6 papers · 1 filter
First, Do No Harm: AI Supervisor Scaffolds Novice Growth in Counselor Education
Chen Xu, Zhenyu Lyu, Tian Lan +12
The most dangerous mistakes a novice counselor makes are not the obvious ones: they are utterances that sound caring while quietly violating professional ethics and leaving vulnera…
A Survey of Automatic Evaluation Methods on Text, Visual and Speech Generations
Tian Lan, Yang-Hao Zhou, Zi-Ao Ma +8
Recent advances in deep learning have significantly enhanced generative AI capabilities across text, images, and audio. However, automatically evaluating the quality of these gener…
DECIDER: A Dual-System Rule-Controllable Decoding Framework for Language Generation
Chen Xu, Tian Lan, Yu Ji +8
Constrained decoding approaches aim to control the meaning or style of text generated by the pre-trained large language models (LLMs or also PLMs) for various tasks at inference ti…
Beyond Exact Match: Semantically Reassessing Event Extraction by Large Language Models
Yi-Fan Lu, Xian-Ling Mao, Tian Lan +3
Event extraction has gained extensive research attention due to its broad range of applications. However, the current mainstream evaluation method for event extraction relies on to…
Training Language Models to Critique With Multi-agent Feedback
Tian Lan, Wenwei Zhang, Chengqi Lyu +6
Critique ability, a meta-cognitive capability of humans, presents significant challenges for LLMs to improve. Recent works primarily rely on supervised fine-tuning (SFT) using crit…
CriticEval: Evaluating Large Language Model as Critic
Tian Lan, Wenwei Zhang, Chen Xu +4
Critique ability, i.e., the capability of Large Language Models (LLMs) to identify and rectify flaws in responses, is crucial for their applications in self-improvement and scalabl…