الباحثون

Jiakai Wang

المنشورات 2

نسخة أولية وصول مفتوح

When Upstream Messages Override Correct Answers: A Controlled Study of Multi-Agent LLM Collaboration

Yaxin Gong, Gangyi Zhang, Chongming Gao وآخرون · 2026

Multi-agent LLM systems rely on message passing among specialized agents to accomplish complex tasks. However, an upstream agent may provide useful information or an incorrect answer that causes a downstream agent to override a correct answer supported by its own evidence. Prior work has not clearly separated the benef …

نسخة أولية وصول مفتوح

Beyond Verbalized Confidence: Calibrating Reasoners with Differentiable Readouts

Chenxiao Fan, Chongming Gao, Gangyi Zhang وآخرون · 2026

Reinforcement learning with verifiable rewards (RLVR) trains reasoning models to produce correct answers, but does not ensure that their stated confidence is calibrated. The resulting models are systematically overconfident. Recent methods train calibration inside the RLVR loop by having the model state a numerical con …

المؤلفون المشاركون