New model TeCQR leverages conversations for better question retrieval

Researchers have developed a new model called TeCQR to improve the retrieval of related questions on community question-answering platforms. This model leverages conversational interactions, specifically by creating conversations through tag-enhanced clarifying questions. TeCQR incorporates a noise-tolerant mechanism to assess semantic similarity between questions and tags, allowing it to handle imperfect feedback effectively. The proposed tag-enhanced two-stage offline training method aims to capture detailed representations of user queries, questions, and tags, ultimately enhancing the accuracy of related question retrieval. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Improves information retrieval in community Q&A platforms by leveraging conversational context.

RANK_REASON Academic paper proposing a new model for question retrieval.

Read on arXiv cs.CL →

• Stack Overflow
• TeCQR
• arXiv

• paper
• other