical Neural Architecture Search Approach
• SHAP values via sparse Fourier representation
• Provable Meta-Learning with Low-Rank Adaptations
• Learn2Mix: Training Neural Networks Using Adaptive Data Integration
• Pareto-Optimal Energy Alignment for Designing Nature-Like Antibodies
• Twilight: Adaptive Attention Sparsity with Hierarchical Top-$p$ Pruning
• From Counterfactuals to Trees: Competitive Analysis of Model Extraction Attacks
• WENDy for Nonlinear-in-Parameters ODEs
• Depth-Bounds for Neural Networks via the Braid Arrangement
• Continuous Diffusion Model for Language Modeling
• Harnessing Feature Resonance under Arbitrary Target Alignment for Out-of-Distribution Node Detection
• Gatekeeper: Improving Model Cascades Through Confidence Tuning
• Training Robust Graph Neural Networks by Modeling Noise Dependencies
• Proper decision trees: An axiomatic framework for solving optimal decision tree problems with arbitrary splitting rules
• Real-Time Cell Sorting with Scalable In Situ FPGA-Accelerated Deep Learning
• Streaming Federated Learning with Markovian Data
• Exploring the Energy Landscape of RBMs: Reciprocal Space Insights into Bosons, Hierarchical Learning and Symmetry Breaking
• CTSketch: Compositional Tensor Sketching for Scalable Neurosymbolic Learning
• Sign-In to the Lottery: Reparameterizing Sparse Training From Scratch
• Adaptive PCA-Based Outlier Detection for Multi-Feature Time Series in Space Missions
• SetONet: A Set-Based Operator Network for Solving PDEs with Variable-Input Sampling
• Reinforcing Multi-Turn Reasoning in LLM Agents via Turn-Level Reward Design
• Improving Energy Natural Gradient Descent through Woodbury, Momentum, and Randomization
• Embedding principle of homogeneous neural network for classification problem
• Deep Learning for Continuous-time Stochastic Control with Jumps
• Tropical Attention: Neural Algorithmic Reasoning for Combinatorial Algorithms
• Wasserstein Transfer Learning
• DesignX: Human-Competitive Algorithm Designer for Black-Box Optimization
• Geometry Aware Operator Transformer as an Efficient and Accurate Neural Surrogate for PDEs on Arbitrary Domains
• Born a Transformer -- Always a Transformer? On the Effect of Pretraining on Architectural Abilities
• Taming Hyperparameter Sensitivity in Data Attribution: Practical Selection Without Costly Retraining
• Blending Complementary Memory Systems in Hybrid Quadratic-Linear Transformers
• KOALA++: Efficient Kalman-Based Optimization of Neural Networks with Gradient-Covariance Products
• Spark Transformer: Reactivating Sparsity in FFN and Attention
• Execution Guided Line-by-Line Code Generation
• What Happens During the Loss Plateau? Understanding Abrupt Learning in Transformers
• S$^2$-Diffusion: Generalizing from Instance-level to Category-level Skills in Robot Manipulation
• Toward Metaphor-Fluid Conversation Design for Voice User Interfaces
• Neural Attention Search
• ExpertLens: Activation steering features are highly interpretable
• Deep Learning-Powered Electrical Brain Signals Analysis: Advancing Neurological Diagnostics
• DIPLI: Deep Image Prior Lucky Imaging for Blind Astronomical Image Restoration
• Token embeddings violate the manifold hypothesis
• Integrating Structural and Semantic Signals in Text-Attributed Graphs with BiGTex
• Fast-Slow Thinking GRPO for Large Vision-Language Model Reasoning
• Don't be lazy: CompleteP enables compute-efficient deep transformers
• PRUNE: A Patching Based Repair Framework for Certifiable Unlearning of Neural Networks
• UMoE: Unifying Attention and FFN with Shared Experts
• Fair Clustering via Alignment
• Superposition Yields Robust Neural Scaling
• Shallow Flow Matching for Coarse-to-Fine Text-to-Speech Synthesis
• CALM-PDE: Continuous and Adaptive Convolutions for Latent Space Modeling of Time-dependent PDEs
• One-Step Offline Distillation of Diffusion-based Models via Koopman Modeling
• CLEVER: A Curated Benchmark for Formally Verified Code Generation
• Text Generation Beyond Discrete Token Sampling
• RL Tango: Reinforcing Generator and Verifier Together for Language Reasoning
• LLM-Explorer: A Plug-in Reinforcement Learning Policy Exploration Enhancement Driven by Large Language Models
• How Ensembles of Distilled Policies Improve Generalisation in Reinforcement Learning
• Breaking mBad! Supervised Fine-tuning for Cross-Lingual Detoxification
• LeCoDe: A Benchmark Dataset for Interactive Legal Consultation Dialogue Evaluation
• Towards Interpretability Without Sacrifice: Faithful Dense Layer Decomposition with Mixture of Decoders
• Autoencoding Random Forests
• Balanced Token Pruning: Accelerating Vision Language Models Beyond Local Optimization
• Train with Perturbation, Infer after Merging: A Two-Stage Framework for Continual Learning
• Machine Unlearning under Overparameterization
• Direct Numerical Layout Generation for 3D Indoor Scene Synthesis via Spatial Reasoning
• FuseUNet: A Multi-Scale Feature Fusion Method for U-like Networks
• LeVo: High-Quality Song Generation with Multi-Preference Alignment
• Edit Flows: Flow Matching with Edit Operations
• AssistedDS: Benchmarking How External Domain Knowledge Assists LLMs in Automated Data Science
• Watermarking Autoregressive Image Generation
• Flow based approach for Dynamic Temporal Causal models with non-Gaussian or Heteroscedastic Noises
• ReDit: Reward Dithering for Improved LLM Policy Optimization
• From High-SNR Radar Signal to ECG: A Transfer Learning Model with Cardio-Focusing Algorithm for Scenarios with Limited Data
• Learning Modular Exponentiation with Transformers
• Serving LLMs in HPC Clusters: A Comparative Study of Qualcomm Cloud AI 100 Ultra and NVIDIA Data Center GPUs
• Symbiosis: Multi-Adapter Inference and Fine-Tuning
• Crafting Imperceptible On-Manifold Adversarial Attacks for Tabular Data
• Quantization-Aware Neuromorphic Architecture for Efficient Skin Disease Classification on Resource-Constrained Devices
• Some Attention is All You Need for Retrieval
• An Integrated Approach to Neural Architecture Search for Deep Q-Networks
• FairGRPO: Fair Reinforcement Learning for Equitable Clinical Reasoning
• Enhancing Diagnostic Accuracy for Urinary Tract Disease through Explainable SHAP-Guided Feature Selection and Classification
• FINDER: Feature Inference on Noisy Datasets using Eigenspace Residuals
• Beyond the Ideal: Analyzing the Inexact Muon Update
• Mitigating Privacy-Utility Trade-off in Decentralized Federated Learning via $f$-Differential Privacy
• Are Greedy Task Orderings Better Than Random in Continual Linear Regression?
• Towards Strong Certified Defense with Universal Asymmetric Randomization
• Abstain Mask Retain Core: Time Series Prediction by Adaptive Masking Loss with Representation Consistency
• No Compute Left Behind: Rethinking Reasoning and Sampling with Masked Diffusion Models
• Machine Learning-Based Localization Accuracy of RFID Sensor Networks via RSSI Decision Trees and CAD Modeling for Defense Applications
• SALT: Step-level Advantage Assignment for Long-horizon Agents via Trajectory Graph
• Speculative Sampling for Parametric Temporal Point Processes
• Learning Personalized Ad Impact via Contextual Reinforcement Learning under Delayed Rewards
• Not-a-Bandit: Provably No-Regret Drafter Selection in Speculative Decoding for LLMs
• A Multi-Layer Machine Learning and Econometric Pipeline for Forecasting Market Risk: Evidence from Cryptoasset Liquidity Spillovers
• Coupled Transformer Autoencoder for Disentangling Multi-Region Neural Latent Dynamics
• Hierarchical Dual-Head Model for Suicide Risk Assessment via MentalRoBERTa
• Competition is the key: A Game Theoretic Causal Discovery Approach
• On pattern classification with weighted dimensions
• Why Prototypes Collapse: Diagnosing and Preventing Partial Collapse in Prototypical Self-Supervised Learning
• There is No "apple" in Timeseries: Rethinking TSFM through the Lens of Invariance
• Understanding Mechanistic Role of Structural and Functional Connectivity in Tau Propagation Through Multi-Layer Modeling
• ADP-VRSGP: Decentralized Learning with Adaptive Differential Privacy via Variance-Reduced Stochastic Gradient Push
• Empowering Targeted Neighborhood Search via Hyper Tour for Large-Scale TSP
• Every Question Has Its Own Value: Reinforcement Learning with Explicit Human Values
• Risk-Averse Constrained Reinforcement Learning with Optimized Certainty Equivalents
• Approximate Replicability in Learning
• CO-PFL: Contribution-Oriented Personalized Federated Learning for Heterogeneous Networks
• Alternatives to the Laplacian for Scalable Spectral Clustering with Group Fairness Constraints
• Sparse Local Implicit Image Function for sub-km Weather Downscaling
• Layer-to-Layer Knowledge Mixing in Graph Neural Network for Chemical Property Prediction
• FedGPS: Statistical Rectification Against Data Heterogeneity in Federated Learning
• Optimistic Task Inference for Behavior Foundation Models
• ImpossibleBench: Measuring LLMs' Propensity of Exploiting Test Cases
• Scalable GPU-Accelerated Euler Characteristic Curves: Optimization and Differentiable Learning for PyTorch
• SynTSBench: Rethinking Temporal Pattern Learning in Deep Learning Models for Time Series
• KCM: KAN-Based Collaboration Models Enhance Pretrained Large Models
• ResearchGPT: Benchmarking and Training LLMs for End-to-End Computer Science Research Workflows
• Quantifying Distributional Invariance in Causal Subgraph for IRM-Free Graph Generalization
• InvDec: Inverted Decoder for Multivariate Time Series Forecasting with Separated Temporal and Variate Modeling
• Synthetic Data for Robust Runway Detection
• Ask a Strong LLM Judge when Your Reward Model is Uncertain
• Hierarchical Time Series Forecasting with Robust Reconciliation
• Why DPO is a Misspecified Estimator and How to Fix It
• Addressing Mark Imbalance in Integration-free Neural Marked Temporal Point Processes
• An Empirical Study of Sample Selection Strategies for Large Language Model Repair
• Explainable Benchmarking through the Lense of Concept Learning
• Intransitive Player Dominance and Market Inefficiency in Tennis Forecasting: A Graph Neural Network Approach
• Bi-CoG: Bi-Consistency-Guided Self-Training for Vision-Language Models
• SheafAlign: A Sheaf-theoretic Framework for Decentralized Multimodal Alignment
• A Unified Framework for Zero-Shot Reinforcement Learning
• Embedding the MLOps Lifecycle into OT Reference Models
• Convergence Analysis of SGD under Expected Smoothness
• Limits of PRM-Guided Tree Search for Mathematical Reasoning with LLMs
• Context-level Language Modeling by Learning Predictive Context Embeddings
• UI-Ins: Enhancing GUI Grounding with Multi-Perspective Instruction-as-Reasoning
• Breakdance Video classification in the age of Generative AI
• A Parameter-Efficient Mixture-of-Experts Framework for Cross-Modal Geo-Localization
• RAG-Stack: Co-Optimizing RAG Quality and Performance From the Vector Database Perspective
• DB-FGA-Net: Dual Backbone Frequency Gated Attention Network for Multi-Class Classification with Grad-CAM Interpretability
• Enhancing Security in Deep Reinforcement Learning: A Comprehensive Survey on Adversarial Attacks and Defenses
• LEGO: A Lightweight and Efficient Multiple-Attribute Unlearning Framework for Recommender Systems
• MemER: Scaling Up Memory for Robot Control via Experience Retrieval
• GhostEI-Bench: Do Mobile Agents Resilience to Environmental Injection in Dynamic On-Device Environments?
• Multi-Task Deep Learning for Surface Metrology
• Teaching Language Models to Reason with Tools
• What do AI-Generated Images Want?
• Evaluating Latent Knowledge of Public Tabular Datasets in Large Language Models
• The Impact of Negated Text on Hallucination with Large Language Models
• VLSP 2025 MLQA-TSR Challenge: Vietnamese Multimodal Legal Question Answering on Traffic Sign Regulation
• Relative-Based Scaling Law for Neural Language Models
• FLAS: a combination of proactive and reactive auto-scaling architecture for distributed services
• Balancing Specialization and Centralization: A Multi-Agent Reinforcement Learning Benchmark for Sequential Industrial Control
• Dynamic Weight Adjustment for Knowledge Distillation: Leveraging Vision Transformer for High-Accuracy Lung Cancer Detection and Real-Time Deployment
• UniSE: A Unified Framework for Decoder-only Autoregressive LM-based Speech Enhancement
• MolBridge: Atom-Level Joint Graph Refinement for Robust Drug-Drug Interaction Event Prediction
• Symbolic Regression and Differentiable Fits in Beyond the Standard Model Physics
• Transferable Black-Box One-Shot Forging of Watermarks via Image Preference Models
• Structures generated in a multiagent system performing information fusion in peer-to-peer resource-constrained networks
• RECALL: REpresentation-aligned Catastrophic-forgetting ALLeviation via Hierarchical Model Merging
• Hurdle-IMDL: An Imbalanced Learning Framework for Infrared Rainfall Retrieval
• Steering Evaluation-Aware Language Models To Act Like They Are Deployed
• Hierarchical Sequence Iteration for Heterogeneous Question Answering
• Metis-HOME: Hybrid Optimized Mixture-of-Experts for Multimodal Reasoning
• Fake-in-Facext: Towards Fine-Grained Explainable DeepFake Analysis
• ARC-Encoder: learning compressed text representations for large language models
• The Dog the Cat Chased Stumped the Model: Measuring When Language Models Abandon Structure for Shortcuts
• GlobalRAG: Enhancing Global Reasoning in Multi-hop Question Answering via Reinforcement Learning
• Structural Invariance Matters: Rethinking Graph Rewiring through Graph Metrics
• AdaDoS: Adaptive DoS Attack via Deep Adversarial Reinforcement Learning in SDN
• Open-o3 Video: Grounded Video Reasoning with Explicit Spatio-Temporal Evidence
• Can ChatGPT Code Communication Data Fairly?: Empirical Evidence from Multiple Collaborative Tasks
• Unsupervised Domain Adaptation via Similarity-based Prototypes for Cross-Modality Segmentation
• Resounding Acoustic Fields with Reciprocity
• OnlineSplatter: Pose-Free Online 3D Reconstruction for Free-Moving Objects
• Generalizable Reasoning through Compositional Energy Minimization
• Practical Code RAG at Scale: Task-Aware Retrieval Design Choices under Compute Budgets
• BUSTED at AraGenEval Shared Task: A Comparative Study of Transformer-Based Models for Arabic AI-Generated Text Detection
• PSO-XAI: A PSO-Enhanced Explainable AI Framework for Reliable Breast Cancer Detection
• Black Box Absorption: LLMs Undermining Innovative Ideas
• Equitable Survival Prediction: A Fairness-Aware Survival Modeling (FASM) Approach
• Quantum Processing Unit (QPU) processing time Prediction with Machine Learning
• Deep Learning in Dental Image Analysis: A Systematic Review of Datasets, Methodologies, and Emerging Challenges
• Why Did Apple Fall To The Ground: Evaluating Curiosity In Large Language Model
• The Reasoning Lingua Franca: A Double-Edged Sword for Multilingual AI
• Finding the Sweet Spot: Trading Quality, Cost, and Speed During Inference-Time LLM Reflection
• GRACE: GRaph-based Addiction Care prEdiction
• R2-SVC: Towards Real-World Robust and Expressive Zero-shot Singing Voice Conversion
• A Scalable, Causal, and Energy Efficient Framework for Neural Decoding with Spiking Neural Networks
• Neural Diversity Regularizes Hallucinations in Small Models
• Exploring Large Language Models for Access Control Policy Synthesis and Summarization
• Fusing Narrative Semantics for Financial Volatility Forecasting
• Real-Time Gait Adaptation for Quadrupeds using Model Predictive Control and Reinforcement Learning
• Unsupervised Anomaly Prediction with N-BEATS and Graph Neural Network in Multi-variate Semiconductor Process Time Series
• User Perceptions of Privacy and Helpfulness in LLM Responses to Privacy-Sensitive Scenarios
• Automated Extraction of Fluoropyrimidine Treatment and Treatment-Related Toxicities from Clinical Notes Using Natural Language Processing
• Co-Designing Quantum Codes with Transversal Diagonal Gates via Multi-Agent Systems
• Thought Communication in Multiagent Collaboration
• Empathic Prompting: Non-Verbal Context Integration for Multimodal LLM Conversations
• Reinforcement Learning and Consumption-Savings Behavior
• RAGRank: Using PageRank to Counter Poisoning in CTI LLM Pipelines
• FieldGen: From Teleoperated Pre-Manipulation Trajectories to Field-Guided Data Generation
• Are Large Reasoning Models Good Translation Evaluators? Analysis and Performance Boost
• A Use-Case Specific Dataset for Measuring Dimensions of Responsible Performance in LLM-generated Text
• Bayesian Inference of Primordial Magnetic Field Parameters from CMB with Spherical Graph Neural Networks
• Simple Context Compression: Mean-Pooling and Multi-Ratio Training
• Compress to Impress: Efficient LLM Adaptation Using a Single Gradient Step on 100 Samples
• The Reality Gap in Robotics: Challenges, Solutions, and Best Practices
• On the Detectability of LLM-Generated Text: What Exactly Is LLM-Generated Text?
• Small Drafts, Big Verdict: Information-Intensive Visual Reasoning via Speculation
• GSWorld: Closed-Loop Photo-Realistic Simulation Suite for Robotic Manipulation
• VAMOS: A Hierarchical Vision-Language-Action Model for Capability-Modulated and Steerable Navigation
• Towards General Modality Translation with Contrastive and Predictive Latent Diffusion Bridge
• MindForge: Empowering Embodied Agents with Theory of Mind for Lifelong Cultural Learning
• MIR-Bench: Can Your LLM Recognize Complicated Patterns via Many-Shot In-Context Reasoning?
• Stop Summation: Min-Form Credit Assignment Is All Process Reward Model Needs for Reasoning
• Towards Machine Learning-based Model Predictive Control for HVAC Control in Multi-Context Buildings at Scale via Ensemble Learning
• Privacy Risks and Preservation Methods in Explainable Artificial Intelligence: A Scoping Review
• SAFEPATH: Preventing Harmful Reasoning in Chain-of-Thought via Early Alignment
• Lessons Learned: A Multi-Agent Framework for Code LLMs to Learn and Improve
• Does Thinking More always Help? Mirage of Test-Time Scaling in Reasoning Models
• Adaptive Learning in Spatial Agent-Based Models for Climate Risk Assessment: A Geospatial Framework with Evolutionary Economic Agents
• Aligning Transformers with Continuous Feedback via Energy Rank Alignment
• Annotation Guidelines-Based Knowledge Augmentation: Towards Enhancing Large Language Models for Educational Text Classification
• Towards Understanding Safety Alignment: A Mechanistic Perspective from Safety Neurons
• Residual Kolmogorov-Arnold Network for Enhanced Deep Learning
• Not All Heads Matter: A Head-Level KV Cache Compression Method with Integrated Retrieval and Reasoning
• Bi-Mamba: Towards Accurate 1-Bit State Space Models
• Making Classic GNNs Strong Baselines Across Varying Homophily: A Smoothness-Generalization Perspective
• Face-Human-Bench: A Comprehensive Benchmark of Face and Human Understanding for Multi-modal Assistants
• DMWM: Dual-Mind World Model with Long-Term Imagination
• A Quantum-Inspired Algorithm for Solving Sudoku Puzzles and the MaxCut Problem
• Benchmarking Reasoning Reliability in Artificial Intelligence Models for Energy-System Analysis
• Branch-and-Browse: Efficient and Controllable Web Exploration with Tree-Structured Reasoning and Action Memory
• DAG-Math: Graph-Guided Mathematical Reasoning in LLMs
• Surfer 2: The Next Generation of Cross-Platform Computer Use Agents
• RELATE: A Schema-Agnostic Perceiver Encoder for Multimodal Relational Graphs
• A new wave of vehicle insurance fraud fueled by generative AI
• AI-Driven Personalized Learning: Predicting Academic Per-formance Through Leadership Personality Traits
• LLMs can hide text in other text of the same length.ipynb
• AI PB: A Grounded Generative Agent for Personalized Investment Insights
• Human-Centered LLM-Agent System for Detecting Anomalous Digital Asset Transactions
• The Verification-Value Paradox: A Normative Critique of Gen AI in Legal Practice
• TRUST: A Decentralized Framework for Auditing Large Language Model Reasoning
• The Lock-In Phase Hypothesis: Identity Consolidation as a Precursor to AGI
• Merge and Conquer: Evolutionarily Optimizing AI for 2048
• Individualized Cognitive Simulation in Large Language Models: Evaluating Different Cognitive Representation Methods
• Using Large Language Models for Abstraction of Planning Domains - Extended Version
• Classical Feature Embeddings Help in BERT-Based Human Mobility Prediction
• Multi-Step Reasoning for Embodied Question Answering via Tool Augmentation
• Bias by Design? How Data Practices Shape Fairness in AI Healthcare Systems
• Collateral Damage Assessment Model for AI System Target Engagement in Military Operations
• LLM-empowered knowledge graph construction: A survey
• IKnow: Instruction-Knowledge-Aware Continual Pretraining for Effective Domain Adaptation
• A computational model and tool for generating more novel opportunities in professional innovation processes
• Neural Reasoning for Robust Instance Retrieval in $\mathcal{SHOIQ}$
• FLORA: Unsupervised Knowledge Graph Alignment by Fuzzy Logic
• Lost in Translation: Policymakers are not really listening to Citizen Concerns about AI
• Transferable Graph Learning for Transmission Congestion Management via Busbar Splitting
• What Defines Good Reasoning in LLMs? Dissecting Reasoning Steps with Multi-Aspect Evaluation
• Efficient Algorithms for Computing Random Walk Centrality
• Towards the Formalization of a Trustworthy AI for Mining Interpretable Models explOiting Sophisticated Algorithms
• Towards Reliable Evaluation of Large Language Models for Multilingual and Multimodal E-Commerce Applications
• Fluidity Index: Next-Generation Super-intelligence Benchmarks
• Integrating Machine Learning into Belief-Desire-Intention Agents: Current Advances and Open Challenges
• The Shape of Reasoning: Topological Analysis of Reasoning Traces in Large Language Models
• Plan Then Retrieve: Reinforcement Learning-Guided Complex Reasoning over Knowledge Graphs
• A Coherence-Based Measure of AGI
• Real Deep Research for AI, Robotics and Beyond
• SLYKLatent: A Learning Framework for Gaze Estimation Using Deep Facial Feature Learning
• SSL-SE-EEG: A Framework for Robust Learning from Unlabeled EEG Data with Self-Supervised Learning and Squeeze-Excitation Networks
• CourtGuard: A Local, Multiagent Prompt Injection Classifier
• Prompt Decorators: A Declarative and Composable Syntax for Reasoning, Formatting, and Control in LLMs
• Can Reasoning Models Obfuscate Reasoning? Stress-Testing Chain-of-Thought Monitorability
• An Evaluation of the Pedagogical Soundness and Usability of AI-Generated Lesson Plans Across Different Models and Prompt Frameworks in High-School Physics
• From Large to Small: Transferring CUDA Optimization Expertise via Reasoning Graph
• Stream: Scaling up Mechanistic Interpretability to Long Context in LLMs via Sparse Attention
• Quantifying Feature Importance for Online Content Moderation
• From Optimization to Prediction: Transformer-Based Path-Flow Estimation to the Traffic Assignment Problem
• Can They Dixit? Yes they Can! Dixit as a Playground for Multimodal Language Model Capabilities
• Large Language Model enabled Mathematical Modeling
• Learning from Supervision with Semantic and Episodic Memory: A Reflective Approach to Agent Adaptation
• Robust Reinforcement Learning in Finance: Modeling Market Impact with Elliptic Uncertainty Sets
• On the Optimal Construction of Unbiased Gradient Estimators for Zeroth-Order Optimization
• LyriCAR: A Difficulty-Aware Curriculum Reinforcement Learning Framework For Controllable Lyric Translation
• A Tutorial on Cognitive Biases in Agentic AI-Driven 6G Autonomous Networks
• Revisiting Zeroth-Order Optimization: Minimum-Variance Two-Point Estimators and Directionally Aligned Perturbations
• LLM-Augmented Symbolic NLU System for More Reliable Continuous Causal Statement Interpretation
• A Framework for the Adoption and Integration of Generative AI in Midsize Organizations and Enterprises (FAIGMOE)
• Beyond MedQA: Towards Real-world Clinical Decision Making in the Era of LLMs
• Forging GEMs: Advancing Greek NLP through Quality-Based Corpus Curation and Specialized Pre-training
• Optimized Distortion in Linear Social Choice
• The Temporal Graph of Bitcoin Transactions
• Beyond One-Way Influence: Bidirectional Opinion Dynamics in Multi-Turn Human-LLM Interactions
• Approximate Model Predictive Control for Microgrid Energy Management via Imitation Learning
• Ask What Your Country Can Do For You: Towards a Public Red Teaming Model
• ShapeX: Shapelet-Driven Post Hoc Explanations for Time Series Classification Models
• CreativityPrism: A Holistic Benchmark for Large Language Model Creativity
• StableSketcher: Enhancing Diffusion Model for Pixel-based Sketch Generation via Visual Question Answering Feedback
• On the Structure of Stationary Solutions to McKean-Vlasov Equations with Applications to Noisy Transformers
• Leveraging the Power of Large Language Models in Entity Linking via Adaptive Routing and Targeted Reasoning
• SAID: Empowering Large Language Models with Self-Activating Internal Defense
• Are Stereotypes Leading LLMs' Zero-Shot Stance Detection ?
• IB-GAN: Disentangled Representation Learning with Information Bottleneck Generative Adversarial Networks
• Collective Communication for 100k+ GPUs
• Mixture-of-Minds: Multi-Agent Reinforcement Learning for Table Understanding
• PPMStereo: Pick-and-Play Memory Construction for Consistent Dynamic Stereo Matching
• Stuck in the Matrix: Probing Spatial Reasoning in Large Language Models
• Assessing the Feasibility of Early Cancer Detection Using Routine Laboratory Data: An Evaluation of Machine Learning Approaches on an Imbalanced Dataset
• Automated Cloud Infrastructure-as-Code Reconciliation with AI Agents
• High-order Interactions Modeling for Interpretable Multi-Agent Q-Learning
• FinCARE: Financial Causal Analysis with Reasoning and Evidence
• QKCV Attention: Enhancing Time Series Forecasting with Static Categorical Embeddings for Both Lightweight and Pre-trained Foundation Models
• Federated Learning via Meta-Variational Dropout
• Why LVLMs Are More Prone to Hallucinations in Longer Responses: The Role of Context
• Multi-Objective Reinforcement Learning with Max-Min Criterion: A Game-Theoretic Approach
• Tri-Modal Severity Fused Diagnosis across Depression and Post-traumatic Stress Disorders
• What Does It Take to Build a Performant Selective Classifier?
• Towards AI Agents for Course Instruction in Higher Education: Early Experiences from the Field