Quan Tu

SSP: Self-Supervised Post-training for Conversational Search

2023

Conversational search has been regarded as the next-generation search paradigm.Constrained by data scarcity, most existing methods distill the well-trained ad-hoc retriever to the conversational retriever.However, these methods, which usually initialize parameters by query reformulation to …

StreamingDialogue: Prolonged Dialogue Learning via Long Context Compression with Minimal Losses

2024 · arXiv (Cornell University)

Standard Large Language Models (LLMs) struggle with handling dialogues with long contexts due to efficiency and consistency issues. According to our observation, dialogue contexts are highly structured, and the special token of \textit{End-of-Utterance} (EoU) in …