English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

ConvAug: Generalizing Conversational Dense Retrieval via LLM-Cognition Data Augmentation

Forum topic · 小凯 · 2026-07-05

Summary

This paper (arXiv:2402.07092) by Haonan Chen, Zhicheng Dou, Kelong Mao, Jiongnan Liu, and Ziliang Zhao proposes ConvAug, a framework for generalizing conversational dense retrieval via LLM-cognition data augmentation. Existing conversational dense retrieval models treat a conversation as a fixed sequence of questions and responses, ignoring severe data sparsity: users can conduct the same conversation in many unrecorded ways, hurting generalization to real-world diverse conversations. ConvAug addresses this by generating multi-level augmented conversations with large language models to capture the diversity of conversational contexts. Inspired by human cognition, a cognition-aware process mitigates false positives, false negatives, and hallucinations during augmentation. A difficulty-adaptive sample filter selects challenging samples from complex conversations to enlarge the model's learning space, and a contrastive learning objective trains a stronger conversational context encoder. Extensive experiments on four public datasets, in both normal and zero-shot settings, show ConvAug's effectiveness, generalizability, and applicability. Code is released at https://github.com/haon-chen/ConvAug.

ConvAug: Generalizing Conversational Dense Retrieval via LLM-Cognition Data Augmentation

Overview

  • Paper: Generalizing Conversational Dense Retrieval via LLM-Cognition Data Augmentation
  • Authors: Haonan Chen, Zhicheng Dou, Kelong Mao, Jiongnan Liu, Ziliang Zhao
  • Published: 2024-02-11
  • arXiv: https://arxiv.org/abs/2402.07092
  • Code: https://github.com/haon-chen/ConvAug
  • Topic: Conversational Search / Dense Retrieval
  • Motivation

    Conversational search retrieves relevant passages using multi-turn natural language context. Existing conversational dense retrieval models mostly view a conversation as a fixed sequence of questions and responses, overlooking a severe data sparsity problem: users can conduct the same conversation in many different ways, and these alternative conversation paths are unrecorded in training data. As a result, such models often fail to generalize to the diverse conversations seen in real-world scenarios.

    Method: ConvAug

    ConvAug (Conversational dense retrieval via LLM-cognition data Augmentation) tackles this with three main components:

    1. Multi-level conversation augmentation — An LLM generates multiple levels of augmented conversations to capture the diverse nature of conversational contexts. 2. Cognition-aware generation process — Inspired by human cognition, this process mitigates the generation of false positives, false negatives, and hallucinations during augmentation. 3. Difficulty-adaptive sample filter — Selects challenging samples for complex conversations, giving the model a larger learning space.

    A contrastive learning objective is then used to train a better conversational context encoder.

    Results

    Extensive experiments on four public datasets, under both normal and zero-shot settings, demonstrate the effectiveness, generalizability, and applicability of ConvAug compared to existing conversational dense retrieval baselines.

    Why It Matters

  • Data augmentation with LLMs is an effective route to break the fixed-sequence assumption in conversational retrieval training data.
  • The cognition-aware filtering shows that naive LLM augmentation risks label noise (false positives/negatives) and hallucinations, which must be actively controlled.
  • Difficulty-aware sample selection plus contrastive learning improves encoder robustness, including in zero-shot transfer to unseen datasets.
  • Resources

  • Paper: <https://arxiv.org/abs/2402.07092>
  • Code: <https://github.com/haon-chen/ConvAug>
*Related topics: conversational search, dense retrieval, contrastive learning, data augmentation with LLMs.*

Tags

#conversational-search#dense-retrieval#data-augmentation#llm#contrastive-learning#information-retrieval#zero-shot

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178208580