mlsanityData to 7 Oct 2026

A Diversity-Promoting Objective Function for Neural Conversation Models

arXiv 1510.03055 ↗First posted 11 Oct 2015cs.CLcited by 1,068

Not on the map. Why?

Cited by 55 in LLM alignment / RLHF / preference optimisation, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Sequence-to-sequence neural network models for generation of conversational responses tend to generate safe, commonplace responses (e.g., "I don't know") regardless of the input. We suggest that the traditional objective function, i.e., the likelihood of output (response) given input (message) is unsuited to response generation tasks. Instead we propose using Maximum Mutual Information (MMI) as the objective function in neural models. Experimental results demonstrate that the proposed MMI models produce more diverse, interesting, and appropriate responses, yielding substantive gains in BLEU scores on two conversational datasets and in human evaluations.

Key facts. Cited by 1,068 papers in the map (data through 7 Oct 2026). First posted on arXiv 11 Oct 2015.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2018: 1; 2019: 0; 2020: 2; 2021: 184; 2022: 221; 2023: 192; 2024: 172; 2025: 152; 2026: 144.

1,068 papers.

Sort:

By area:

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  2. 2021Calibrate Before Use: Improving Few-Shot Performance of Language Modelscited by 1,371
  3. 2022LaMDA: Language Models for Dialog Applicationscited by 1,218
  4. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,054
  5. 2023RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignmentcited by 672
  6. 2022Contrastive Decoding: Open-ended Text Generation as Optimizationcited by 598
  7. 2021DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Expertscited by 434
  8. 2023Is ChatGPT a Good NLG Evaluator? A Preliminary Studycited by 420
  9. 2021Large Language Models Can Be Strong Differentially Private Learnerscited by 405
  10. 2023Understanding the Effects of RLHF on LLM Generalisation and Diversitycited by 386
  11. 2021FUDGE: Controlled Text Generation With Future Discriminatorscited by 374
  12. 2023Trusting Your Evidence: Hallucinate Less with Context-aware Decodingcited by 319
38 more
  1. 2022The Vendi Score: A Diversity Evaluation Metric for Machine Learningcited by 299
  2. 2021Surface Form Competition: Why the Highest Probability Answer Isn't Always Rightcited by 256
  3. 2022CoAuthor: Designing a Human-AI Collaborative Writing Dataset for Exploring Language Model Capabilitiescited by 245
  4. 2023Pretraining Language Models with Human Preferencescited by 245
  5. 2022Is Reinforcement Learning (Not) for Natural Language Processing: Benchmarks, Baselines, and Building Blocks for Natural Language Policy Optimizationcited by 244
  6. 2021The GEM Benchmark: Natural Language Generation, its Evaluation and Metricscited by 239
  7. 2022A Contrastive Framework for Neural Text Generationcited by 224
  8. 2022Factuality Enhanced Language Models for Open-Ended Text Generationcited by 200
  9. 2024Multi-Modal Hallucination Control by Visual Information Groundingcited by 196
  10. 2022SSD-LM: Semi-autoregressive Simplex-based Diffusion Language Model for Text Generation and Modular Controlcited by 191
  11. 2023Beyond Reverse KL: Generalizing Direct Preference Optimization with Diverse Divergence Constraintscited by 174
  12. 2024A Survey on Recent Advances in LLM-Based Multi-turn Dialogue Systemscited by 170
  13. 2021Challenges in Detoxifying Language Modelscited by 169
  14. 2024COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllabilitycited by 162
  15. 2022A Survey on Retrieval-Augmented Text Generationcited by 157
  16. 2021Recent Advances in Deep Learning Based Dialogue Systems: A Systematic Surveycited by 130
  17. 2023MiniLLM: On-Policy Distillation of Large Language Modelscited by 129
  18. 2022A Survey of Controllable Text Generation using Transformer-based Pre-trained Language Modelscited by 128
  19. 2022Generative Spoken Dialogue Language Modelingcited by 125
  20. 2023Black-Box Prompt Optimization: Aligning Large Language Models without Model Trainingcited by 122
  21. 2022Pretrained Language Models for Text Generation: A Surveycited by 121
  22. 2021Advances and Challenges in Conversational Recommender Systems: A Surveycited by 118
  23. 2023Aligning Language Models with Preferences through f-divergence Minimizationcited by 118
  24. 2023f-Divergence Minimization for Sequence-Level Knowledge Distillationcited by 112
  25. 2022Locally Typical Samplingcited by 108
  26. 2022Why Exposure Bias Matters: An Imitation Learning Perspective of Error Accumulation in Language Generationcited by 106
  27. 2023Lift Yourself Up: Retrieval-augmented Text Generation with Self Memorycited by 103
  28. 2023The Curious Decline of Linguistic Diversity: Training Language Models on Synthetic Textcited by 97
  29. 2021PsyQA: A Chinese Dataset for Generating Long Counseling Text for Mental Health Supportcited by 92
  30. 2021A Comprehensive Assessment of Dialog Evaluation Metricscited by 90
  31. 2021CEM: Commonsense-aware Empathetic Response Generationcited by 89
  32. 2022On Decoding Strategies for Neural Text Generatorscited by 89
  33. 2022MISC: A MIxed Strategy-Aware Model Integrating COMET for Emotional Support Conversationcited by 84
  34. 2022Coder Reviewer Reranking for Code Generationcited by 83
  35. 2022On Reinforcement Learning and Distribution Matching for Fine-Tuning Language Models with no Catastrophic Forgettingcited by 80
  36. 2022AugESC: Dialogue Augmentation with Large Language Models for Emotional Support Conversationcited by 79
  37. 2023Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Modelcited by 79
  38. 2025Jointly Reinforcing Diversity and Quality in Language Model Generationscited by 79

50 most cited of 1,068 All 1,068 (JSON)