# An Effective Domain Adaptive Post-Training Method for BERT in Response Selection

arXiv: https://arxiv.org/abs/1908.04812 · HTML: https://mlsanity.com/paper/1908.04812/

## Key facts

- Cited by 7 papers in the map (data through 7 Oct 2026).
- First posted on arXiv 13 Aug 2019.

## Built on by (most cited first, 7 of 7)

- [NaturalConv: A Chinese Dialogue Dataset Towards Multi-turn Topic-driven Conversation](https://mlsanity.com/paper/2103.02548/) (2021)
- [On the Calibration and Uncertainty of Neural Learning to Rank Models](https://mlsanity.com/paper/2101.04356/) (2021)
- [Cluster & Tune: Boost Cold Start Performance in Text Classification](https://mlsanity.com/paper/2203.10581/) (2022)
- [Multi-turn Dialogue Reading Comprehension with Pivot Turns and Knowledge](https://mlsanity.com/paper/2102.05474/) (2021)
- [Contextual Fine-to-Coarse Distillation for Coarse-grained Response Selection in Open-Domain Conversations](https://mlsanity.com/paper/2109.13087/) (2021)
- [Improving Cross-Modal Understanding in Visual Dialog via Contrastive Learning](https://mlsanity.com/paper/2204.07302/) (2022)
- [Channel-aware Decoupling Network for Multi-turn Dialogue Comprehension](https://mlsanity.com/paper/2301.03953/) (2023)

## Abstract (arXiv)

We focus on multi-turn response selection in a retrieval-based dialog system. In this paper, we utilize the powerful pre-trained language model Bi-directional Encoder Representations from Transformer (BERT) for a multi-turn dialog system and propose a highly effective post-training method on domain-specific corpus. Although BERT is easily adopted to various NLP tasks and outperforms previous baselines of each task, it still has limitations if a task corpus is too focused on a certain domain. Post-training on domain-specific corpus (e.g., Ubuntu Corpus) helps the model to train contextualized representations and words that do not appear in general corpus (e.g., English Wikipedia). Experimental results show that our approach achieves new state-of-the-art on two response selection benchmarks (i.e., Ubuntu Corpus V1, Advising Corpus) performance improvement by 5.9% and 6% on R@1.