الباحثون

Weijie Zhu

المنشورات 2

نسخة أولية وصول مفتوح

MeSD: Multi-Evidence Self-Distillation for VideoLLM

Weijie Zhu, Han Fang, Hanyu Fu وآخرون · 2026

While reinforcement learning with verifiable rewards provides reliable outcome supervision for VideoLLMs, sequence-level rewards offer limited token-level guidance. On-policy self-distillation addresses this limitation by conditioning a self-teacher on privileged information to provide dense token-level supervision. Ho …

نسخة أولية وصول مفتوح

CypherTurn: A Multi-Turn Benchmark for Conversational Text-to-Cypher Evaluation and the Autonomy Divergence

Yuzhe Zhang, Weijie Zhu, Haolin Yang وآخرون · 2026

Graph databases are increasingly queried through natural language, yet every existing benchmark evaluates isolated single-turn queries rather than the multi-turn sessions through which analysts actually work. We introduce CypherTurn, the first benchmark for conversational Text-to-Cypher evaluation, comprising 721 sessi …

المؤلفون المشاركون