AI.info

Liang Wang

Explore Liang Wang on AI.info.

• Beyond Closed-Pool Video Retrieval: A Benchmark and Agent Framework for Real-World Video Search and Moment Localization
• OmniSIFT: Modality-Asymmetric Token Compression for Efficient Omni-modal Large Language Models
• How Well Do Models Follow Visual Instructions? VIBE: A Systematic Benchmark for Visual Instruction-Driven Image Editing
• ShotFinder: Imagination-Driven Open-Domain Video Shot Retrieval via Web Search
• ToolWeaver: Weaving Collaborative Semantics for Scalable Tool Use in Large Language Models
• Predict the Retrieval! Test time adaptation for Retrieval Augmented Generation
• Two Pathways to Truthfulness: On the Intrinsic Encoding of LLM Hallucinations
• EchoTrail-GUI: Building Actionable Memory for GUI Agents via Critic-Guided Self-Exploration
• KBQA-R1: Reinforcing Large Language Models for Knowledge Base Question Answering
• SimScale: Learning to Drive via Real-World Simulation at Scale
• GenPilot: A Multi-Agent System for Test-Time Prompt Optimization in Image Generation