mlsanityData to 7 Oct 2026

Negative Training for Neural Dialogue Response Generation

arXiv 1903.02134 ↗First posted 6 Mar 2019cs.CL cs.LGcited by 24

Not on the map. Why?

Abstract (arXiv)

Although deep learning models have brought tremendous advancements to the field of open-domain dialogue response generation, recent research results have revealed that the trained models have undesirable generation behaviors, such as malicious responses and generic (boring) responses. In this work, we propose a framework named "Negative Training" to minimize such behaviors. Given a trained model, the framework will first find generated samples that exhibit the undesirable behavior, and then use them to feed negative training signals for fine-tuning the model. Our experiments show that negative training can significantly reduce the hit rate of malicious responses, or discourage frequent responses and improve response diversity.

Key facts. Cited by 24 papers in the map (data through 7 Oct 2026). First posted on arXiv 6 Mar 2019.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 7; 2022: 11; 2023: 3; 2024: 1; 2025: 2.

24 papers.

Sort:

By area:

  1. 2022Red Teaming Language Models with Language Modelscited by 983
  2. 2021Recent Advances in Deep Learning Based Dialogue Systems: A Systematic Surveycited by 130
  3. 2021CLIFF: Contrastive Learning for Improving Faithfulness and Factuality in Abstractive Summarizationcited by 120
  4. 2022The CRINGE Loss: Learning what language not to modelcited by 39
  5. 2022Rethinking and Refining the Distinct Metriccited by 27
  6. 2022ConTinTin: Continual Learning from Task Instructionscited by 26
  7. 2021Diversifying Dialog Generation via Adaptive Label Smoothingcited by 25
  8. 2023Evolving Diverse Red-team Language Models in Multi-round Multi-agent Gamescited by 17
  9. 2022Measuring and Improving Semantic Diversity of Dialogue Generationcited by 16
  10. 2021Identifying Untrustworthy Samples: Data Filtering for Open-domain Dialogues with Bayesian Optimizationcited by 15
  11. 2022Diversifying Neural Dialogue Generation via Negative Distillationcited by 6
  12. 2021Automatically Exposing Problems with Neural Dialog Modelscited by 5
12 more
  1. 2022Unified Multimodal Model with Unlikelihood Training for Visual Dialogcited by 5
  2. 2022Controlling the Focus of Pretrained Language Generation Modelscited by 4
  3. 2022MOCHA: A Multi-Task Training Approach for Coherent Text Generation from Cognitive Perspectivecited by 4
  4. 2023Learning Descriptive Image Captioning via Semipermeable Maximum Likelihood Estimationcited by 4
  5. 2025IdealTSF: Can Non-Ideal Data Contribute to Enhancing the Performance of Time Series Forecasting Models?cited by 3
  6. 2021Diversifying Neural Text Generation with Part-of-Speech Guided Softmax and Samplingcited by 2
  7. 2021Similarity Based Label Smoothing For Dialogue Generationcited by 2
  8. 2022Near-Negative Distinction: Giving a Second Life to Human Evaluation Datasetscited by 1
  9. 2022Modeling Complex Dialogue Mappings via Sentence Semantic Segmentation Guided Conditional Variational Auto-Encodercited by 1
  10. 2024Towards Minimal Targeted Updates of Language Models with Targeted Negative Trainingcited by 1
  11. 2023Learning to Diversify Neural Text Generation via Degenerative Modelcited by 0
  12. 2025Reducing the Probability of Undesirable Outputs in Language Models Using Probabilistic Inferencecited by 0