AI Research News Feeds for October 31st, 2025

AI RESEARCH PAPERS & ACADEMIC SOURCES

• BikeScenes: Online LiDAR Semantic Segmentation for Bicycles
• Generative Image Restoration and Super-Resolution using Physics-Informed Synthetic Data for Scanning Tunneling Microscopy
• SplitFlow: Flow Decomposition for Inversion-Free Text-to-Image Editing
• Fine-tuning Segment Anything for Real-Time Tumor Tracking in Cine-MRI
• Larger Hausdorff Dimension in Scanning Pattern Facilitates Mamba-Based Methods in Low-Light Image Enhancement
• Enhancing Temporal Understanding in Video-LLMs through Stacked Temporal Attention in Vision Encoders
• FlexICL: A Flexible Visual In-context Learning Framework for Elbow and Wrist Ultrasound Segmentation
• OracleAgent: A Multimodal Reasoning Agent for Oracle Bone Script Research
• JOGS: Joint Optimization of Pose Estimation and 3D Gaussian Splatting
• Exploring Object-Aware Attention Guided Frame Association for RGB-D SLAM
• FullPart: Generating each 3D Part at Full Resolution
• BasicAVSR: Arbitrary-Scale Video Super-Resolution via Image Priors and Enhanced Motion Compensation
• Detecting Unauthorized Vehicles using Deep Learning for Smart Cities: A Case Study on Bangladesh
• CRAG-MM: Multi-modal Multi-turn Comprehensive RAG Benchmark
• MoTDiff: High-resolution Motion Trajectory estimation from a single blurred image using Diffusion models
• Sketch2PoseNet: Efficient and Generalized Sketch to 3D Human Pose Prediction
• Developing a Multi-task Ensemble Geometric Deep Network for Supply Chain Sustainability and Risk Management
• OmniLayout: Enabling Coarse-to-Fine Learning with LLMs for Universal Document Layout Generation
• Revisiting Generative Infrared and Visible Image Fusion Based on Human Cognitive Laws
• Exploring Complementarity and Explainability in CNNs for Periocular Verification Across Acquisition Distances
• Beyond Imitation: Constraint-Aware Trajectory Generation with Flow Matching For End-to-End Autonomous Driving
• Leveraging Large-Scale Face Datasets for Deep Periocular Recognition via Ocular Cropping
• Towards Realistic Earth-Observation Constellation Scheduling: Benchmark and Methodology
• Exploring the correlation between the type of music and the emotions evoked: A study using subjective questionnaires and EEG
• A Hybrid Framework Bridging CNN and ViT based on Theory of Evidence for Diabetic Retinopathy Grading
• EEG-Driven Image Reconstruction with Saliency-Guided Diffusion Models
• A-TPT: Angular Diversity Calibration Properties for Test-Time Prompt Tuning of Vision-Language Models
• PointSt3R: Point Tracking through 3D Grounded Correspondence
• Towards Fine-Grained Vision-Language Alignment for Few-Shot Anomaly Detection
• Analysis of the Robustness of an Edge Detector Based on Cellular Automata Optimized by Particle Swarm
• SA$^{2}$Net: Scale-Adaptive Structure-Affinity Transformation for Spine Segmentation from Ultrasound Volume Projection Imaging
• AdSum: Two-stream Audio-visual Summarization for Automated Video Advertisement Clipping
• Dynamic Context-Aware Scene Reasoning Using Vision-Language Alignment in Zero-Shot Real-World Scenarios
• CATCH: A Modular Cross-domain Adaptive Template with Hook
• Emu3.5: Native Multimodal Models are World Learners
• Spiking Patches: Asynchronous, Sparse, and Efficient Tokens for Event Cameras
• PT-DETR: Small Target Detection Based on Partially-Aware Detail Focus
• All You Need for Object Detection: From Pixels, Points, and Prompts to Next-Gen Fusion and Multimodal LLMs/VLMs in Autonomous Vehicles
• Towards Reliable Sea Ice Drift Estimation in the Arctic Deep Learning Optical Flow on RADARSAT-2
• Improving Classification of Occluded Objects through Scene Context
• The Impact and Outlook of 3D Gaussian Splatting
• ChartAB: A Benchmark for Chart Grounding & Dense Alignment
• The Quest for Generalizable Motion Generation: Data, Model, and Evaluation
• SEE4D: Pose-Free 4D Generation via Auto-Regressive Video Inpainting
• Masked Diffusion Captioning for Visual Feature Learning
• Groupwise Registration with Physics-Informed Test-Time Adaptation on Multi-parametric Cardiac MRI
• StructLayoutFormer:Conditional Structured Layout Generation via Structure Serialization and Disentanglement
• Self-localization on a 3D map by fusing global and local features from a monocular camera
• AgriGS-SLAM: Orchard Mapping Across Seasons via Multi-View Gaussian Splatting SLAM
• Comparative Analysis of Deep Learning Models for Olive Tree Crown and Shadow Segmentation Towards Biovolume Estimation
• SAMRI: Segment Anything Model for MRI
• BRIQA: Balanced Reweighting in Image Quality Assessment of Pediatric Brain MRI
• ProstNFound+: A Prospective Study using Medical Foundation Models for Prostate Cancer Detection
• MORE: Multi-Organ Medical Image REconstruction Dataset
• Quality-Aware Prototype Memory for Face Representation Learning
• Dynamic Traceback Learning for Medical Report Generation
• EmoAttack: Emotion-to-Image Diffusion Models for Emotional Backdoor Generation
• NerfBaselines: Consistent and Reproducible Evaluation of Novel View Synthesis Methods
• Static for Dynamic: Towards a Deeper Understanding of Dynamic Facial Expressions Using Static Expression Data
• A Continuous and Interpretable Morphometric for Robust Quantification of Dynamic Biological Shapes
• Defending Multimodal Backdoored Models by Repulsive Visual Prompt Tuning
• Language-guided Open-world Video Anomaly Detection under Weak Supervision
• SD-ReID: View-aware Stable Diffusion for Aerial-Ground Person Re-Identification
• DOVE: Efficient One-Step Diffusion Model for Real-World Video Super-Resolution
• Unleashing Diffusion Transformers for Visual Correspondence by Modulating Massive Activations
• LODGE: Level-of-Detail Large-Scale Gaussian Splatting with Efficient Rendering
• RRCANet: Recurrent Reusable-Convolution Attention Network for Infrared Small Target Detection
• MoralCLIP: Contrastive Alignment of Vision-and-Language Representations with Moral Foundations Theory
• Boosting Generative Adversarial Transferability with Self-supervised Vision Transformer Features
• DDL: A Large-Scale Datasets for Deepfake Detection and Localization in Diversified Real-World Scenarios
• ScoreAdv: Score-based Targeted Generation of Natural Adversarial Examples via Diffusion Models
• From One to More: Contextual Part Latents for 3D Generation
• Disentangled 4D Gaussian Splatting: Rendering High-Resolution Dynamic World at 343 FPS
• CronusVLA: Towards Efficient and Robust Manipulation via Multi-Frame Vision-Language-Action Modeling
• Bridging the Gap between Empirical Welfare Maximization and Conditional Average Treatment Effect Estimation in Policy Learning
• SteerVLM: Robust Model Control through Lightweight Activation Steering for Vision Language Models
• Surpassing state of the art on AMD area estimation from RGB fundus images through careful selection of U-Net architectures and loss functions for class imbalance
• A Unified Theory for Causal Inference: Direct Debiased Machine Learning via Bregman-Riesz Regression
• HEIR: Learning Graph-Based Motion Hierarchies
• Scaling Image Geo-Localization to Continent Level
• OmniX: From Unified Panoramic Generation and Perception to Graphics-Ready 3D Scenes
• Time Weaver: A Conditional Time Series Generation Model
• Parallel Unlearning in Inherited Model Networks
• Dolphin: A Programmable Framework for Scalable Neurosymbolic Learning
• Stability and Sharper Risk Bounds with Convergence Rate $\tilde{O}(1/n^2)$
• Hysteresis Activation Function for Efficient Inference
• HoGA: Higher-Order Graph Attention via Diversity-Aware k-Hop Sampling
• Omni-Mol: Multitask Molecular Model for Any-to-any Modalities
• Decoding for Punctured Convolutional and Turbo Codes: A Deep Learning Solution for Protocols Compliance
• Experiments with Optimal Model Trees
• Explainable post-training bias mitigation with distribution-based fairness metrics
• Smart Exploration in Reinforcement Learning using Bounded Uncertainty Models
• Advancing Local Clustering on Graphs via Compressive Sensing: Semi-supervised and Unsupervised Methods
• AnomalyMatch: Discovering Rare Objects of Interest with Semi-supervised and Active Learning
• A Robust and Non-Iterative Tensor Decomposition Method with Automatic Thresholding
• Improving the Euclidean Diffusion Generation of Manifold Data by Mitigating Score Function Singularity
• Is Grokking a Computational Glass Relaxation?
• Neurosymbolic Diffusion Models
• On the creation of narrow AI: hierarchy and nonlocality of neural network skills
• C-LoRA: Contextual Low-Rank Adaptation for Uncertainty Estimation in Large Language Models
• TabSTAR: A Tabular Foundation Model for Tabular Data with Text Fields
• Oryx: a Scalable Sequence Model for Many-Agent Coordination in Offline MARL
• Rethinking Neural Combinatorial Optimization for Vehicle Routing Problems with Different Constraint Tightness Degrees
• Improving Generalization of Neural Combinatorial Optimization for Vehicle Routing Problems via Test-Time Projection Learning
• MTL-KD: Multi-Task Learning Via Knowledge Distillation for Generalizable Neural Vehicle Routing Solver
• AANet: Virtual Screening under Structural Uncertainty via Alignment and Aggregation
• Mind the Gap: Removing the Discretization Gap in Differentiable Logic Gate Networks
• When Kernels Multiply, Clusters Unify: Fusing Embeddings with the Kronecker Product
• A geometric framework for momentum-based optimizers for low-rank training
• LLM-Driven Treatment Effect Estimation Under Inference Time Text Confounding
• Generalized Linear Bandits: Almost Optimal Regret with One-Pass Update
• FEDONet : Fourier-Embedded DeepONet for Spectrally Accurate Operator Learning
• Linearized Optimal Transport for Analysis of High-Dimensional Point-Cloud and Single-Cell Data
• Hierarchical Graph Networks for Accurate Weather Forecasting via Lightweight Training
• GSE: Group-wise Sparse and Explainable Adversarial Attacks
• SafEDMD: A Koopman-based data-driven controller design framework for nonlinear dynamical systems
• Random pairing MLE for estimation of item parameters in Rasch model
• Infinite-dimensional Mahalanobis Distance with Applications to Kernelized Novelty Detection
• Diffusion Map Autoencoder
• Integrating Protein Sequence and Expression Level to Analysis Molecular Characterization of Breast Cancer Subtypes
• Unified Error Correction Code Transformer with Low Complexity
• OnlyFlow: Optical Flow based Motion Conditioning for Video Diffusion Models
• Beyond likelihood ratio bias: Nested multi-time-scale stochastic approximation for likelihood-free parameter estimation
• HyGen: Efficient LLM Serving via Elastic Online-Offline Request Co-location
• Model Provenance Testing for Large Language Models
• On the Impact of Performative Risk Minimization for Binary Random Variables
• Improving LLM Safety Alignment with Dual-Objective Optimization
• Accurate predictive model of band gap with selected important features based on explainable machine learning
• Cybersecurity threat detection based on a UEBA framework using Deep Autoencoders
• Optimal Online Change Detection via Random Fourier Features
• Partially-Supervised Neural Network Model For Quadratic Multiparametric Programming
• Predictive Causal Inference via Spatio-Temporal Modeling and Penalized Empirical Likelihood
• Machine-learning competition to grade EEG background patterns in newborns with hypoxic-ischaemic encephalopathy
• Direct Debiased Machine Learning via Bregman Divergence Minimization
• LISTEN to Your Preferences: An LLM Framework for Multi-Objective Selection
• Beyond Length: Quantifying Long-Range Information for Long-Context LLM Pretraining Data
• Ideology-Based LLMs for Content Moderation
• A Survey on Efficient Large Language Model Training: From Data-centric Perspectives
• RECAP: Reproducing Copyrighted Data from LLMs Training with an Agentic Pipeline
• Semantic Label Drift in Cross-Cultural Translation
• SymCode: A Neurosymbolic Approach to Mathematical Reasoning via Verifiable Code Generation
• NeuronMM: High-Performance Matrix Multiplication for LLM Inference on AWS Trainium
• QCoder Benchmark: Bridging Language Generation and Quantum Hardware through Simulator-Based Feedback
• Reasoning Path Divergence: A New Metric and Curation Strategy to Unlock LLM Diverse Thinking
• On the Influence of Discourse Relations in Persuasive Texts
• MossNet: Mixture of State-Space Experts is a Multi-Head Attention
• Similarity-Distance-Magnitude Language Models
• RCScore: Quantifying Response Consistency in Large Language Models
• Pragmatic Theories Enhance Understanding of Implied Meanings in LLMs
• Language Models Are Borrowing-Blind: A Multilingual Evaluation of Loanword Identification across 10 Languages
• Distilling Multilingual Vision-Language Models: When Smaller Models Stay Multilingual
• Do LLMs Signal When They're Right? Evidence from Neuron Agreement
• SCRIBE: Structured Chain Reasoning for Interactive Behaviour Explanations using Tool Calling
• On the Role of Context for Discourse Relation Classification in Scientific Writing
• OmniEduBench: A Comprehensive Chinese Benchmark for Evaluating Large Language Models in Education
• 1+1>2: A Synergistic Sparse and Low-Rank Compression Method for Large Language Models
• A Multi-agent Large Language Model Framework to Automatically Assess Performance of a Clinical AI Triage Tool
• Hebrew Diacritics Restoration using Visual Representation
• SlideAgent: Hierarchical Agentic Framework for Multi-Page Visual Document Understanding
• Encoder-Decoder or Decoder-Only? Revisiting Encoder-Decoder Large Language Model
• Enhancing Underwater Object Detection through Spatio-Temporal Analysis and Spatial Attention Networks
• FakeZero: Real-Time, Privacy-Preserving Misinformation Detection for Facebook and X
• CAVE: Detecting and Explaining Commonsense Anomalies in Visual Environments
• ORBIT - Open Recommendation Benchmark for Reproducible Research with Hidden Tests
• SP-MCQA: Evaluating Intelligibility of TTS Beyond the Word Level
• Which Way Does Time Flow? A Psychophysics-Grounded Evaluation for Vision-Language Models
• Nexus: Execution-Grounded Multi-Agent Test Oracle Synthesis
• Rethinking Text-to-SQL: Dynamic Multi-turn SQL Interaction for Real-world Database Exploration
• Are LLMs Rigorous Logical Reasoners? Empowering Natural Language Proof Generation by Stepwise Decoding with Contrastive Learning
• The LSCD Benchmark: a Testbed for Diachronic Word Meaning Tasks
• Unstructured Evidence Attribution for Long Context Query Focused Summarization
• Dependency Structure Augmented Contextual Scoping Framework for Multimodal Aspect-Based Sentiment Analysis
• ChartMuseum: Testing Visual Reasoning Capabilities of Large Vision-Language Models
• ClueAnchor: Clue-Anchored Knowledge Reasoning Exploration and Optimization for Retrieval-Augmented Generation
• AI Debate Aids Assessment of Controversial Claims
• SPARTA ALIGNMENT: Collectively Aligning Multiple Language Models through Combat
• Adversarial Paraphrasing: A Universal Attack for Humanizing AI-Generated Text
• Large Language Models Have Intrinsic Meta-Cognition, but Need a Good Lens
• Comparing human and LLM politeness strategies in free production
• Massive Supervised Fine-tuning Experiments Reveal How Data, Layer, and Training Factors Shape LLM Alignment Quality
• Beyond Isolated Dots: Benchmarking Structured Table Construction as Deep Knowledge Extraction
• Enhancing Reasoning Skills in Small Persian Medical Language Models Can Outperform Large-Scale Data Training
• Quality Over Quantity? LLM-Based Curation for a Data-Efficient Audio-Video Foundation Model
• GradEscape: A Gradient-Based Evader Against AI-Generated Text Detectors
• Ditch the Denoiser: Emergence of Noise Robustness in Self-Supervised Learning from Data Curriculum
• Curriculum Abductive Learning
• Seek in the Dark: Reasoning via Test-Time Instance-Level Policy Gradient in Latent Space
• Learning to Insert for Constructive Neural Vehicle Routing Solver
• Nek Minit: Harnessing Pragmatic Metacognitive Prompting for Explainable Sarcasm Detection of Australian and Indian English
• StyleGuard: Preventing Text-to-Image-Model-based Style Mimicry Attacks by Style Perturbations
• Paper2Poster: Towards Multimodal Poster Automation from Scientific Papers
• Learning World Models for Interactive Video Generation
• Towards Predicting Any Human Trajectory In Context
• Efficient Regression-Based Training of Normalizing Flows for Boltzmann Generators
• Incentivizing LLMs to Self-Verify Their Answers
• UniSite: The First Cross-Structure Dataset and Learning Framework for End-to-End Ligand Binding Site Detection
• GenIR: Generative Visual Feedback for Mental Image Retrieval
• Human-assisted Robotic Policy Refinement via Action Preference Optimization
• SAFE: Multitask Failure Detection for Vision-Language-Action Models
• SPARKE: Scalable Prompt-Aware Diversity and Novelty Guidance in Diffusion Models via RKE Score
• The Scales of Justitia: A Comprehensive Survey on Safety Evaluation of LLMs
• Unveiling the Learning Mind of Language Models: A Cognitive Framework and Empirical Study
• AIMeter: Measuring, Analyzing, and Visualizing Energy and Carbon Footprint of AI Workloads
• Controlling Thinking Speed in Reasoning Models
• Vision-and-Language Training Helps Deploy Taxonomic Knowledge but Does Not Fundamentally Alter It
• DSDE: Dynamic Speculative Decoding with KLD Stability for Real-World Serving
• FASL-Seg: Anatomy and Tool Segmentation of Surgical Scenes
• SHA-256 Infused Embedding-Driven Generative Modeling of High-Energy Molecules in Low-Data Regimes
• Optimal Information Combining for Multi-Agent Systems Using Adaptive Bias Learning
• FreIE: Low-Frequency Spectral Bias in Neural Networks for Time-Series Tasks
• Mixture-of-Experts Operator Transformer for Large-Scale PDE Pre-Training
• PRESTO: Preimage-Informed Instruction Optimization for Prompting Black-Box LLMs
• MedVLSynther: Synthesizing High-Quality Visual Question Answering from Medical Documents with Generator-Verifier LMMs
• $\pi_\texttt{RL}$: Online RL Fine-tuning for Flow-based Vision-Language-Action Models
• Topology-Aware Active Learning on Graphs
• Active Learning with Task-Driven Representations for Messy Pools
• Robust GNN Watermarking via Implicit Perception of Topological Invariants
• Modular Linear Tokenization (MLT)
• On the Dataless Training of Neural Networks
• Contrastive Predictive Coding Done Right for Mutual Information Estimation
• A General and Streamlined Differentiable Optimization Framework
• Efficient Online Learning with Predictive Coding Networks: Exploiting Temporal Correlations
• Infrequent Exploration in Linear Bandits
• Exploring Human-AI Conceptual Alignment through the Prism of Chess
• Towards Scaling Laws for Symbolic Regression
• New Money: A Systematic Review of Synthetic Data Generation for Finance
• LLMBisect: Breaking Barriers in Bug Bisection with A Comparative Analysis Pipeline
• Do Not Step Into the Same River Twice: Learning to Reason from Trial and Error
• maxVSTAR: Maximally Adaptive Vision-Guided CSI Sensing with Closed-Loop Edge Model Adaptation for Robust Human Activity Recognition
• STAR: A Privacy-Preserving, Energy-Efficient Edge AI Framework for Human Activity Recognition via Wi-Fi CSI in Mobile and Pervasive Computing Environments
• A Game-Theoretic Spatio-Temporal Reinforcement Learning Framework for Collaborative Public Resource Allocation
• Likely Interpolants of Generative Models
• Empirical Bayesian Multi-Bandit Learning
• Offline Clustering of Preference Learning with Active-data Augmentation
• Model Inversion with Layer-Specific Modeling and Alignment for Data-Free Continual Learning
• On the Impact of Weight Discretization in QUBO-Based SVM Training
• Agent Skills Enable a New Class of Realistic and Trivially Simple Prompt Injections
• UnifiedFL: A Dynamic Unified Learning Framework for Equitable Federation
• Towards Explainable and Reliable AI in Finance
• CorVS: Person Identification via Video Trajectory-Sensor Correspondence in a Real-World Warehouse
• Efficient Generative AI Boosts Probabilistic Forecasting of Sudden Stratospheric Warmings
• Adaptive Context Length Optimization with Low-Frequency Truncation for Multi-Agent Reinforcement Learning
• Multi-Task Learning Based on Support Vector Machines and Twin Support Vector Machines: A Comprehensive Survey
• Co-Evolving Latent Action World Models
• ReSpec: Towards Optimizing Speculative Decoding in Reinforcement Learning Systems
• Quantum Gated Recurrent GAN with Gaussian Uncertainty for Network Anomaly Detection
• Data-Efficient RLVR via Off-Policy Influence Guidance
• Enhancing ECG Classification Robustness with Lightweight Unsupervised Anomaly Detection Filters
• LLMs as In-Context Meta-Learners for Model and Hyperparameter Selection
• Think Outside the Policy: In-Context Steered Policy Optimization
• Polybasic Speculative Decoding Through a Theoretical Perspective
• Higher-Order Regularization Learning on Hypergraphs
• A Three-Stage Bayesian Transfer Learning Framework to Improve Predictions in Data-Scarce Domains
• Boosted Trees on a Diet: Compact Models for Resource-Constrained Devices
• On Measuring Localization of Shortcuts in Deep Networks
• Wasserstein Regression as a Variational Approximation of Probabilistic Trajectories through the Bernstein Basis
• Omnipresent Yet Overlooked: Heat Kernels in Combinatorial Bayesian Optimization
• MSAD: A Deep Dive into Model Selection for Time series Anomaly Detection
• Curly Flow Matching for Learning Non-gradient Field Dynamics
• Tight Differentially Private PCA via Matrix Coherence
• LoRAQuant: Mixed-Precision Quantization of LoRA to Ultra-Low Bits
• How Regularization Terms Make Invertible Neural Networks Bayesian Point Estimators
• Budgeted Multiple-Expert Deferral
• An All-Reduce Compatible Top-K Compressor for Communication-Efficient Distributed Learning
• LSM-MS2: A Foundation Model Bridging Spectral Identification and Biological Interpretation
• On Purely Private Covariance Estimation
• Pre-trained Forecasting Models: Strong Zero-Shot Feature Extractors for Time Series Classification
• Learning Pseudorandom Numbers with Transformers: Permuted Congruential Generators, Curricula, and Interpretability
• RNAGenScape: Property-guided Optimization and Interpolation of mRNA Sequences with Manifold Langevin Dynamics
• Pulsar Detection with Deep Learning
• StreetMath: Study of LLMs' Approximation Behaviors
• Review Based Entity Ranking using Fuzzy Logic Algorithmic Approach: Analysis
• Attention Augmented GNN RNN-Attention Models for Advanced Cybersecurity Intrusion Detection
• Discovering Interpretable Biological Concepts in Single-cell RNA-seq Foundation Models
• Flex-GAD : Flexible Graph Anomaly Detection
• Multimodal Bandits: Regret Lower Bounds and Optimal Algorithms
• Optimizing Mirror-Image Peptide Sequence Design for Data Storage via Peptide Bond Cleavage Prediction
• Beyond Long Context: When Semantics Matter More than Tokens
• Debate2Create: Robot Co-design via Large Language Model Debates
• MIRO: MultI-Reward cOnditioned pretraining improves T2I quality and efficiency
• InputDSA: Demixing then Comparing Recurrent and Externally Driven Dynamics
• Risks and Opportunities in Human-Machine Teaming in Operationalizing Machine Learning Target Variables
• AttnCache: Accelerating Self-Attention Inference for LLM Prefill via Attention Cache
• Enabling Fast and Accurate Neutral Atom Readout through Image Denoising
• Detecting Anomalies in Machine Learning Infrastructure via Hardware Telemetry
• Conformal Prediction Beyond the Horizon: Distribution-Free Inference for Policy Evaluation
• Accelerating Real-World Overtaking in F1TENTH Racing Employing Reinforcement Learning Methods
• $L_1$-norm Regularized Indefinite Kernel Logistic Regression
• Bias-Corrected Data Synthesis for Imbalanced Learning
• ALMGuard: Safety Shortcuts and Where to Find Them as Guardrails for Audio-Language Models
• Robust Super-Capacity SRS Channel Inpainting via Diffusion Models
• Uncertainty-Aware Diagnostics for Physics-Informed Machine Learning
• PVMark: Enabling Public Verifiability for LLM Watermarking Schemes
• A Survey of Heterogeneous Graph Neural Networks for Cybersecurity Anomaly Detection
• SABER: Symbolic Regression-based Angle of Arrival and Beam Pattern Estimator
• Multi-Output Robust and Conjugate Gaussian Processes
• Vectorized Context-Aware Embeddings for GAT-Based Collaborative Filtering
• Representation-Level Counterfactual Calibration for Debiased Zero-Shot Recognition
• Inference-Cost-Aware Dynamic Tree Construction for Efficient Inference in Large Language Models
• Physics-Informed Mixture Models and Surrogate Models for Precision Additive Manufacturing
• Hybrid Physical-Neural Simulator for Fast Cosmological Hydrodynamics
• CYPRESS: Crop Yield Prediction via Regression on Prithvi's Encoder for Satellite Sensing
• Heuristic Adaptation of Potentially Misspecified Domain Support for Likelihood-Free Inference in Stochastic Dynamical Systems
• Action-Driven Processes for Continuous-Time Control
• FlowQ-Net: A Generative Framework for Automated Quantum Circuit Design
• Kimi Linear: An Expressive, Efficient Attention Architecture
• Assessment of the conditional exchangeability assumption in causal machine learning models: a simulation study
• Value Drifts: Tracing Value Alignment During LLM Post-Training
• Multi-Agent Reinforcement Learning for Market Making: Competition without Collusion
• A Process Mining-Based System For The Analysis and Prediction of Software Development Workflows
• Revisiting Multilingual Data Mixtures in Language Model Pretraining
• Application and Validation of Geospatial Foundation Model Data for the Prediction of Health Facility Programmatic Outputs -- A Case Study in Malawi
• WaveVerif: Acoustic Side-Channel based Verification of Robotic Workflows
• Brain-IT: Image Reconstruction from fMRI via Brain-Interaction Transformer
• Supervised Reinforcement Learning: From Expert Trajectories to Step-wise Reasoning
• DARTS: A Drone-Based AI-Powered Real-Time Traffic Incident Detection System
• The Quest for Reliable Metrics of Responsible AI
• Dual Mixture-of-Experts Framework for Discrete-Time Survival Analysis
• Climate Adaptation-Aware Flood Prediction for Coastal Cities Using Deep Learning
• RADRON: Cooperative Localization of Ionizing Radiation Sources by MAVs with Compton Cameras
• PORTool: Tool-Use LLM Training with Rewarded Tree
• Rethinking Cross-lingual Alignment: Balancing Transfer and Cultural Erasure in Multilingual LLMs
• Artificial Intelligence-Enabled Analysis of Radiology Reports: Epidemiology and Consequences of Incidental Thyroid Findings
• SIRAJ: Diverse and Efficient Red-Teaming for LLM Agents via Distilled Structured Reasoning
• Do Students Debias Like Teachers? On the Distillability of Bias Mitigation Methods
• Dynamic VLM-Guided Negative Prompting for Diffusion Models
• Data-driven Projection Generation for Efficiently Solving Heterogeneous Quadratic Programming Problems
• Learning Geometry: A Framework for Building Adaptive Manifold Models through Metric Optimization
• Nirvana: A Specialized Generalist Model With Task-Aware Memory Mechanism
• Network-Constrained Policy Optimization for Adaptive Multi-agent Vehicle Routing
• SAFE: A Novel Approach to AI Weather Evaluation through Stratified Assessments of Forecasts over Earth
• Security Risk of Misalignment between Text and Image in Multi-modal Model
• EgoExo-Con: Exploring View-Invariant Video Temporal Understanding
• WOD-E2E: Waymo Open Dataset for End-to-End Driving in Challenging Long-tail Scenarios
• Beyond Synthetic Benchmarks: Evaluating LLM Performance on Real-World Class-Level Code Generation
• MV-MLM: Bridging Multi-View Mammography and Language for Breast Cancer Diagnosis and Risk Prediction
• Bridging the Gap Between Molecule and Textual Descriptions via Substructure-aware Alignment
• Segmentation over Complexity: Evaluating Ensemble and Hybrid Approaches for Anomaly Detection in Industrial Time Series
• Learning to Manage Investment Portfolios beyond Simple Utility Functions
• Linking Heterogeneous Data with Coordinated Agent Flows for Social Media Analysis
• Accumulative SGD Influence Estimation for Data Attribution
• ConceptScope: Characterizing Dataset Bias via Disentangled Visual Concepts
• Predicting All-Cause Hospital Readmissions from Medical Claims Data of Hospitalised Patients
• Don't Let It Fade: Preserving Edits in Diffusion Language Models via Token Timestep Allocation
• What's In My Human Feedback? Learning Interpretable Descriptions of Preference Data
• Towards Global Retrieval Augmented Generation: A Benchmark for Corpus-Level Reasoning
• Hybrid LLM and Higher-Order Quantum Approximate Optimization for CSA Collateral Management
• Test-Time Alignment of LLMs via Sampling-Based Optimal Control in pre-logit space
• MPRU: Modular Projection-Redistribution Unlearning as Output Filter for Classification Pipelines
• Angular Steering: Behavior Control via Rotation in Activation Space
• A Research Roadmap for Augmenting Software Engineering Processes and Software Products with Generative AI
• Distributional Multi-objective Black-box Optimization for Diffusion-model Inference-time Multi-Target Generation
• Unravelling the Mechanisms of Manipulating Numbers in Language Models
• Can Agent Conquer Web? Exploring the Frontiers of ChatGPT Atlas Agent in Web Games
• Understanding Hardness of Vision-Language Compositionality from A Token-level Causal Lens
• Implicit Bias of Per-sample Adam on Separable Data: Departure from the Full-batch Regime
• Posterior Sampling by Combining Diffusion Models with Annealed Langevin Dynamics
• From Amateur to Master: Infusing Knowledge into LLMs via Automated Curriculum Learning
• GLYPH-SR: Can We Achieve Both High-Quality Image Super-Resolution and High-Fidelity Text Recovery via VLM-guided Latent Diffusion Model?
• Linear Causal Discovery with Interventional Constraints
• MisSynth: Improving MISSCI Logical Fallacies Classification with Synthetic Data
• Reinforcement Learning for Pollution Detection in a Randomized, Sparse and Nonstationary Environment with an Autonomous Underwater Vehicle
• The Geometry of Dialogue: Graphing Language Models to Reveal Synergistic Teams for Multi-Agent Collaboration
• SPG-CDENet: Spatial Prior-Guided Cross Dual Encoder Network for Multi-Organ Segmentation
• Human-in-the-loop Online Rejection Sampling for Robotic Manipulation
• LoCoT2V-Bench: A Benchmark for Long-Form and Complex Text-to-Video Generation
• SSCL-BW: Sample-Specific Clean-Label Backdoor Watermarking for Dataset Ownership Verification
• Personalized Treatment Outcome Prediction from Scarce Data via Dual-Channel Knowledge Distillation and Adaptive Fusion
• Robust Graph Condensation via Classification Complexity Mitigation
• SecureReviewer: Enhancing Large Language Models for Secure Code Review through Secure-aware Fine-tuning
• Counteracting Matthew Effect in Self-Improvement of LVLMs through Head-Tail Re-balancing
• Bayesian Network Fusion of Large Language Models for Sentiment Analysis
• Simulating and Experimenting with Social Media Mobilization Using LLM Agents
• Inside CORE-KG: Evaluating Structured Prompting and Coreference Resolution for Knowledge Graphs
• The Structure of Relation Decoding Linear Operators in Large Language Models
• Adaptive Inverse Kinematics Framework for Learning Variable-Length Tool Manipulation in Robotics
• Multiclass Local Calibration With the Jensen-Shannon Distance
• InfoFlow: Reinforcing Search Agent Via Reward Density Optimization
• Stop Wasting Your Tokens: Towards Efficient Runtime Multi-Agent Systems
• ResMatching: Noise-Resilient Computational Super-Resolution via Guided Conditional Flow Matching
• Aeolus: A Multi-structural Flight Delay Dataset
• Hybrid DQN-TD3 Reinforcement Learning for Autonomous Navigation in Dynamic Environments
• Evontree: Ontology Rule-Guided Self-Evolution of Large Language Models
• Process Integrated Computer Vision for Real-Time Failure Prediction in Steel Rolling Mill
• The End of Manual Decoding: Towards Truly End-to-End Language Models
• On the limitation of evaluating machine unlearning using only a single training seed
• Non-Convex Over-the-Air Heterogeneous Federated Learning: A Bias-Variance Trade-off
• ExpertFlow: Adaptive Expert Scheduling and Memory Coordination for Efficient MoE Inference
• A General Incentives-Based Framework for Fairness in Multi-agent Resource Allocation
• Deep sequence models tend to memorize geometrically; it is unclear why
• AMO-Bench: Large Language Models Still Struggle in High School Math Competitions
• STaMP: Sequence Transformation and Mixed Precision for Low-Precision Activation Quantization
• Faithful and Fast Influence Function via Advanced Sampling
• Clone Deterministic 3D Worlds with Geometrically-Regularized World Models
• Remote Labor Index: Measuring AI Automation of Remote Work
• Defeating the Training-Inference Mismatch via FP16
• Gistify! Codebase-Level Understanding via Runtime Execution
• Are Video Models Ready as Zero-Shot Reasoners? An Empirical Study with the MME-CoF Benchmark
• Plasticity as the Mirror of Empowerment
• Rethinking Optimal Verification Granularity for Compute-Efficient Test-Time Scaling
• MedAgentBoard: Benchmarking Multi-Agent Collaboration with Conventional Methods for Diverse Medical Tasks
• Self-Evolving Curriculum for LLM Reasoning
• Embracing Contradiction: Theoretical Inconsistency Will Not Impede the Road of Building Responsible AI Systems
• TERAG: Token-Efficient Graph-Based Retrieval-Augmented Generation
• Reward Collapse in Aligning Large Language Models
• VerifIoU - Robustness of Object Detection to Perturbations
• Chaos-based reinforcement learning with TD3
• Chain-of-Scrutiny: Detecting Backdoor Attacks for Large Language Models
• A mathematical certification for positivity conditions in Neural Networks with applications to partial monotonicity and Trustworthy AI
• AI's Social Forcefield: Reshaping Distributed Cognition in Human-AI Teams
• Speak & Spell: LLM-Driven Controllable Phonetic Error Augmentation for Robust Dialogue State Tracking
• Constrained Posterior Sampling: Time Series Generation with Hard Constraints
• Language Model Preference Evaluation with Multiple Weak Evaluators
• Vital Insight: Assisting Experts' Context-Driven Sensemaking of Multi-modal Personal Tracking Data Using Visualization and Human-In-The-Loop LLM
• In Defence of Post-hoc Explainability
• UV-Attack: Physical-World Adversarial Attacks for Person Detection via Dynamic-NeRF-based UV Mapping
• Diversity as a Reward: Fine-Tuning LLMs on a Mixture of Domain-Undetermined Data
• More of the Same: Persistent Representational Harms Under Increased Representation
• Reflection on Data Storytelling Tools in the Generative AI Era from the Human-AI Collaboration Perspective
• Language Models can Self-Improve at State-Value Estimation for Better Search
• MindGYM: What Matters in Question Synthesis for Thinking-Centric Fine-Tuning?
• Guided Model Merging for Hybrid Data Learning: Leveraging Centralized Data to Refine Decentralized Models
• Zero-shot Benchmarking: A Framework for Flexible and Scalable Automatic Evaluation of Language Models
• M-Prometheus: A Suite of Open Multilingual LLM Judges
• Empowering Agentic Video Analytics Systems with Video Language Models
• Toward a Public and Secure Generative AI: A Comparative Analysis of Open and Closed LLMs
• Towards Piece-by-Piece Explanations for Chess Positions with SHAP
• An Agentic Framework for Rapid Deployment of Edge AI Solutions in Industry 5.0
• Symbolically Scaffolded Play: Designing Role-Sensitive Prompts for Generative NPC Dialogue
• Through the Judge's Eyes: Inferred Thinking Traces Improve Reliability of LLM Raters
• The Information-Theoretic Imperative: Compression and the Epistemic Foundations of Intelligence
• Approximating Human Preferences Using a Multi-Judge Learned System
• SciTrust 2.0: A Comprehensive Framework for Evaluating Trustworthiness of Large Language Models in Scientific Applications
• FinOps Agent -- A Use-Case for IT Infrastructure and Cost Optimization
• Humains-Junior: A 3.8B Language Model Achieving GPT-4o-Level Factual Accuracy by Directed Exoskeleton Reasoning
• Estimating cognitive biases with attention-aware inverse planning
• From Queries to Insights: Agentic LLM Pipelines for Spatio-Temporal Text-to-SQL
• AutoSurvey2: Empowering Researchers with Next Level Automated Literature Surveys
• Large Language Model-assisted Autonomous Vehicle Recovery from Immobilization
• Can AI be Accountable?
• Lean4Physics: Comprehensive Reasoning Framework for College-level Physics in Lean4
• GUI Knowledge Bench: Revealing the Knowledge Gap Behind VLM Failures in GUI Tasks
• Beyond Benchmarks: The Economics of AI Inference
• Reasoning Curriculum: Bootstrapping Broad LLM Reasoning from Math
• The FM Agent
• One Model to Critique Them All: Rewarding Agentic Tool-Use via Efficient Reasoning
• Questionnaire meets LLM: A Benchmark and Empirical Study of Structural Skills for Understanding Questions and Responses
• Retrieval Augmented Generation-Enhanced Distributed LLM Agents for Generalizable Traffic Signal Control with Emergency Vehicles
• Graph-Enhanced Policy Optimization in LLM Agent Training
• GraphCompliance: Aligning Policy and Context Graphs for LLM-Based Regulatory Compliance
• Discovering State Equivalences in UCT Search Trees By Action Pruning
• BOTS: A Unified Framework for Bayesian Online Task Selection in LLM Reinforcement Finetuning
• AI Mathematician as a Partner in Advancing Mathematical Discovery - A Case Study in Homogenization Theory
• Scales++: Compute Efficient Evaluation Subset Selection with Cognitive Scales Embeddings
• A Pragmatic View of AI Personhood
• Autograder+: A Multi-Faceted AI Framework for Rich Pedagogical Feedback in Programming Education
• MedSAE: Dissecting MedCLIP Representations with Sparse Autoencoders
• Chain-of-Thought Hijacking
• Who Has The Final Say? Conformity Dynamics in ChatGPT's Selections
• LINK-KG: LLM-Driven Coreference-Resolved Knowledge Graphs for Human Smuggling Networks
• Context Engineering 2.0: The Context of Context Engineering
• Human-AI Complementarity: A Goal for Amplified Oversight
• EdgeRunner 20B: Military Task Parity with GPT-5 while Running on the Edge
• Agentic AI Home Energy Management System: A Large Language Model Framework for Residential Load Scheduling
• Normative Reasoning in Large Language Models: A Comparative Benchmark from Logical and Modal Perspectives
• The Era of Agentic Organization: Learning to Organize with Language Models
• Delegated Authorization for Agents Constrained to Semantic Task-to-Scope Matching
• Unveiling Intrinsic Text Bias in Multimodal Large Language Models through Attention Key-Space Analysis
• Cross-Platform Evaluation of Reasoning Capabilities in Foundation Models
• The Oversight Game: Learning to Cooperatively Balance an AI Agent's Safety and Autonomy
• LLMs Process Lists With General Filter Heads
• Magentic Marketplace: An Open-Source Environment for Studying Agentic Markets
• A Practitioner's Guide to Kolmogorov-Arnold Networks
• LASTIST: LArge-Scale Target-Independent STance dataset
• zFLoRA: Zero-Latency Fused Low-Rank Adapters
• HiMAE: Hierarchical Masked Autoencoders Discover Resolution-Specific Structure in Wearable Time Series
• BlackboxNLP-2025 MIB Shared Task: Improving Circuit Faithfulness via Better Edge Selection
• Unsupervised local learning based on voltage-dependent synaptic plasticity for resistive and ferroelectric synapses
• The Kinetics of Reasoning: How Chain-of-Thought Shapes Learning in Transformers?
• Non-myopic Matching and Rebalancing in Large-Scale On-Demand Ride-Pooling Systems Using Simulation-Informed Reinforcement Learning
• MemEIC: A Step Toward Continual and Compositional Knowledge Editing
• Metis-SPECS: Decoupling Multimodal Learning via Self-distilled Preference-based Cold Start
• ScaleDiff: Higher-Resolution Image Synthesis via Efficient and Model-Agnostic Diffusion
• Identity Management for Agentic AI: The new frontier of authorization, authentication, and security for an AI agent world
• AAGATE: A NIST AI RMF-Aligned Governance Platform for Agentic AI
• PRISM: Proof-Carrying Artifact Generation through LLM x MDE Synergy and Stratified Constraints
• Evaluating the Impact of LLM-Assisted Annotation in a Perspectivized Setting: the Case of FrameNet Annotation
• Transferring Causal Effects using Proxies