---
title: "WeChat Researchers Built F2RVLM to Find Buried Chat Fragments Across Thousands of Turns"
subtitle: "A new retrieval framework tackles the messy reality of long group chats by finding coherent multi-message, multi-image fragments instead of isolated utterances"
url: "https://alphasignal.ai/news/wechat-researchers-built-f2rvlm-to-find-buried-chat-fragments-across-thousands"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "AlphaSignal"
company_url: "https://alphasignal.ai/company/alphasignal"
published: "2026-06-03T00:00:00.000Z"
updated: "2026-06-12T00:58:07.761Z"
type: "Paper"
topics:
  - "Retrieval"
  - "Image"
subtopics:
  - "Rag"
  - "Embeddings"
  - "Vision Language"
upvotes: 1388
source_url: "https://arxiv.org/abs/2606.04591"
comments: 0
truncated: true
---

# WeChat Researchers Built F2RVLM to Find Buried Chat Fragments Across Thousands of Turns

> A new retrieval framework tackles the messy reality of long group chats by finding coherent multi-message, multi-image fragments instead of isolated utterances

Anyone who has tried to find a specific conversation thread in a long group chat knows the pain: you remember the gist of what was said, but the relevant messages are scattered across dozens of turns, mixed with images, and buried under unrelated tangents. Standard search tools return individual messages in isolation, stripping away the context that made them meaningful in the first place. A new paper from researchers at WeChat's parent organization tackles this head-on with a framework called [Fine-grained Fragment Retrieval (FFR)](https://arxiv.org/abs/2606.04591) \-- a system designed to find coherent, multi-message, multi-image conversation fragments from long-form dialogues.

## The gap that existing retrieval misses

Traditional dialogue retrieval aims to select the most appropriate utterance or image from recent dialogue history, but these approaches often fail to meet users' actual needs for revisiting semantically coherent content scattered across long-form conversations. Think of it like the difference between a search engine returning a single sentence versus returning the whole relevant paragraph. Even leading Vision-Language Models (VLMs) struggle here -- they often retrieve incoherent utterance-image fragments because, while optimized for generating responses from visual-textual inputs, they lack explicit supervision to ensure semantic coherence within retrieved fragments.

The problem is compounded by scale. Real-world WeChat conversations reflect authentic user interactions with noisy inputs, informal expressions, and frequent topic shifts, posing greater challenges for robust retrieval than any clean benchmark can simulate. Retrieving a fragment that spans, say, a planning discussion with shared photos requires understanding topical boundaries, not just keyword overlap.

![Comparison of traditional utterance retrieval vs fine-grained fragment retrieval approaches](https://arxiv.org/html/2606.04591v1/Retrieval_all.jpg) 

## Two problems, two systems

The paper defines FFR and then splits it into two distinct settings, each with its own solution:

* **FFR within a Single Dialogue:** Given one long conversation, find the fragment most relevant to a query. The model here is **F2RVLM**, a generation-based retrieval model that reasons over the full dialogue and outputs the relevant span.
* **FFR within a Dialogue Corpus:** Given a large collection of dialogues (think: an entire organization's chat history), find the right fragment across all of them. This is handled by

_Preview only. Full text requires an AlphaSignal Pro account._

## Sources

- [Fine-grained Fragment Retrieval in Multi-modal Long-form Dialogues (arXiv 2606.04591)](https://arxiv.org/abs/2606.04591)
- [F2RVLM: Boosting Fine-grained Fragment Retrieval for Multi-Modal Long-form Dialogue with Vision Language Model (arXiv 2508.17714)](https://arxiv.org/abs/2508.17714)
- [F2RVLM HTML paper (arXiv 2508.17714v1)](https://arxiv.org/html/2508.17714v1)
- [Fine-grained Fragment Retrieval HTML paper (arXiv 2606.04591)](https://arxiv.org/html/2606.04591)
