AI.info

Latest Research — Page 98

Browse Latest Research on AI.info.

• CG-MLLM: Captioning and Generating 3D content via Multi-modal Large Language Models
• Neighbor GRPO: Contrastive ODE Policy Optimization Aligns Flow Models
• Characterizing Human Semantic Navigation in Concept Production as Trajectories in Embedding Space
• Curly Flow Matching for Learning Non-gradient Field Dynamics
• When Is a Task Vector Enough? An Empirical Theory of Implicit Multimodal ICL
• Constrained CTC Decoding for Efficient Diacritic Restoration
• Imputation Uncertainty in Interpretable Machine Learning Methods
• g-DPO: Scalable Preference Optimization for Protein Language Models
• Interpretable Embeddings with Sparse Autoencoders: A Data Analysis Toolkit
• Scaling Spatial Intelligence with Multimodal Foundation Models
• Quadratic Direct Forecast for Training Multi-Step Time-Series Forecast Models
• Adapting from Downturns: Prediction of Long-Term Conversational-Skill Development in Mental-Health Crisis Counselors
• Learning Spatial-Aware Manipulation Ordering
• Adobe Summit Concierge Evaluation with Human in the Loop
• ActFovea: Runtime Safeguarding for VLA Policies via Spatiotemporal Visual-Action Consistency
• Maximum Mean Discrepancy with Unequal Sample Sizes via Generalized U-Statistics
• Lying with Truths: Open-Channel Multi-Agent Collusion for Belief Manipulation via Generative Montage
• PosterVerse: A Full-Workflow Framework for Commercial-Grade Poster Generation with HTML-Based Scalable Typography
• Temporal Graph Pattern Machine
• MobileFetalCLIP: Selective Repulsive Knowledge Distillation for Mobile Fetal Ultrasound Analysis
• MiNER: A Two-Stage Pipeline for Metadata Extraction from Municipal Meeting Minutes
• HGATSolver: A Heterogeneous Graph Attention Solver for Fluid-Structure Interaction
• Teaching Models to Teach Themselves: Reasoning at the Edge of Learnability
• Flexible Camera Calibration using a Collimator System
• A Decomposable Forward Process in Diffusion Models for Time-Series Forecasting
• Prompt-Based Safety Guidance Is Ineffective for Unlearned Text-to-Image Diffusion Models
• MedTutor: A Retrieval-Augmented LLM System for Case-Based Medical Education
• Agent Inheritance Protocol: Speculating on Feralized Agents After Principals Die
• Any3D-VLA: Enhancing VLA Robustness via Diverse Point Clouds
• TAMEing Long Contexts in Personalization: Towards Training-Free and State-Aware MLLM Personalized Assistant
• Bayesian Evaluation of Large Language Model Behavior
• PerTouch: VLM-Driven Agent for Personalized and Semantic Image Retouching
• AISPA: User-Centric System Prompt Auditing for Large Language Model Applications
• FedPS: Federated Preprocessing for structured data via aggregated Statistics
• Entropy-Guided Dynamic Tokens for Graph-LLM Alignment in Molecular Understanding
• Quadratic Upper Bound for Boosting Robustness
• Routing Matters in MoE: Scaling Diffusion Transformers with Explicit Routing Guidance
• Robust Watermarking on Gradient Boosting Decision Trees
• No One Wins in Nuclear War: A Social Simulation of Military Decision-making
• When Shared Knowledge Hurts: Spectral Over-Accumulation in Model Merging