# Findings on Conversation Disentanglement

arXiv: https://arxiv.org/abs/2112.05346 · HTML: https://mlsanity.com/paper/2112.05346/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 19 references that are in the map.
- First posted on arXiv 10 Dec 2021.

## Builds on (most cited first, 9 of 9)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://mlsanity.com/paper/1909.11942/) (2019)
- [The Ubuntu Dialogue Corpus: A Large Dataset for Research in Unstructured Multi-Turn Dialogue Systems](https://mlsanity.com/paper/1506.08909/) (2015)
- [Poly-encoders: Transformer Architectures and Pre-training Strategies for Fast and Accurate Multi-sentence Scoring](https://mlsanity.com/paper/1905.01969/) (2019)
- [Molweni: A Challenge Multiparty Dialogues-based Machine Reading Comprehension Dataset with Discourse Structure](https://mlsanity.com/paper/2004.05080/) (2020)
- [Multi-turn Response Selection using Dialogue Dependency Relations](https://mlsanity.com/paper/2010.01502/) (2020)
- [Who did They Respond to? Conversation Structure Modeling using Masked Hierarchical Transformer](https://mlsanity.com/paper/1911.10666/) (2019)
- [Utterance-level Dialogue Understanding: An Empirical Study](https://mlsanity.com/paper/2009.13902/) (2020)

## Built on by (most cited first, 6 of 6)

- [Conversation Disentanglement with Bi-Level Contrastive Learning](https://mlsanity.com/paper/2210.15265/) (2022)
- [Multimodal Conversation Structure Understanding](https://mlsanity.com/paper/2505.17536/) (2025)
- [Dramatic Conversation Disentanglement](https://mlsanity.com/paper/2305.16648/) (2023)
- [Revisiting Conversation Discourse for Dialogue Disentanglement](https://mlsanity.com/paper/2306.03975/) (2023)
- [Computational Analysis of Conversation Dynamics through Participant Responsivity](https://mlsanity.com/paper/2509.16464/) (2025)
- [Language Models as Measurement Apparatus for Culture](https://mlsanity.com/paper/2607.02459/) (2026)

## Abstract (arXiv)

Conversation disentanglement, the task to identify separate threads in conversations, is an important pre-processing step in multi-party conversational NLP applications such as conversational question answering and conversation summarization. Framing it as a utterance-to-utterance classification problem -- i.e. given an utterance of interest (UOI), find which past utterance it replies to -- we explore a number of transformer-based models and found that BERT in combination with handcrafted features remains a strong baseline. We then build a multi-task learning model that jointly learns utterance-to-utterance and utterance-to-thread classification. Observing that the ground truth label (past utterance) is in the top candidates when our model makes an error, we experiment with using bipartite graphs as a post-processing step to learn how to best match a set of UOIs to past utterances. Experiments on the Ubuntu IRC dataset show that this approach has the potential to outperform the conventional greedy approach of simply selecting the highest probability candidate for each UOI independently, indicating a promising future research direction.