Papers

New on arXiv

Sat 14 Jun 2025

192 papers

CORONA: A Coarse-to-Fine Framework for Graph-based Recommendation with Large Language ModelsLLM reasoning → LLM reasoning for recommendation
Step-by-Step Reasoning Attack: Revealing 'Erased' Knowledge in Large Language ModelsLLM reasoning → Reasoning for agents, planning and code +1
DeepSeq: High-Throughput Single-Cell RNA Sequencing Data Labeling via Web Search-Augmented Agentic Generative AI Foundation ModelsLLM agents and tool use → Medical and scientific domain agents
LIFELONG SOTOPIA: Evaluating Social Intelligence of Language Agents Over Lifelong Social InteractionsLLM agents and tool use → General LLM agent frameworks and benchmarks
Behavioral Generative Agents for Energy OperationsLLM agents and tool use → Financial trading and forecasting agents
Mapping Neural Signals to Agent Performance, A Step Towards Reinforcement Learning from Neural FeedbackDeep reinforcement learning methods → Deep RL algorithms and representations
Performance Plateaus in Inference-Time Scaling for Text-to-Image Diffusion Without External ModelsDiffusion models → Diffusion foundations and sampling
DR-SAC: Distributionally Robust Soft Actor-Critic for Reinforcement Learning under UncertaintyDeep reinforcement learning methods → RL theory and sample complexity
OpenUnlearning: Accelerating LLM Unlearning via Unified Benchmarking of Methods and MetricsLLM safety and security → Machine unlearning
Evaluating AI Alignment in LLMs: Output Analysis of Value Priorities Across 75 Models with Human BenchmarkingLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Not All Tokens and Heads Are Equally Important: Dual-Level Attention Intervention for Hallucination MitigationMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
Towards Building General Purpose Embedding Models for Industry 4.0 AgentsLLM agents and tool use → Web, search, and research agents
An Exploration of Mamba for Speech Self-Supervised ModelsState-space sequence models (S4, Mamba) → Speech and audio SSMs
Automatic Expert Discovery in LLM Upcycling via Sparse Interpolated Mixture-of-ExpertsMixture-of-experts → MoE large language models
On the Robustness of Derivative-free Methods for Linear Quadratic RegulatorDeep reinforcement learning methods → RL for games and combinatorial optimization
A Comprehensive Survey of Deep Research: Systems, Methodologies, and ApplicationsLLM reasoning → Knowledge- and tool-augmented reasoning
OneEval: Benchmarking LLM Knowledge-intensive Reasoning over Diverse Knowledge BasesLLM reasoning → Reasoning benchmarks and evaluation
Calibrating Lightweight Sparse Autoencoder Feature SteeringMechanistic interpretability → SAE features and their applications
DoTA-RAG: Dynamic of Thought Aggregation RAGRetrieval-augmented generation → RAG pipeline design and evaluation
MEraser: An Effective Fingerprint Erasure Approach for Large Language ModelsLLM safety and security → Watermarking generated content and models
Constrained Diffusers for Safe Planning and ControlDiffusion models → Diffusion and flow policies for robotics
PLD: A Choice-Theoretic List-Wise Knowledge DistillationKnowledge distillation → General knowledge distillation methods
Similarity as Reward Alignment: Robust and Versatile Preference-based Reinforcement LearningLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Good Noise Makes Good Edits: A Training-Free Diffusion-Based Video Editing with Image and Text PromptsDiffusion models → Video generation
Graph of Verification: Structured Verification of LLM Reasoning with Directed Acyclic GraphsLLM reasoning → Reasoning benchmarks and evaluation
AgentOrchestra: Orchestrating Multi-Agent Intelligence with the Tool-Environment-Agent(TEA) ProtocolLLM agents and tool use → General LLM agent frameworks and benchmarks
FlexRAG: A Flexible and Comprehensive Framework for Retrieval-Augmented GenerationRetrieval-augmented generation → RAG pipeline design and evaluation
Robust LLM Unlearning with MUDMAN: Meta-Unlearning with Disruption Masking And NormalizationLLM safety and security → Machine unlearning
Tiered Agentic Oversight: A Hierarchical Multi-Agent System for Healthcare SafetyLLM agents and tool use → Medical and scientific domain agents
Generalizable Trajectory Prediction via Inverse Reinforcement Learning with Mamba-Graph ArchitectureState-space sequence models (S4, Mamba) → Video, motion and generative Mamba models +2
CLIP-HandID: Vision-Language Model for Hand-Based Person IdentificationMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
From Outcomes to Processes: Guiding PRM Learning from ORM for Inference-Time AlignmentLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
MS-UMamba: An Improved Vision Mamba Unet for Fetal Abdominal Medical Image SegmentationState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Pushing the Limits of Safety: A Technical Report on the ATLAS Challenge 2025Multimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
Wireless Channel Identification via Conditional Diffusion ModelDiffusion models → Diffusion for compression and wireless
HarMoEny: Efficient Multi-GPU Inference of MoE ModelsMixture-of-experts → Efficient MoE training and inference systems
Cross-Domain Conditional Diffusion Models for Time Series ImputationDiffusion models → Flow matching, time series and science
InverTune: Removing Backdoors from Multimodal Contrastive Learning Models via Trigger Inversion and Activation TuningLLM safety and security → Backdoor and data poisoning attacks
Branch, or Layer? Zeroth-Order Optimization for Continual Learning of Vision-Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
Group then Scale: Dynamic Mixture-of-Experts Multilingual Language ModelMixture-of-experts → MoE large language models
Exploring the Secondary Risks of Large Language ModelsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks
Ghost Policies: A New Paradigm for Understanding and Learning from Failure in Deep Reinforcement LearningDeep reinforcement learning methods → RL for robot locomotion and manipulation
Advances in LLMs with Focus on Reasoning, Adaptability, Efficiency and EthicsLLM reasoning → Reasoning benchmarks and evaluation +1
MM-R5: MultiModal Reasoning-Enhanced ReRanker via Reinforcement Learning for Document RetrievalLLM reasoning → Knowledge- and tool-augmented reasoning
Relative Entropy Regularized Reinforcement Learning for Efficient Encrypted Policy SynthesisDeep reinforcement learning methods → RL for energy, networks and operations
Efficient Reasoning Through Suppression of Self-Affirmation Reflections in Large Reasoning ModelsLLM reasoning → Test-time compute and efficient thinking
Theoretical Tensions in RLHF: Reconciling Empirical Success with Inconsistencies in Social Choice TheoryLLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Why AI Agents Still Need You: Findings from Developer-Agent Collaborations in the WildLLM agents and tool use → Software engineering coding agents
Refract ICL: Rethinking Example Selection in the Era of Million-Token ModelsLLM reasoning → Reasoning benchmarks and evaluation
Image Corruption-Inspired Membership Inference Attacks against Large Vision-Language ModelsMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness +1
SheetMind: Actions Set Accuracy, Agents Set the Failure ModeLLM agents and tool use → Financial trading and forecasting agents
Benchmarking the Trustworthiness in Multimodal LLMs for Video UnderstandingMultimodal LLMs / vision-language models → Video understanding with MLLMs
IndoorWorld: Integrating Physical Task Solving and Social Simulation in A Heterogeneous Multi-Agent EnvironmentLLM agents and tool use → Social simulation and multi-agent debate
GSDNet: Revisiting Incomplete Multimodal-Diffusion from Graph Spectrum Perspective for Conversation Emotion RecognitionDiffusion models → Audio, speech and music
Doctor Approved: Generating Medically Accurate Skin Disease Images through AI-Expert FeedbackLLM alignment / RLHF / preference optimisation → Medical LLM alignment
Med-U1: Incentivizing Unified Medical Reasoning in LLMs via Large-scale Reinforcement LearningLLM reasoning → Clinical and biomedical applications
SPIRE: Conditional Personalization for Federated Diffusion Generative ModelsDiffusion models → Robustness, privacy and adversarial aspects
Unveiling Confirmation Bias in Chain-of-Thought ReasoningLLM reasoning → Prompting and chain-of-thought reasoning
QGuard:Question-based Zero-shot Guard for Multi-modal LLM SafetyLLM safety and security → Safety alignment, refusal and guardrails
The SWE-Bench Illusion: When State-of-the-Art LLMs Remember Instead of ReasonLLM reasoning → Reasoning for agents, planning and code
Deep Fictitious Play-Based Potential Differential Games for Learning Human-Like Interaction at Unsignalized IntersectionsEnd-to-end autonomous driving → Imitation-learning end-to-end planners
Lightweight LSTM Model for Energy Theft Detection via Input Data Reduction
Gradient-based Fine-Tuning through Pre-trained Model Regularization
Microelectrode Signal Dynamics as Biomarkers of Subthalamic Nucleus Entry on Deep Brain Stimulation: A Nonlinear Feature Approach
Instruction Learning Paradigms: A Dual Perspective on White-box and Black-box LLMs
A Framework for Generating Conversational Recommendation Datasets from Behavioral Interactions
A Theoretical Framework for Virtual Power Plant Integration with Gigawatt-Scale AI Data Centers: Multi-Timescale Control and Stability Analysis
Detecting Narrative Shifts through Persistent Structures: A Topological Analysis of Media Discourse
MonoVQD: Monocular 3D Object Detection with Variational Query Denoising and Self-Distillation
Deploying and Evaluating Multiple Deep Learning Models on Edge Devices for Diabetic Retinopathy Detection
Real-Time, Low-Latency Surveillance Using Entropy-Based Adaptive Buffering and MobileNetV2 on Edge Devices
ArchShapeNet:An Interpretable 3D-CNN Framework for Evaluating Architectural Shapes
ReFrame: Layer Caching for Accelerated Inference in Real-Time Rendering
ANIRA: An Architecture for Neural Network Inference in Real-Time Audio Applications
INTERPOS: Interaction Rhythm Guided Positional Morphing for Mobile App Recommender Systems
Synthetic Socratic Debates: Examining Persona Effects on Moral Decision and Persuasion Dynamics
Beyond Sin-Squared Error: Linear-Time Entrywise Uncertainty Quantification for Streaming PCA
Learning Mappings in Mesh-based Simulations
Glocal Smoothness: Line search and adaptive step sizes can help in theory too!
Optimizing Blood Transfusions and Predicting Shortages in Resource-Constrained Areas
How Grounded is Wikipedia? A Study on Structured Evidential Support and Retrieval
Between Predictability and Randomness: Seeking Artistic Inspiration from AI Generative Models
MS4UI: A Dataset for Multi-modal Summarization of User Interface Instructional Videos
Semivalue-based data valuation is arbitrary and gameable
Konooz: Multi-domain Multi-dialect Corpus for Named Entity Recognition
Existence of Adversarial Examples for Random Convolutional Networks via Isoperimetric Inequalities on mathbbso(d)
OscNet v1.5: Energy Efficient Hopfield Network on CMOS Oscillators for Image Classification
Trust-MARL: Trust-Based Multi-Agent Reinforcement Learning Framework for Cooperative On-Ramp Merging Control in Heterogeneous Traffic Flow
Are We Really Measuring Progress? Transferring Insights from Evaluating Recommender Systems to Temporal Link Prediction
DejaVid: Encoder-Agnostic Learned Temporal Matching for Video Classification
Overview of the NLPCC 2025 Shared Task: Gender Bias Mitigation Challenge
StreamMel: Real-Time Zero-shot Text-to-Speech via Interleaved Continuous Autoregressive Modeling
MVP-CBM:Multi-layer Visual Preference-enhanced Concept Bottleneck Model for Explainable Medical Image Classification
Benchmarking Image Similarity Metrics for Novel View Synthesis Applications
Parkinson's Disease Freezing of Gait (FoG) Symptom Detection Using Machine Learning from Wearable Sensor Data
RAW-Explainer: Post-hoc Explanations of Graph Neural Networks on Knowledge Graphs
Language Models Enable Data-Augmented Synthesis Planning for Inorganic Materials
Algorithmic Fairness: Not a Purely Technical but Socio-Technical Property
Neuromorphic Online Clustering and Its Application to Spike Sorting
Beyond Laplace and Gaussian: Exploring the Generalized Gaussian Mechanism for Private Machine Learning

All 192