Yin Cai

The paper introduces PARL, a framework that learns personalized evaluation rubrics directly from raw user interaction histories to accurately assess how well LLM outputs align with subjective, user-specific preferences.

This paper investigates the scaling behavior of homogeneous LLM-driven Multi-Agent Systems (MAS) and finds that performance exhibits diminishing returns due to coordination overhead, rather than scaling monotonically with agent count.

The paper proposes Deep Research as Rubric (DR-rubric), a novel evidence-driven framework that treats rubric construction itself as a research problem to generate fine-grained, scalable reward signals for open-ended reasoning tasks.