3 papers
cs.CL2024
Aligning Language Models Using Follow-up Likelihood as Reward Signal
Chen Zhang, Dading Chong, Feng Jiang +4
In natural human-to-human conversations, participants often receive feedback signals from one another based on their follow-up reactions. These reactions can include verbal respons…
cs.CL2024
TS-Align: A Teacher-Student Collaborative Framework for Scalable Iterative Finetuning of Large Language Models
Chen Zhang, Chengguang Tang, Dading Chong +4
Mainstream approaches to aligning large language models (LLMs) heavily rely on human preference data, particularly when models require periodic updates. The standard process for it…
cs.CL2023
xDial-Eval: A Multilingual Open-Domain Dialogue Evaluation Benchmark
Chen Zhang, Luis Fernando D'Haro, Chengguang Tang +3
Recent advancements in reference-free learned metrics for open-domain dialogue evaluation have been driven by the progress in pre-trained language models and the availability of di…