# Response Selection for Multi-Party Conversations with Dynamic Topic Tracking

arXiv: https://arxiv.org/abs/2010.07785 · HTML: https://mlsanity.com/paper/2010.07785/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Oct 2020.

## Built on by (most cited first, 26 of 26)

- [MPC-BERT: A Pre-Trained Language Model for Multi-Party Conversation Understanding](https://mlsanity.com/paper/2106.01541/) (2021)
- [Improving Unsupervised Dialogue Topic Segmentation with Utterance-Pair Coherence Scoring](https://mlsanity.com/paper/2106.06719/) (2021)
- [HeterMPC: A Heterogeneous Graph Neural Network for Response Generation in Multi-Party Conversations](https://mlsanity.com/paper/2203.08500/) (2022)
- [Dialogue History Matters! Personalized Response Selectionin Multi-turn Retrieval-based Chatbots](https://mlsanity.com/paper/2103.09534/) (2021)
- [Self- and Pseudo-self-supervised Prediction of Speaker and Key-utterance for Multi-party Dialogue Reading Comprehension](https://mlsanity.com/paper/2109.03772/) (2021)
- [Enhanced Speaker-aware Multi-party Multi-turn Dialogue Comprehension](https://mlsanity.com/paper/2109.04066/) (2021)
- [Is ChatGPT a Good Multi-Party Conversation Solver?](https://mlsanity.com/paper/2310.16301/) (2023)
- [One Agent To Rule Them All: Towards Multi-agent Conversational AI](https://mlsanity.com/paper/2203.07665/) (2022)
- [GIFT: Graph-Induced Fine-Tuning for Multi-Party Conversation Understanding](https://mlsanity.com/paper/2305.09360/) (2023)
- [Dialogue-oriented Pre-training](https://mlsanity.com/paper/2106.00420/) (2021)
- [Do LLMs suffer from Multi-Party Hangover? A Diagnostic Approach to Addressee Recognition and Response Selection in Conversations](https://mlsanity.com/paper/2409.18602/) (2024)
- [Unsupervised Conversation Disentanglement through Co-Training](https://mlsanity.com/paper/2109.03199/) (2021)
- [Mixed-Session Conversation with Egocentric Memory](https://mlsanity.com/paper/2410.02503/) (2024)
- [Structural Characterization for Dialogue Disentanglement](https://mlsanity.com/paper/2110.08018/) (2021)
- [EM Pre-training for Multi-party Dialogue Response Generation](https://mlsanity.com/paper/2305.12412/) (2023)
- [Multi-Party Conversational Agents: A Survey](https://mlsanity.com/paper/2505.18845/) (2025)
- [Conversation Disentanglement with Bi-Level Contrastive Learning](https://mlsanity.com/paper/2210.15265/) (2022)
- [MADNet: Maximizing Addressee Deduction Expectation for Multi-Party Conversation Generation](https://mlsanity.com/paper/2305.12733/) (2023)
- [Multi-turn Dialogue Comprehension from a Topic-aware Perspective](https://mlsanity.com/paper/2309.09666/) (2023)
- [Multimodal Conversation Structure Understanding](https://mlsanity.com/paper/2505.17536/) (2025)
- [Zero-Shot Dialogue Disentanglement by Self-Supervised Entangled Response Selection](https://mlsanity.com/paper/2110.12646/) (2021)
- [Dramatic Conversation Disentanglement](https://mlsanity.com/paper/2305.16648/) (2023)
- [Advancing Multi-Party Dialogue Framework with Speaker-ware Contrastive Learning](https://mlsanity.com/paper/2501.11292/) (2025)
- [SS-MPC: A Sequence-Structured Multi-Party Conversation System](https://mlsanity.com/paper/2502.16920/) (2025)
- [Multi-party Response Generation with Relation Disentanglement](https://mlsanity.com/paper/2403.10827/) (2024)
- [CobSeg: Coherence Boundary Modeling for Dialogue Topic Segmentation](https://mlsanity.com/paper/2605.30668/) (2026)

## Abstract (arXiv)

While participants in a multi-party multi-turn conversation simultaneously engage in multiple conversation topics, existing response selection methods are developed mainly focusing on a two-party single-conversation scenario. Hence, the prolongation and transition of conversation topics are ignored by current methods. In this work, we frame response selection as a dynamic topic tracking task to match the topic between the response and relevant conversation context. With this new formulation, we propose a novel multi-task learning framework that supports efficient encoding through large pretrained models with only two utterances at once to perform dynamic topic disentanglement and response selection. We also propose Topic-BERT an essential pretraining step to embed topic information into BERT with self-supervised learning. Experimental results on the DSTC-8 Ubuntu IRC dataset show state-of-the-art results in response selection and topic disentanglement tasks outperforming existing methods by a good margin.