AI.info

Latest Research — Page 39

Browse Latest Research on AI.info.

• Creator Incentives in Recommender Systems: A Cooperative Game-Theoretic Approach for Stable and Fair Collaboration in Multi-Agent Bandits
• ResWorld: Temporal Residual World Model for End-to-End Autonomous Driving
• SynthAgent: A Multi-Agent LLM Framework for Realistic Patient Simulation -- A Case Study in Obesity with Mental Health Comorbidities
• From Interpretability to Performance: Optimizing Retrieval Heads for Long-Context Language Models
• IntelliProof: An Argumentation Network-based Conversational Helper for Organized Reflection
• Do LLMs Really Struggle at NL-FOL Translation? Revealing their Strengths via a Novel Benchmarking Strategy
• Learning Time-Scale Invariant Population-Level Neural Representations
• ReFIne: A Framework for Trustworthy Large Reasoning Models with Reliability, Faithfulness, and Interpretability
• RIMRULE: Improving Tool-Using Language Agents via MDL-Guided Rule Learning
• Leveraging Generic Time Series Foundation Models for EEG Classification
• Distribution-Aligned Sequence Distillation for Superior Long-CoT Reasoning
• CrochetBench: Can Vision-Language Models Move from Describing to Doing in Crochet Domain?
• Stabilizing Native Low-Rank LLM Pretraining
• FAPO: Flawed-Aware Policy Optimization for Efficient and Reliable Reasoning
• Robust Tabular Foundation Models
• Conformal Prediction in The Loop: A Feedback-Based Uncertainty Model for Trajectory Optimization
• Aligning Tree-Search Policies with Fixed Token Budgets in Test-Time Scaling of LLMs
• Q3R: Quadratic Reweighted Rank Regularizer for Effective Low-Rank Training
• MobileManiBench: Simplifying Model Verification for Mobile Manipulation
• Position: It is Time to Virtualize Foundation Models with a Self-evolving Operating System Layer
• RoadSceneVQA: Benchmarking Visual Question Answering in Roadside Perception Systems for Intelligent Transportation System
• All You Need is One: Capsule Prompt Tuning with a Single Vector
• Enhancing Control Policy Smoothness by Aligning Actions with Predictions from Preceding States
• Improved Balanced Classification with Theoretically Grounded Loss Functions
• Enhancing Multimodal Misinformation Detection by Replaying the Whole Story from Image Modality Perspective
• GSPN-2: Efficient Parallel Sequence Modeling
• SS4D: Native 4D Generative Model via Structured Spacetime Latents
• Inference-Time Scaling of Diffusion Models via Progressive Seed Pruning
• RAMEN: Resolution-Adjustable Multimodal Encoder for Earth Observation
• Twin: Playing an Unknown Game with a Test-Time Digital Twin
• How Robust Are Automated Fact-Checking Systems? A Cross-Benchmark Evaluation
• PolySAE: Modeling Feature Interactions in Sparse Autoencoders via Polynomial Decoding
• Layer-Wise Modality Decomposition for Interpretable Multimodal Sensor Fusion
• HIDISC: A Hyperbolic Framework for Domain Generalization with Generalized Category Discovery
• Identifying Slug Formation in Oil Well Pipelines: A Use Case from Industrial Analytics
• LinCa: Accelerating Diffusion Models via Learnable Decomposed Feature Caching
• Flowing Backwards: Improving Normalizing Flows via Reverse Representation Alignment
• The Bitter Lesson of Diffusion Language Models for Agentic Workflows: A Comprehensive Reality Check
• SafeMIL: Learning Offline Safe Imitation Policy from Non-Preferred Trajectories
• Bridging the Copyright Gap: Do Large Vision-Language Models Recognize and Respect Copyrighted Content?