Papers

New on arXiv

Thu 25 Dec 2025

163 papers

Context Discipline and Performance Correlation: Analyzing LLM Performance and Quality Degradation Under Varying Context LengthsMixture-of-experts → MoE large language models
Horizon Reduction as Information Loss in Offline Reinforcement LearningOffline RL and imitation learning from data → Offline RL algorithms
Can We Trust AI Explanations? Evidence of Systematic Underreporting in Chain-of-Thought ReasoningLLM reasoning → Prompting and chain-of-thought reasoning
Can Generative Models Actually Forge Realistic Identity Documents?Diffusion models → Image generation, text-to-image and editing
Enforcing Temporal Constraints for LLM AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
Hybrid Quantum-Classical Mixture of Experts: Unlocking Topological Advantage via Interference-Based RoutingMixture-of-experts → MoE theory and gating functions
Learning from Negative Examples: Why Warning-Framed Training Data Teaches What It Warns AgainstMechanistic interpretability → Probing deception, bias and beliefs
Co-GRPO: Co-Optimized Group Relative Policy Optimization for Masked Diffusion ModelLLM reasoning → Reinforcement learning for reasoning +1
FETAL-GAUGE: A Benchmark for Assessing Vision-Language Models in Fetal UltrasoundMultimodal LLMs / vision-language models → Medical vision-language models
The Illusion of Clinical Reasoning: A Benchmark Reveals the Pervasive Gap in Vision-Language Models for Clinical CompetencyLLM reasoning → Clinical and biomedical applications +1
Human-Aligned Generative Perception: Bridging Psychophysics and Generative ModelsDiffusion models → 3D, point cloud and scene generation
CellMamba: Adaptive Mamba for Accurate and Efficient Cell DetectionState-space sequence models (S4, Mamba) → Mamba for medical image analysis
Diffusion Posterior Sampling for Super-Resolution under Gaussian Measurement NoiseDiffusion models → Inverse problems and image restoration
InstructMoLE: Instruction-Guided Mixture of Low-rank Experts for Multi-Conditional Image GenerationDiffusion models → Diffusion foundations and sampling
Accelerating Scientific Discovery with Autonomous Goal-evolving AgentsLLM agents and tool use → Web, search, and research agents
Scene-VLM: Multimodal Video Scene Segmentation via Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Assessing the Effectiveness of Membership Inference on Generative MusicLLM safety and security → Membership inference attacks
Knot Forcing: Taming Autoregressive Video Diffusion Models for Real-time Infinite Interactive Portrait AnimationDiffusion models → Video generation
PERELMAN: Pipeline for scientific literature meta-analysis. Technical reportLLM agents and tool use → Web, search, and research agents
MAction-SocialNav: Multi-Action Socially Compliant Navigation via Reasoning-enhanced Prompt TuningLLM reasoning → Spatial and embodied reasoning
An Information Theoretic Perspective on Agentic System DesignLLM agents and tool use → Agent memory systems
AstraNav-World: World Model for Foresight Control and ConsistencyWorld models → World models for navigation
Do Latent Tokens Think? A Causal and Adversarial Analysis of Chain-of-Continuous-ThoughtLLM reasoning → Prompting and chain-of-thought reasoning
MoRAgent: Parameter Efficient Agent Tuning with Mixture-of-RolesLLM agents and tool use → General LLM agent frameworks and benchmarks
Spatiotemporal-Untrammelled Mixture of Experts for Multi-Person Motion PredictionMixture-of-experts → MoE in reinforcement learning and robotics +1
Enabling Conversational Behavior Reasoning Capabilities in Full-Duplex SpeechLLM reasoning → Audio-language model reasoning
Towards Responsible and Explainable AI Agents with Consensus-Driven ReasoningLLM reasoning → Reasoning for agents, planning and code +1
Exploring the Security Threats of Retriever Backdoors in Retrieval-Augmented Code GenerationRetrieval-augmented generation → Privacy and security of RAG +1
UniPercept: Towards Unified Perceptual-Level Image Understanding across Aesthetics, Quality, Structure, and TextureMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
The Deepfake Detective: Interpreting Neural Forensics Through Sparse Features and ManifoldsMechanistic interpretability → Sparse autoencoder training and evaluation
Rethinking Output Alignment For 1-bit Post-Training Quantization of Large Language ModelsLLM reasoning → Test-time compute and efficient thinking
Variance-Aware Prior-Based Tree Policies for Monte Carlo Tree SearchDeep reinforcement learning methods → RL theory and sample complexity
Rethinking Sample Polarity in Reinforcement Learning with Verifiable RewardsLLM reasoning → Reinforcement learning for reasoning
TAMEing Long Contexts in Personalization: Towards Training-Free and State-Aware MLLM Personalized AssistantMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
AMS-IO-Bench and AMS-IO-Agent: Benchmarking and Structured Reasoning for Analog and Mixed-Signal Integrated Circuit Input/Output DesignLLM agents and tool use → Software engineering coding agents
Residual Prior Diffusion: A Probabilistic Framework Integrating Coarse Latent Priors with Diffusion ModelsDiffusion models → Diffusion foundations and sampling
UltraLBM-UNet: Ultralight Bidirectional Mamba-based Model for Skin Lesion SegmentationState-space sequence models (S4, Mamba) → Mamba for medical image analysis
A Medical Multimodal Diagnostic Framework Integrating Vision-Language Models and Logic Tree ReasoningLLM reasoning → Clinical and biomedical applications +1
NEMO-4-PAYPAL: Leveraging NVIDIA's Nemo Framework for empowering PayPal's Commerce AgentLLM agents and tool use → Social simulation and multi-agent debate
A Unified Definition of Hallucination: It's The World Model, Stupid!World models → World models for language agents
Towards Long-window Anchoring in Vision-Language Model DistillationMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
Beyond Heuristics: A Decision-Theoretic Framework for Agent Memory ManagementLLM agents and tool use → Agent memory systems
Human-AI Interaction Alignment: Designing, Evaluating, and Evolving Value-Centered AI For Reciprocal Human-AI FuturesLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
Leash: Adaptive Length Penalty and Reward Shaping for Efficient Large Reasoning ModelLLM reasoning → Test-time compute and efficient thinking
Hierarchy-Aware Fine-Tuning of Vision-Language ModelsMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
Generative Actor CriticDeep reinforcement learning methods → RL with diffusion and flow models
Selective LLM-Guided Regularization for Enhancing Recommendation ModelsLLM reasoning → LLM reasoning for recommendation
DiverseGRPO: Mitigating Mode Collapse in Image Generation via Diversity-Aware GRPOLLM reasoning → Reinforcement learning for reasoning
Efficient MoE Inference with Fine-Grained Scheduling of Disaggregated Expert ParallelismMixture-of-experts → Efficient MoE training and inference systems +1
LogicLens: Visual-Logical Co-Reasoning for Text-Centric Forgery AnalysisLLM reasoning → Multimodal reasoning
The AI Committee: A Multi-Agent Framework for Automated Validation and Remediation of Web-Sourced DataLLM agents and tool use → Web, search, and research agents
Analysis of LLM Vulnerability to GPU Soft Errors: An Instruction-Level Fault Injection Study
Compass-Embedding v4: Robust Contrastive Learning for Multilingual E-commerce Embeddings
Context-Aware Classification and Grading of Sensitive Information in Online Conversational Health Data
Bayesian Recovery for Probabilistic Coalition Structures
Self-Supervised Masked Autoencoders with Dense-Unet for Coronary Calcium Removal in limited CT Data
WearVox: An Egocentric Multichannel Voice Assistant Benchmark for Wearables
A Knowledge Graph and Deep Learning-Based Semantic Recommendation Database System for Advertisement Retrieval and Personalization
ShrimpXNet: A Transfer Learning Framework for Shrimp Disease Classification with Augmented Regularization, Adversarial Training, and Explainable AI
Towards representation agnostic probabilistic programming
Break Out the Silverware -- Semantic Understanding of Stored Household Items
Learning Dynamic Scene Reconstruction with Sinusoidal Geometric Priors
A Three-Level Alignment Framework for Large-Scale 3D Retrieval and Controlled 4D Generation
Multi-Head Spectral-Adaptive Graph Anomaly Detection
When Algorithms Manage Humans: A Double Machine Learning Approach to Estimating Nonlinear Effects of Algorithmic Control on Gig Worker Performance and Wellbeing
Cluster Aggregated GAN (CAG): A Cluster-Based Hybrid Model for Appliance Pattern Generation
A General Weighting Theory for Ensemble Learning: Beyond Variance Reduction via Spectral and Geometric Structure
On Fibonacci Ensembles: An Alternative Approach to Ensemble Learning Inspired by the Timeless Architecture of the Golden Ratio
Synergizing Kolmogorov-Arnold Networks with Dynamic Adaptive Weighting for High-Frequency and Multi-Scale PDE Solutions
Nonnegative matrix factorizations and related compositional models: Equivalence, identifiability, and an application on the grain-size analysis of sediments
Valori: A Deterministic Memory Substrate for AI Systems
Hierarchical Stacking Optimization Using Dirichlet's Process (SoDip): Towards Accelerated Design for Graft Polymerization
GeCo: Evaluating Geometric Consistency for Video Generation via Motion and Structure
S&P 500 Stock's Movement Prediction using CNN
Smart IoT-Based Leak Forecasting and Detection for Energy-Efficient Liquid Cooling in AI Data Centers
Deep Generative Models for Synthetic Financial Data: Applications to Portfolio and Risk Modeling
Multi-agent Adaptive Mechanism Design
AI for Mycetoma Diagnosis in Histopathological Images: The MICCAI 2024 Challenge
Synthetic Financial Data Generation for Enhanced Financial Modelling
Five Years of SciCap: What We Learned and Future Directions for Scientific Figure Captioning
Ara-HOPE: Human-Centric Post-Editing Evaluation for Dialectal Arabic to Modern Standard Arabic Translation
VAMP-Net: An Interpretable Multi-Path Network of Genomic Permutation-Invariant Set Attention and Quality-Aware 1D-CNN for MTB Drug Resistance
Inference-based GAN Video Generation
Compliance Rating Scheme: A Data Provenance Framework for Generative AI Datasets
BertsWin: Resolving Topological Sparsity in 3D Masked Autoencoders via Component-Balanced Structural Optimization
A-QCF-Net: An Adaptive Quaternion Cross-Fusion Network for Multimodal Liver Tumor Segmentation from Unpaired Datasets
How Do Agents Perform Code Optimization? An Empirical Study
Approximation Capabilities of Feedforward Neural Networks with GELU Activations
Modified TSception for Analyzing Driver Drowsiness and Mental Workload from EEG
A Model of Causal Explanation on Neural Networks for Tabular Data
Dynamic Feedback Engines: Layer-Wise Control for Self-Regulating Continual Learning
SyncAnyone: Implicit Disentanglement via Progressive Self-Correction for Lip-Syncing in the wild
HELP: Hierarchical Embodied Language Planner for Household Tasks
Multiconnectivity for SAGIN: Current Trends, Challenges, AI-driven Solutions, and Opportunities
CATCH: A Controllable Theme Detection Framework with Contextualized Clustering and Hierarchical Generation
RAPTOR: Real-Time High-Resolution UAV Video Prediction with Efficient Video Attention
Detecting AI-Generated Paraphrases in Bengali: A Comparative Study of Zero-Shot and Fine-Tuned Transformers
Zero-Shot to Zero-Lies: Detecting Bengali Deepfake Audio through Transfer Learning
FUSE: Unifying Spectral and Semantic Cues for Robust AI-Generated Image Detection
BeHGAN: Bengali Handwritten Word Generation from Plain Text Using Generative Adversarial Networks

All 163