Han Li

VCap introduces a novel Witness-Adjudicator reward mechanism that provides highly precise, factually grounded feedback for visual captioning, enabling state-of-the-art performance in RL-trained multimodal models.

VCap introduces a novel Witness-Adjudicator reward mechanism that provides highly precise, factually grounded feedback for visual captioning, enabling state-of-the-art performance in RL-trained multimodal models.

The paper introduces MMG2Skill, a closed-loop framework that converts noisy, human-oriented web guides into editable, executable skills, significantly improving agent performance across diverse tasks.

The paper introduces MMG2Skill, a closed-loop framework that converts noisy, human-oriented web guides into editable, executable skills, significantly improving agent performance across diverse tasks.

The paper proposes OneReason, a framework that enhances the reasoning capability of generative recommendation models by focusing on improving item perception and structuring user behavior into coherent latent interests.

The paper proposes OneReason, a framework that enhances the reasoning capability of generative recommendation models by focusing on improving item perception and structuring user behavior into coherent latent interests.

The paper proposes OneReason, a framework that enhances the reasoning capability of generative recommendation models by focusing on improving item perception and structuring user behavior into coherent latent interests.

EARS is a framework that improves the reliability of sub-agents in centralized multi-agent systems by enabling them to abstain from answering ambiguous or unsupported requests and provide actionable feedback.

The paper presents ASAP, an asynchronous inference system designed to accelerate the prefill phase of Mixture-of-Experts (MoE) models.

The paper proposes WebSwarm, a multi-agent search system that dynamically instantiates agentic search nodes for task decomposition, recursive expansion, and agent collaboration.

The paper proposes VEGAS, a metric for video captioning that uses test-time gaze to sample personalized captions aligning with human focus.

A framework for selective retrieval in iterative graph repair is proposed to improve knowledge graph repair with the help of a researcher agent and a router.