Zhou Liu

LLMs struggle with semantic fidelity, achieving only a 30% alignment with scientific paper specifications, revealing a critical gap in AI-driven research reproducibility.

ReproAgent achieves unprecedented accuracy in translating research papers into executable code, outperforming existing methods by leveraging a dual-channel contract system.

TR-GS achieves superior volumetric rendering quality in sparse-view CT by leveraging t-distribution primitives and a novel ray-confidence model, significantly reducing artifacts and uncertainty.

Executable roles derived from team trajectories can boost multi-agent language model performance by over 16 points, reshaping agent interactions.

Visual Skill Cards transform raw interaction data into actionable insights, boosting GUI action prediction accuracy by over 11 points.

Multi-LLM reasoning gets a boost with TRACER, a framework that learns when to speak and what to say, outperforming fixed collaboration protocols without excessive training costs.

Counterintuitively, simply smoothing boundary probabilities in a localized manner can dramatically improve annotation-free skin lesion segmentation, rivaling supervised methods on some datasets.

Stop reinventing the wheel: OpenWorldLib offers a unified framework and codebase for advanced world models, finally bringing standardization to a fragmented field.

Stop wrestling with finicky evaluation codebases: One-Eval lets you specify LLM evaluation tasks in natural language and automatically executes them end-to-end.