ACL Findings2022
Mitigating Contradictions in Dialogue Based on Contrastive Learning
Abstract
Chatbot models have achieved remarkable progress in recent years but tend to yield contradictory responses. In this paper, we exploit the advantage of contrastive learning technique to mitigate this issue. To endow the model with the ability of discriminating contradictory patterns, we minimize the similarity between the target response and contradiction related negative example. The negative example is generated with learnable latent noise, which receives contradiction related feedback from the pretrained critic. Experimental results show that our method helps to avoid contradictions in response generation while preserving response fluency, outperforming existing methods on both automatic and human evaluation.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
3 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- ReIn: Conversational Error Recovery with Reasoning Inception cites Red Teaming Language Models for Processing Contradictory Dialogues
- Contrastive Learning for Inference in Dialogue cites SimCSE: Simple Contrastive Learning of Sentence Embeddings
- SimCSE: Simple Contrastive Learning of Sentence Embeddings cites A Simple Framework for Contrastive Learning of Visual Representations
Cited by top-tier papers3
- ReIn: Conversational Error Recovery with Reasoning InceptionTakyoung Kim, Jinseok Nam, Chandrayee Basu, Xing Fan et al.ICLR 2026 · 1 citation
- Red Teaming Language Models for Processing Contradictory DialoguesXiaofei Wen, Bangzheng Li, Tenghao Huang, Muhao ChenEMNLP 2024 · 1 citation
- Contrastive Learning for Inference in DialogueEtsuko Ishii, Yan Xu, Bryan Wilie, Ziwei Ji et al.EMNLP 2023 · 1 citation
Builds on13
- A Simple Framework for Contrastive Learning of Visual RepresentationsTing Chen, Simon Kornblith, Mohammad Norouzi, Geoffrey E. HintonICML 2020 · 24,064 citations
- SimCSE: Simple Contrastive Learning of Sentence EmbeddingsTianyu Gao, Xingcheng Yao, Danqi ChenEMNLP 2021 · 2,496 citations
- Neural Text Generation With Unlikelihood TrainingSean Welleck, Ilia Kulikov, Stephen Roller, Emily Dinan et al.ICLR 2020 · 683 citations
- Stabilizing Transformers for Reinforcement LearningEmilio Parisotto, H. Francis Song, Jack W. Rae, Razvan Pascanu et al.ICML 2020 · 464 citations
- PLATO: Pre-trained Dialogue Generation Model with Discrete Latent VariableSiqi Bao, Huang He, Fan Wang, Hua Wu et al.ACL 2020 · 229 citations
Related papers
- Group-wise Contrastive Learning for Neural Dialogue GenerationHengyi Cai, Hongshen Chen, Yonghao Song, Zhuoye Ding et al.EMNLP Findings 2020 · 30 citations
- Mitigating Negative Style Transfer in Hybrid Dialogue SystemShimin Li, Qinyuan Cheng, Linyang Li, Xipeng QiuAAAI 2023 · 1 citation
- Customizing Language Model Responses with Contrastive In-Context LearningXiang Gao, Kamalika DasAAAI 2024 · 23 citations
- Contrastive Learning Reduces Hallucination in ConversationsWeiwei Sun, Zhengliang Shi, Shen Gao, Pengjie Ren et al.AAAI 2023 · 92 citations
- A Large Collection of Model-generated Contradictory Responses for Consistency-aware Dialogue SystemsShiki Sato, Reina Akama, Jun Suzuki, Kentaro InuiACL Findings 2024 · 1 citation
