mlsanityData to 14 Oct 2026

Towards Controllable Biases in Language Generation

arXiv 2005.00268 ↗First posted 1 May 2020cs.CLcited by 112

Not on the map. Why?

Abstract (arXiv)

We present a general approach towards controllable societal biases in natural language generation (NLG). Building upon the idea of adversarial triggers, we develop a method to induce societal biases in generated text when input prompts contain mentions of specific demographic groups. We then analyze two scenarios: 1) inducing negative biases for one demographic and positive biases for another demographic, and 2) equalizing biases between demographics. The former scenario enables us to detect the types of biases present in the model. Specifically, we show the effectiveness of our approach at facilitating bias analysis by finding topics that correspond to demographic inequalities in generated text and comparing the relative effectiveness of inducing biases for different demographics. The second scenario is useful for mitigating biases in downstream applications such as dialogue generation. In our experiments, the mitigation technique proves to be effective at equalizing the amount of biases across demographics while simultaneously generating less negatively biased text overall.

Key facts. Cited by 112 papers in the map (data through 14 Oct 2026). First posted on arXiv 1 May 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 23; 2022: 32; 2023: 26; 2024: 14; 2025: 10; 2026: 7.

112 papers.

Sort:

By area:

  1. 2021Deduplicating Training Data Makes Language Models Bettercited by 714
  2. 2023Bias and Fairness in Large Language Models: A Surveycited by 639
  3. 2021DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Expertscited by 441
  4. 2021BOLD: Dataset and Metrics for Measuring Biases in Open-Ended Language Generationcited by 412
  5. 2021Towards Understanding and Mitigating Social Biases in Language Modelscited by 296
  6. 2022Quark: Controllable Text Generation with Reinforced Unlearningcited by 237
  7. 2022LinkBERT: Pretraining Language Models with Document Linkscited by 211
  8. 2023REFINER: Reasoning Feedback on Intermediate Representationscited by 205
  9. 2021Societal Biases in Language Generation: Progress and Challengescited by 202
  10. 2023"Kelly is a Warm Person, Joseph is a Role Model": Gender Biases in LLM-Generated Reference Letterscited by 199
  11. 2022GreaseLM: Graph REASoning Enhanced Language Models for Question Answeringcited by 171
  12. 2022Deep Bidirectional Language-Knowledge Graph Pretrainingcited by 156
38 more
  1. 2022A Survey of Controllable Text Generation using Transformer-based Pre-trained Language Modelscited by 129
  2. 2024On Prompt-Driven Safeguarding for Large Language Modelscited by 116
  3. 2021Detoxifying Language Models Risks Marginalizing Minority Voicescited by 107
  4. 2022How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?cited by 92
  5. 2021A Survey on Gender Bias in Natural Language Processingcited by 89
  6. 2021Quantifying Social Biases in NLP: A Generalization and Empirical Comparison of Extrinsic Fairness Metricscited by 83
  7. 2021Mitigating Political Bias in Language Models Through Reinforced Calibrationcited by 73
  8. 2022Theories of "Gender" in NLP Bias Researchcited by 62
  9. 2024Self-Debiasing Large Language Models: Zero-Shot Recognition and Reduction of Stereotypescited by 59
  10. 2023GraphCare: Enhancing Healthcare Predictions with Personalized Knowledge Graphscited by 58
  11. 2021Just Say No: Analyzing the Stance of Neural Dialogue Generation in Offensive Contextscited by 56
  12. 2023"I'm fully who I am": Towards Centering Transgender and Non-Binary Voices to Measure Biases in Open Language Generationcited by 56
  13. 2024Disclosure and Mitigation of Gender Bias in LLMscited by 49
  14. 2021Ethical-Advice Taker: Do Language Models Understand Natural Language Interventions?cited by 46
  15. 2021Generating Syntactically Controlled Paraphrases without Using Annotated Parallel Pairscited by 45
  16. 2022A Causal Lens for Controllable Text Generationcited by 45
  17. 2021Towards Knowledge-Grounded Counter Narrative Generation for Hate Speechcited by 42
  18. 2022Controllable Text Generation with Neurally-Decomposed Oraclecited by 41
  19. 2021Fairness for Unobserved Characteristics: Insights from Technological Impacts on Queer Communitiescited by 40
  20. 2022Multilingual Generative Language Models for Zero-Shot Cross-Lingual Event Argument Extractioncited by 40
  21. 2022Mitigating Gender Bias in Distilled Language Models via Counterfactual Role Reversalcited by 40
  22. 2021Efficient (Soft) Q-Learning for Text Generation with Limited Good Datacited by 37
  23. 2023Are Personalized Stochastic Parrots More Dangerous? Evaluating Persona Biases in Dialogue Systemscited by 37
  24. 2023Evaluation of African American Language Bias in Natural Language Generationcited by 36
  25. 2021StylePTB: A Compositional Benchmark for Fine-grained Controllable Text Style Transfercited by 34
  26. 2022Why So Toxic? Measuring and Triggering Toxic Behavior in Open-Domain Chatbotscited by 33
  27. 2022Unified Detoxifying and Debiasing in Language Generation via Inference-time Adaptive Optimizationcited by 33
  28. 2021Uncovering Implicit Gender Bias in Narratives through Commonsense Inferencecited by 32
  29. 2023Probing Explicit and Implicit Gender Bias through LLM Conditional Text Generationcited by 31
  30. 2021Lawyers are Dishonest? Quantifying Representational Harms in Commonsense Knowledge Resourcescited by 30
  31. 2023Sociodemographic Bias in Language Models: A Survey and Forward Pathcited by 28
  32. 2024Identifying Implicit Social Biases in Vision-Language Modelscited by 27
  33. 2023Survey of Social Bias in Vision-Language Modelscited by 26
  34. 2023Denevil: Towards Deciphering and Navigating the Ethical Values of Large Language Models via Instruction Learningcited by 26
  35. 2022ClidSum: A Benchmark Dataset for Cross-Lingual Dialogue Summarizationcited by 25
  36. 2023Uncovering and Quantifying Social Biases in Code Generationcited by 25
  37. 2021Detect and Perturb: Neutral Rewriting of Biased and Sensitive Text via Gradient-based Decodingcited by 23
  38. 2022Robust Conversational Agents against Imperceptible Toxicity Triggerscited by 23

50 most cited of 112 All 112 (JSON)