Papers

New on arXiv

Sun 23 Feb 2025

184 papers

Optimizing Retrieval-Augmented Generation of Medical Content for Spaced Repetition LearningRetrieval-augmented generation → Medical and biomedical RAG
Swallowing the Poison Pills: Insights from Vulnerability Disparity Among LLMsLLM safety and security → Backdoor and data poisoning attacks
RewardDS: Privacy-Preserving Fine-Tuning for Large Language Models via Reward Driven Data SynthesisSynthetic data and self-improvement for LLMs → Privacy-preserving tabular data synthesis
A Reverse Mamba Attention Network for Pathological Liver SegmentationState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Ensemble RL through Classifier Models: Enhancing Risk-Return Trade-offs in Trading StrategiesDeep reinforcement learning methods → Adversarial robustness and RL in finance
Guardians of the Agentic System: Preventing Many Shots Jailbreak with Agentic SystemLLM agents and tool use → Agent threats, guardrails and governance +1
Keeping up with dynamic attackers: Certifying robustness to adaptive online data poisoningLLM safety and security → Backdoor and data poisoning attacks
Towards Optimal Adversarial Robust Reinforcement Learning with Infinity Measurement ErrorDeep reinforcement learning methods → Adversarial robustness and RL in finance
RapidPen: Fully Automated IP-to-Shell Penetration Testing with LLM-based AgentsLLM agents and tool use → Agent threats, guardrails and governance
Layer-Wise Evolution of Representations in Fine-Tuned Transformers: Insights from Sparse AutoEncodersMechanistic interpretability → SAE features and their applications
Exploring Incremental Unlearning: Techniques, Challenges, and Future DirectionsLLM safety and security → Machine unlearning
Reflective Planning: Vision-Language Models for Multi-Stage Long-Horizon Robotic ManipulationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Evaluating the Robustness and Accuracy of Text Watermarking Under Real-World Cross-Lingual ManipulationsLLM safety and security → Watermarking generated content and models
Responsible Federated LLMs via Safety Filtering and Constitutional AILLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
From Text to Space: Mapping Abstract Spatial Models in LLMs during a Grid-World Navigation TaskLLM agents and tool use → Multimodal and embodied agents
Are Sparse Autoencoders Useful? A Case Study in Sparse ProbingMechanistic interpretability → SAE features and their applications
SBSC: Step-By-Step Coding for Improving Mathematical Olympiad PerformanceLLM reasoning → Mathematical reasoning
BioMaze: Benchmarking and Enhancing Large Language Models for Biological Pathway ReasoningLLM reasoning → Clinical and biomedical applications
CODESYNC: Synchronizing Large Language Models with Dynamic Code Evolution at ScaleLLM alignment / RLHF / preference optimisation → Process reward models for reasoning
Retrieval-Augmented Visual Question Answering via Built-in Autoregressive Search EnginesRetrieval-augmented generation → Multimodal RAG
Visual-RAG: Benchmarking Text-to-Image Retrieval Augmented Generation for Visual Knowledge Intensive QueriesRetrieval-augmented generation → Multimodal RAG
OptionZero: Planning with Learned OptionsDeep reinforcement learning methods → RL for games and combinatorial optimization
Can Large Vision-Language Models Detect Images Copyright Infringement from GenAI?Multimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
CodeCriticBench: A Holistic Code Critique Benchmark for Large Language ModelsLLM reasoning → Reasoning for agents, planning and code
Reasoning about Affordances: Causal and Compositional Reasoning in LLMsLLM reasoning → Reasoning benchmarks and evaluation
VidLBEval: Benchmarking and Mitigating Language Bias in Video-Involved LVLMsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Diagnosing Moral Reasoning Acquisition in Language Models: Pragmatics and GeneralizationLLM reasoning → Reasoning benchmarks and evaluation
Tracking the Copyright of Large Vision-Language Models through Parameter Learning Adversarial ImagesMultimodal LLMs / vision-language models → MLLM jailbreaks and adversarial robustness
Multimodal Large Language Models for Text-rich Image Understanding: A Comprehensive ReviewMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Can Indirect Prompt Injection Attacks Be Detected and Removed?LLM safety and security → Prompt injection and agent security
LawPal : A Retrieval Augmented Generation Based System for Enhanced Legal Accessibility in IndiaRetrieval-augmented generation → Legal and financial RAG
The Hidden Strength of Disagreement: Unraveling the Consensus-Diversity Tradeoff in Adaptive Multi-Agent SystemsLLM agents and tool use → Social simulation and multi-agent debate
Retrieval-Augmented Fine-Tuning With Preference Optimization For Visual Program GenerationRetrieval-augmented generation → RAG for code +1
OpenVox: Real-time Instance-level Open-vocabulary Probabilistic Voxel RepresentationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
Intrinsic Model Weaknesses: How Priming Attacks Unveil Vulnerabilities in Large Language ModelsLLM safety and security → Safety alignment, refusal and guardrails
MAPN: Enhancing Heterogeneous Sparse Graph Representation by Mamba-based Asynchronous AggregationState-space sequence models (S4, Mamba) → Video, motion and generative Mamba models
Iterative Flow Matching: Path Correction and Gradual Refinement for Enhanced Generative ModelingDiffusion models → Flow matching, time series and science
Sequence-level Large Language Model Training with Contrastive Preference OptimizationLLM alignment / RLHF / preference optimisation → LLM preference optimization methods
Visual Reasoning Evaluation of Grok, Deepseek Janus, Gemini, Qwen, Mistral, and ChatGPTLLM reasoning → Multimodal reasoning +1
Fine-Grained Captioning of Long Videos through Scene Graph ConsolidationMultimodal LLMs / vision-language models → Video understanding with MLLMs
Unified Prompt Attack Against Text-to-Image Generation ModelsLLM safety and security → Vision-language and text-to-image safety
Quadruped Robot Simulation Using Deep Reinforcement Learning -- A step towards locomotion policyDeep reinforcement learning methods → RL for robot locomotion and manipulation
FedNIA: Noise-Induced Activation Analysis for Mitigating Data Poisoning in Federated LearningLLM safety and security → Federated unlearning and poisoning
Enhancing the Aesthetic Appeal of AI-Generated Physical Product Designs through LoRA Fine-Tuning with Human Feedback
Predictive Modeling: BIM Command Recommendation Based on Large-scale Usage Logs
Efficient Multi-Task Learning via Generalist Recommender
Beyond Final Answers: Evaluating Large Language Models for Math Tutoring
Users Favor LLM-Generated Content -- Until They Know It's AI
Optimizing Coverage-Driven Verification Using Machine Learning and PyUVM: A Novel Approach
The Lazy Student's Dream: ChatGPT Passing an Engineering Course on Its Own
MV-CLAM: Multi-View Molecular Interpretation with Cross-Modal Projection via Language Model
A Review of Artificial Intelligence Impacting Statistical Process Monitoring and Future Directions
A Systematic Review of Open Datasets Used in Text-to-Image (T2I) Gen AI Model Safety
A Comparative Tutorial of the Histogram-based Image Segmentation Methods
End-to-End Deep Learning for Structural Brain Imaging: A Unified Framework
Rewards-based image analysis in microscopy
Optimized Custom CNN for Real-Time Tomato Leaf Disease Detection
Class-Conditional Neural Polarizer: A Lightweight and Effective Backdoor Defense by Purifying Poisoned Features
FreeTumor: Large-Scale Generative Tumor Synthesis in Computed Tomography Images for Improving Tumor Recognition
Gradient entropy (GradEn): The two dimensional version of slope entropy for image analysis
Liver Cirrhosis Stage Estimation from MRI with Deep Learning
Multimodal Bearing Fault Classification Under Variable Conditions: A 1D CNN with Transfer Learning
UNCA: A Neutrosophic-Based Framework for Robust Clustering and Enhanced Data Interpretation
Spectral Theory for Edge Pruning in Asynchronous Recurrent Graph Neural Networks
Recent Advances in Large Langauge Model Benchmarks against Data Contamination: From Static to Dynamic Evaluation
On Neural Inertial Classification Networks for Pedestrian Activity Recognition
Attention-based UAV Trajectory Optimization for Wireless Power Transfer-assisted IoT Systems
GS-TransUNet: Integrated 2D Gaussian Splatting and Transformer UNet for Accurate Skin Lesion Analysis
SQLong: Enhanced NL2SQL for Longer Contexts with LLMs
BAGEL: Adversarially Constrained Online Convex Optimization under Separation Oracle Access
Adaptive Conformal Guidance for Learning under Uncertainty
Coreset Selection via LLM-based Concept Bottlenecks
DeepSeek reshaping healthcare in China's tertiary hospitals
Fitting networks with a cancellation trick
DOSE3 : Diffusion-based Out-of-distribution detection on SE(3) trajectories
To Share or Not to Share: Investigating Weight Sharing in Variational Graph Autoencoders
Speed and Conversational Large Language Models: Not All Is About Tokens per Second
NatSGLD: A Dataset with Speech, Gesture, Logic, and Demonstration for Robot Learning in Natural Human-Robot Interaction
Beyond Pattern Recognition: Probing Mental Representations of LMs
Understanding the Impact of Artificial Intelligence in Academic Writing: Metadata to the Rescue
FedDA-TSformer: Federated Domain Adaptation with Vision TimeSformer for Left Ventricle Segmentation on Gated Myocardial Perfusion SPECT Image
DISC: Dynamic Decomposition Improves LLM Inference Scaling
Can ChatGPT Learn to Count Letters?
Code Summarization Beyond Function Level
Subsampling Graphs with GNN Performance Guarantees
Beyond Release: Access Considerations for Generative AI Systems
Interpretable Retinal Disease Prediction Using Biology-Informed Heterogeneous Graph Representations
Dynamic LLM Routing and Selection based on User Preferences: Balancing Performance, Cost, and Ethics
Analyzing Factors Influencing Driver Willingness to Accept Advanced Driver Assistance Systems
WildLong: Synthesizing Realistic Long-Context Instruction Data at Scale
Automatic Input Rewriting Improves Translation with Large Language Models
AeroReformer: Aerial Referring Transformer for UAV-based Referring Image Segmentation
MimeQA: Towards Socially-Intelligent Nonverbal Foundation Models
MetaSym: A Symplectic Meta-learning Framework for Physical Intelligence
A new framework for X-ray absorption spectroscopy data analysis based on machine learning: XASDAML
Geometric Kolmogorov-Arnold Superposition Theorem
Saarthi: The First AI Formal Verification Engineer
Optimizing Input Data Collection for Ranking and Selection
Volume Optimality in Conformal Prediction with Structured Prediction Sets
VPNeXt -- Rethinking Dense Decoding for Plain Vision Transformer

All 184