Papers

New on arXiv

Wed 20 Aug 2025

289 papers

Sealing The Backdoor: Unlearning Adversarial Text Triggers In Diffusion Models Using Knowledge DistillationDiffusion models → Robustness, privacy and adversarial aspects +1
GM-Skip: Metric-Guided Transformer Block Skipping for Efficient Vision-Language ModelsLLM reasoning → Multimodal reasoning +1
Trust but Verify! A Survey on Verification Design for Test-time ScalingLLM reasoning → Test-time compute and efficient thinking
QA-VLM: Providing human-interpretable quality assessment for wire-feed laser additive manufacturing parts with Vision Language ModelsMultimodal LLMs / vision-language models → Chart understanding and anomaly detection
Enabling Multi-Agent Systems as Learning Designers: Applying Learning Sciences to AI Instructional DesignLLM agents and tool use → Educational agents and math problem solving
MSNav: Zero-Shot Vision-and-Language Navigation with Dynamic Memory and LLM Spatial ReasoningLLM reasoning → Spatial and embodied reasoning
Do VLMs Have Bad Eyes? Diagnosing Compositional Failures via Mechanistic InterpretabilityMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP +1
XFinBench: Benchmarking LLMs in Complex Financial Problem Solving and ReasoningLLM reasoning → Reasoning for agents, planning and code
QU-NLP at QIAS 2025 Shared Task: A Two-Phase LLM Fine-Tuning and Retrieval-Augmented Generation Approach for Islamic Inheritance ReasoningLLM reasoning → Reasoning for agents, planning and code +1
DocHop-QA: Towards Multi-Hop Reasoning over Multimodal Document CollectionsLLM reasoning → Knowledge- and tool-augmented reasoning
MedCoT-RAG: Causal Chain-of-Thought RAG for Medical Question AnsweringLLM reasoning → Clinical and biomedical applications
Universal Reinforcement Learning in Coalgebras: Asynchronous Stochastic Computation via ConductionDeep reinforcement learning methods → RL theory and sample complexity
Towards Source-Free Machine UnlearningLLM safety and security → Machine unlearning
Side Effects of Erasing Concepts from Diffusion ModelsDiffusion models → Robustness, privacy and adversarial aspects
Flexible Agent Alignment with Goal Inference from Open-Ended DialogLLM agents and tool use → General LLM agent frameworks and benchmarks +1
LLMs and Agentic AI in Insurance Decision-Making: Opportunities and Challenges For AfricaLLM agents and tool use → Financial trading and forecasting agents
Nemotron-CC-Math: A 133 Billion-Token-Scale High Quality Math Pretraining DatasetLLM reasoning → Mathematical reasoning
Evaluating Sparse Autoencoders for Monosemantic RepresentationMechanistic interpretability → SAE features and their applications
CurveFlow: Curvature-Guided Flow Matching for Image GenerationDiffusion models → Flow matching, time series and science
S3LoRA: Safe Spectral Sharpness-Guided Pruning in Adaptation of Agent PlannerLLM agents and tool use → Agent threats, guardrails and governance
Osprey: Production-Ready Agentic AI for Safety-Critical Control SystemsLLM agents and tool use → Agent threats, guardrails and governance
Don't Think Twice! Over-Reasoning Impairs Confidence CalibrationLLM reasoning → Test-time compute and efficient thinking
Emergent Crowds Dynamics from Language-Driven Multi-Agent InteractionsLLM agents and tool use → Social simulation and multi-agent debate
Reward-Shifted Speculative Sampling Is An Efficient Test-Time Weak-to-Strong AlignerLLM alignment / RLHF / preference optimisation → LLM preference optimization methods +1
MoEcho: Exploiting Side-Channel Attacks to Compromise User Privacy in Mixture-of-Experts LLMsMixture-of-experts → Expert routing strategies
Collab-REC: An LLM-based Agentic Framework for Balancing Recommendations in TourismLLM agents and tool use → Agentic recommender systems
Reversible Unfolding Network for Concealed Visual Perception with Generative RefinementDiffusion models → 3D, point cloud and scene generation
In-Context Iterative Policy Improvement for Dynamic ManipulationLLM reasoning → Spatial and embodied reasoning
TAIGen: Training-Free Adversarial Image Generation via Diffusion ModelsDiffusion models → Robustness, privacy and adversarial aspects
Goals and the Structure of ExperienceWorld models → World models for language agents
Aura-CAPTCHA: A Reinforcement Learning and GAN-Enhanced Multi-Modal CAPTCHA SystemDeep reinforcement learning methods → RL training of language and agent models
Improving LLMs for Machine Translation Using Synthetic Preference DataLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Linear Preference Optimization: Decoupled Gradient Control via Absolute RegularizationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Robust Symbolic Reasoning for Visual Narratives via Hierarchical and Semantically Normalized Knowledge GraphsLLM reasoning → Reasoning benchmarks and evaluation
Compute-Optimal Scaling for Value-Based Deep RLDeep reinforcement learning methods → Deep RL algorithms and representations
Squeezed Diffusion ModelsDiffusion models → Diffusion foundations and sampling
Universal and Transferable Adversarial Attack on Large Language Models Using Exponentiated Gradient DescentLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
Long Chain-of-Thought Reasoning Across LanguagesLLM reasoning → Prompting and chain-of-thought reasoning
From Passive Tool to Socio-cognitive Teammate: A Conceptual Framework for Agentic AI in Human-AI Collaborative LearningLLM agents and tool use → Educational agents and math problem solving
Evaluating Retrieval-Augmented Generation vs. Long-Context Input for Clinical Reasoning over EHRsLLM reasoning → Clinical and biomedical applications +1
Tinker: Diffusion's Gift to 3D--Multi-View Consistent Editing From Sparse Inputs without Per-Scene OptimizationDiffusion models → 3D, point cloud and scene generation
Source-Guided Flow MatchingDiffusion models → Flow matching, time series and science
Synthetic Adaptive Guided Embeddings (SAGE): A Novel Knowledge Distillation MethodKnowledge distillation → General knowledge distillation methods
PepThink-R1: LLM for Interpretable Cyclic Peptide Optimization with CoT SFT and Reinforcement LearningLLM reasoning → Clinical and biomedical applications
Reliable generation of isomorphic physics problems using Generative AI with prompt-chaining and tool useLLM agents and tool use → Web, search, and research agents
HERAKLES: Hierarchical Skill Compilation for Open-ended LLM AgentsLLM agents and tool use → Agent skill learning and evolution
Cross-Modality Controlled Molecule Generation with Diffusion Language ModelDiffusion models → Flow matching, time series and science
MissionHD: Hyperdimensional Refinement of Distribution-Deficient Reasoning Graphs for Video Anomaly DetectionLLM reasoning → Chain-of-thought reasoning for driving
AFABench: A Generic Framework for Benchmarking Active Feature AcquisitionDeep reinforcement learning methods → RL theory and sample complexity
Multiscale Video Transformers for Class Agnostic Segmentation in Autonomous DrivingEnd-to-end autonomous driving → VLMs for autonomous driving
GSFix3D: Diffusion-Guided Repair of Novel Views in Gaussian SplattingDiffusion models → 3D, point cloud and scene generation
ShizhenGPT: Towards Multimodal LLMs for Traditional Chinese MedicineMultimodal LLMs / vision-language models → Medical vision-language models
MCP-Universe: Benchmarking Large Language Models with Real-World Model Context Protocol ServersWorld models → World models for language agents +1
Virtual Multiplex Staining for Histological Images using a Marker-wise Conditioned Diffusion ModelDiffusion models → Medical imaging and segmentation
Entropy-Constrained Strategy Optimization in Urban Floods: A Multi-Agent Framework with LLM and Knowledge Graph IntegrationLLM agents and tool use → Multimodal and embodied agents
LeanGeo: Formalizing Competitional Geometry problems in LeanLLM reasoning → Spatial and embodied reasoning
Can LLM Agents Solve Collaborative Tasks? A Study on Urgency-Aware Planning and CoordinationLLM agents and tool use → General LLM agent frameworks and benchmarks
Who Sees What? Structured Thought-Action Sequences for Epistemic Reasoning in LLMsLLM reasoning → Reasoning for agents, planning and code
Beyond ReLU: Chebyshev-DQN for Enhanced Deep Q-NetworksDeep reinforcement learning methods → Deep RL algorithms and representations
Adversarial Generation and Collaborative Evolution of Safety-Critical Scenarios for Autonomous VehiclesEnd-to-end autonomous driving → Adversarial scenarios and safety testing
DGenCTR: Towards a Universal Generative Paradigm for Click-Through Rate Prediction via Discrete DiffusionDiffusion models → Diffusion recommenders
Vivid-VR: Distilling Concepts from Text-to-Video Diffusion Transformer for Photorealistic Video RestorationDiffusion models → Inverse problems and image restoration
Ouroboros: Single-step Diffusion Models for Cycle-consistent Forward and Inverse RenderingDiffusion models → 3D, point cloud and scene generation
DuPO: Enabling Reliable LLM Self-Verification via Dual Preference OptimizationLLM alignment / RLHF / preference optimisation → Process reward models for reasoning
NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning ModelLLM reasoning → Test-time compute and efficient thinking +1
MUSE: Multi-Subject Unified Synthesis via Explicit Layout Semantic ExpansionDiffusion models → Image generation, text-to-image and editing
FOCUS: Frequency-Optimized Conditioning of DiffUSion Models for mitigating catastrophic forgetting during Test-Time AdaptationDiffusion models → Image generation, text-to-image and editing
HyperDiff: Hypergraph Guided Diffusion Model for 3D Human Pose EstimationDiffusion models → 3D, point cloud and scene generation
Knowledge Graph-Infused Fine-Tuning for Structured Reasoning in Large Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning
Disentanglement in T-space for Faster and Distributed Training of Diffusion Models with Fewer Latent-statesDiffusion models → Diffusion foundations and sampling
ORThought: Benchmarking and Automating Logistics Optimization ModelingLLM reasoning → Reasoning for agents, planning and code
Hallucination-Resistant Relation Extraction via Dependency-Aware Sentence Simplification and Two-tiered Hierarchical RefinementLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Offline Imitation Learning upon Arbitrary Demonstrations by Pre-Training Dynamics RepresentationsOffline RL and imitation learning from data → Imitation learning methods
Action-Constrained Imitation LearningOffline RL and imitation learning from data → Imitation learning methods
Physics-Constrained Diffusion Reconstruction with Posterior Correction for Quantitative and Fast PET ImagingDiffusion models → Medical imaging and segmentation
Organ-Agents: Virtual Human Physiology Simulator via LLMsLLM agents and tool use → Medical and scientific domain agents
SBGD: Improving Graph Diffusion Generative Model via Stochastic Block DiffusionDiffusion models → Flow matching, time series and science
Generative AI Against Poaching: Latent Composite Flow Matching for Wildlife ConservationDiffusion models → Flow matching, time series and science
A Comparative Evaluation of Teacher-Guided Reinforcement Learning Techniques for Autonomous Cyber OperationsDeep reinforcement learning methods → Adversarial robustness and RL in finance
MoVieDrive: Urban Scene Synthesis with Multi-Modal Multi-View Video Diffusion TransformerDiffusion models → Diffusion and flow policies for robotics
Beyond Semantic Similarity: Reducing Unnecessary API Calls via Behavior-Aligned RetrieverLLM agents and tool use → Tool learning and function calling
Beyond Value Elicitation: Towards Moral Profiles in Early Requirements Engineering via Role-Playing Games and Anthropologist LLMs
Understanding Practitioners Perspectives on Monitoring Machine Learning Systems
PerFairX: Is There a Balance Between Fairness and Personality in Large Language Model Recommendations?
Gaussian Process Regression of Steering Vectors With Physics-Aware Deep Composite Kernels for Augmented Listening
SignBind-LLM: Multi-Stage Modality Fusion for Sign Language Translation
From Sound to Sight: Towards AI-authored Music Videos
Mitigating Data Exfiltration Attacks through Layer-Wise Learning Rate Decay Fine-Tuning
Diagnosing Psychiatric Patients: Can Large Language and Machine Learning Models Perform Effectively in Emergency Cases?
DeepEmoNet: Building Machine Learning Models for Automatic Emotion Recognition in Human Speeches
The Variance Paradox: How AI Reduces Diversity but Increases Novelty
Lossless Compression of Neural Network Components: Weights, Checkpoints, and K/V Caches in Low-Precision Formats
Toward Responsible ASR for African American English Speakers: A Scoping Review of Bias and Equity in Speech Technology
Human-like Content Analysis for Generative AI with Language-Grounded Sparse Encoders
Can AI Have a Personality? Prompt Engineering for AI Personality Simulation: A Chatbot Case Study in Gender-Affirming Voice Therapy Training
Pixie: Fast and Generalizable Supervised Learning of 3D Physics from Pixels
The Loupe: A Plug-and-Play Attention Module for Amplifying Discriminative Features in Vision Transformers
Optimizing Hyper parameters in CNN for Soil Classification using PSO and Whale Optimization Algorithm
Leveraging Multi-Source Textural UGC for Neighbourhood Housing Quality Assessment: A GPT-Enhanced Framework
OASIS: Open-world Adaptive Self-supervised and Imbalanced-aware System

All 289