Papers

New on arXiv

Fri 4 Apr 2025

265 papers

How Social is It? A Benchmark for LLMs' Capabilities in Multi-user Multi-turn Social Agent TasksLLM agents and tool use → Social simulation and multi-agent debate
ScreenSpot-Pro: GUI Grounding for Professional High-Resolution Computer UseMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
Improving Mixed-Criticality Scheduling with Reinforcement LearningDeep reinforcement learning methods → RL for energy, networks and operations
I Can Hear You Coming: RF Sensing for Uncooperative Satellite EvasionDeep reinforcement learning methods → RL for robot navigation and UAV control
VideoComp: Advancing Fine-Grained Compositional and Temporal Alignment in Video-Text ModelsMultimodal LLMs / vision-language models → Video understanding with MLLMs
Optimizing UAV Aerial Base Station Flights Using DRL-based Proximal Policy OptimizationDeep reinforcement learning methods → RL for robot navigation and UAV control
Practical Poisoning Attacks against Retrieval-Augmented GenerationRetrieval-augmented generation → Privacy and security of RAG
TGraphX: Tensor-Aware Graph Neural Network for Multi-Dimensional Feature LearningLLM reasoning → Multimodal reasoning
Distillation and Refinement of Reasoning in Small Language Models for Document Re-rankingLLM reasoning → Knowledge- and tool-augmented reasoning +1
YaleNLP @ PerAnsSumm 2025: Multi-Perspective Integration via Mixture-of-Agents for Enhanced Healthcare QA SummarizationLLM agents and tool use → Medical and scientific domain agents
Have Large Language Models Learned to Reason? A Characterization via 3-SAT Phase TransitionLLM reasoning → Reasoning benchmarks and evaluation
Concept-based Rubrics Improve LLM Formative Assessment and Data SynthesisSynthetic data and self-improvement for LLMs → LLM synthetic data generation
HeterMoE: Efficient Training of Mixture-of-Experts Models on Heterogeneous GPUsMixture-of-experts → Efficient MoE training and inference systems
Improving World Models using Deep Supervision with Linear ProbesWorld models → World models for language agents +1
Detection Limits and Statistical Separability of Tree Ring Watermarks in Rectified Flow-based Text-to-Image Generation ModelsDiffusion models → Robustness, privacy and adversarial aspects +1
Do LLM Evaluators Prefer Themselves for a Reason?LLM reasoning → Reasoning benchmarks and evaluation +1
A Hybrid Wavelet-Fourier Method for Next-Generation Conditional Diffusion ModelsDiffusion models → Diffusion foundations and sampling
Hierarchically Encapsulated Representation for Protocol Design in Self-Driving LabsEnd-to-end autonomous driving → Large language models for driving
Decision SpikeFormer: Spike-Driven Transformer for Decision MakingOffline RL and imitation learning from data → Decision transformers and sequence modeling
Sample, Don't Search: Rethinking Test-Time Alignment for Language ModelsLLM reasoning → Reinforcement learning for reasoning +1
MME-Unify: A Comprehensive Benchmark for Unified Multimodal Understanding and Generation ModelsLLM reasoning → Multimodal reasoning
Finding the Minimal Parameter Budget for Implicit Reasoning: A Data Complexity Driven Scaling Law for Language ModelsLLM reasoning → Test-time compute and efficient thinking
Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer ModelsLLM reasoning → Test-time compute and efficient thinking +1
Align to Structure: Aligning Large Language Models with Structural InformationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Multilingual Retrieval-Augmented Generation for Knowledge-Intensive TaskRetrieval-augmented generation → Generation and context integration
AIR: A Systematic Analysis of Annotations, Instructions, and Response Pairs in Preference DatasetLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Multimodal Diffusion Bridge with Attention-Based SAR Fusion for Satellite Image Cloud RemovalDiffusion models → Inverse problems and image restoration
APIGen-MT: Agentic Pipeline for Multi-Turn Data Generation via Simulated Agent-Human InterplayLLM agents and tool use → General LLM agent frameworks and benchmarks
SynWorld: Virtual Scenario Synthesis for Agentic Action Knowledge RefinementLLM agents and tool use → General LLM agent frameworks and benchmarks
Agentic Knowledgeable Self-awarenessLLM agents and tool use → General LLM agent frameworks and benchmarks
RANa: Retrieval-Augmented NavigationRetrieval-augmented generation → Multimodal RAG
Dexterous Manipulation through Imitation Learning: A SurveyOffline RL and imitation learning from data → Robot manipulation imitation learning
Learning Dual-Arm Coordination for Grasping Large Flat ObjectsDeep reinforcement learning methods → RL for robot locomotion and manipulation
Diffusion Active Learning: Towards Data-Driven Experimental Design in Computed TomographyDiffusion models → Medical imaging and segmentation
BUFF: Bayesian Uncertainty Guided Diffusion Probabilistic Model for Single Image Super-ResolutionDiffusion models → Inverse problems and image restoration
Dynamic Importance in Diffusion U-Net for Enhanced Image SynthesisDiffusion models → Diffusion foundations and sampling
D-Garment: Physically Grounded Latent Diffusion for Dynamic Garment DeformationsDiffusion models → 3D, point cloud and scene generation
Generating ensembles of spatially-coherent in-situ forecasts using flow matchingDiffusion models → Flow matching, time series and science
Pyramid-based Mamba Multi-class Unsupervised Anomaly DetectionState-space sequence models (S4, Mamba) → Time series forecasting with SSMs
Know What You do Not Know: Verbalized Uncertainty Estimation Robustness on Corrupted Images in Vision-Language ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Optimizing Quantum Circuits via ZX Diagrams using Reinforcement Learning and Graph Neural NetworksDeep reinforcement learning methods → Quantum reinforcement learning
The AI Cosmologist I: An Agentic System for Automated Data AnalysisLLM agents and tool use → Web, search, and research agents
Autonomous state-space segmentation for Deep-RL sparse reward scenariosDeep reinforcement learning methods → Deep RL algorithms and representations
Online Difficulty Filtering for Reasoning Oriented Reinforcement LearningLLM reasoning → Reinforcement learning for reasoning
Decentralized Collective World Model for Emergent Communication and CoordinationWorld models → World models for language agents
FaR: Enhancing Multi-Concept Text-to-Image Diffusion via Concept Fusion and Localized RefinementDiffusion models → Image generation, text-to-image and editing
Do Large Language Models Solve the Problems of Agent-Based Modeling? A Critical Review of Generative Social SimulationsLLM agents and tool use → Social simulation and multi-agent debate
Inherent and emergent liability issues in LLM-based agentic systems: a principal-agent perspectiveLLM agents and tool use → Agent threats, guardrails and governance
SARLANG-1M: A Benchmark for Vision-Language Modeling in SAR Image UnderstandingMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Think When You Need: Self-Adaptive Chain-of-Thought LearningLLM reasoning → Prompting and chain-of-thought reasoning
Enhancing Personalized Multi-Turn Dialogue with Curiosity RewardLLM alignment / RLHF / preference optimisation → Preference alignment for recommenders and RAG
Explain with Visual Keypoints Like a Real Mentor! A Benchmark for Multimodal Solution ExplanationLLM reasoning → Multimodal reasoning
Mamba as a Bridge: Where Vision Foundation Models Meet Vision Language Models for Domain-Generalized Semantic SegmentationMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP +1
On the Connection Between Diffusion Models and Molecular DynamicsDiffusion models → Flow matching, time series and science
Learning Natural Language Constraints for Safe Reinforcement Learning of Language AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
Multi-lingual Multi-turn Automated Red Teaming for LLMsLLM safety and security → Prompt injection and agent security
PPFPL: Cross-silo Privacy-preserving Federated Prototype Learning Against Data Poisoning AttacksLLM safety and security → Federated unlearning and poisoning
RingMoE: Mixture-of-Modality-Experts Multi-Modal Foundation Models for Universal Remote Sensing Image InterpretationMixture-of-experts → Vision mixture-of-experts
Efficient Dynamic Clustering-Based Document Compression for Retrieval-Augmented-GenerationRetrieval-augmented generation → Generation and context integration
NuScenes-SpatialQA: A Spatial Understanding and Reasoning Benchmark for Vision-Language Models in Autonomous DrivingLLM reasoning → Chain-of-thought reasoning for driving +2
Enhanced Penalty-based Bidirectional Reinforcement Learning AlgorithmsDeep reinforcement learning methods → Deep RL algorithms and representations
DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world EnvironmentsLLM reasoning → Knowledge- and tool-augmented reasoning
TokenFLEX: Unified VLM Training for Flexible Visual Tokens InferenceMultimodal LLMs / vision-language models → Visual token pruning and efficiency
MORAL: A Multimodal Reinforcement Learning Framework for Decision Making in Autonomous LaboratoriesDeep reinforcement learning methods → RL for robot locomotion and manipulation
Why Reasoning Matters? A Survey of Advancements in Multimodal Reasoning (v1)LLM reasoning → Multimodal reasoning
Model Reveals What to Cache: Profiling-Based Feature Reuse for Video Diffusion ModelsDiffusion models → Video generation
LightPROF: A Lightweight Reasoning Framework for Large Language Model on Knowledge GraphLLM reasoning → Knowledge- and tool-augmented reasoning
Les Dissonances: Cross-Tool Harvesting and Polluting in Pool-of-Tools Empowered LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance
Toward a digital twin of U.S. Congress
Measuring Mental Health Variables in Computational Research: Toward Validated, Dimensional, and Transdiagnostic Approaches
Maestoso: An Intelligent Educational Sketching Tool for Learning Music Theory
Kanji Workbook: A Writing-Based Intelligent Tutoring System for Learning Proper Japanese Kanji Writing Technique with Instructor-Emulated Assessment
AI as a deliberative partner fosters intercultural empathy for Americans but fails for Latin American participants
Towards a Multimodal Document-grounded Conversational AI System for Education
Can AI Master Construction Management (CM)? Benchmarking State-of-the-Art Large Language Models on CM Certification Exams
From Tokens to Lattices: Emergent Lattice Structures in Language Models
The Lyme Disease Controversy: An AI-Driven Discourse Analysis of a Quarter Century of Academic Debate and Divides
Metamorphic Testing for Fairness Evaluation in Large Language Models: Identifying Intersectional Bias in LLaMA and GPT
Generative AI Enhanced Financial Risk Management Information Retrieval
Temporal-contextual Event Learning for Pedestrian Crossing Intent Prediction
Optimizing 4D Gaussians for Dynamic Scene Video from Single Landscape Images
Level Generation with Constrained Expressive Range
When is using AI the rational choice? The importance of counterfactuals in AI deployment decisions
Not someone, but something: Rethinking trust in the age of medical AI
Regression Discontinuity Design with Distribution-Valued Outcomes
Algorithmic Prompt Generation for Diverse Human-like Teaming and Communication with Large Language Models
CORTEX-AVD: A Framework for CORner Case Testing and EXploration in Autonomous Vehicle Development
Structured Extraction of Process Structure Properties Relationships in Materials Science
V-CEM: Bridging Performance and Intervenability in Concept-based Models
OLAF: An Open Life Science Analysis Framework for Conversational Bioinformatics Powered by Large Language Models
GREATERPROMPT: A Unified, Customizable, and High-Performing Open-Source Toolkit for Prompt Optimization
Bridging LMS and generative AI: dynamic course content integration (DCCI) for enhancing student satisfaction and engagement via the ask ME assistant
Clinical ModernBERT: An efficient and long context encoder for biomedical text
DeepOHeat-v1: Efficient Operator Learning for Fast and Trustworthy Thermal Simulation and Optimization in 3D-IC Design
A New Approach to Controlling Linear Dynamical Systems
Understanding EFX Allocations: Counting and Variants
ProbRes: Probabilistic Jump Diffusion for Open-World Egocentric Activity Recognition
Multi-Variable Batch Bayesian Optimization in Materials Research: Synthetic Data Analysis of Noise Sensitivity and Problem Landscape Effects
Analysis of Robustness of a Large Game Corpus
Language Models Are Implicitly Continuous

All 265