mlsanityData to 7 Oct 2026

PREADD: Prefix-Adaptive Decoding for Controlled Text Generation

arXiv 2307.03214 ↗First posted 6 Jul 2023cs.CLcited by 29

Not on the map. Why?

Abstract (arXiv)

We propose Prefix-Adaptive Decoding (PREADD), a flexible method for controlled text generation. Unlike existing methods that use auxiliary expert models to control for attributes, PREADD does not require an external model, instead relying on linearly combining output logits from multiple prompts. Specifically, PREADD contrasts the output logits generated using a raw prompt against those generated using a prefix-prepended prompt, enabling both positive and negative control with respect to any attribute encapsulated by the prefix. We evaluate PREADD on three tasks -- toxic output mitigation, gender bias reduction, and sentiment control -- and find that PREADD outperforms not only prompting baselines, but also an auxiliary-expert control method, by 12% or more in relative gain on our main metrics for each task.

Key facts. Cited by 29 papers in the map (data through 7 Oct 2026). Builds on 27 of its 51 references that are in the map. First posted on arXiv 6 Jul 2023.

Builds on

27 of 51 references.

References · 27

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  5. 2022Training language models to follow instructions with human feedbackcited by 15,712
  6. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
  7. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  8. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
  9. 2020The Pile: An 800GB Dataset of Diverse Text for Language Modelingcited by 2,346
  10. 2019HuggingFace's Transformers: State-of-the-art Natural Language Processingcited by 2,201
  11. 2021Calibrate Before Use: Improving Few-Shot Performance of Language Modelscited by 1,371
  12. 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,235
15 more
  1. 2020Language (Technology) is Power: A Critical Survey of "Bias" in NLPcited by 951
  2. 2019Plug and Play Language Models: A Simple Approach to Controlled Text Generationcited by 793
  3. 2018Gender Bias in Coreference Resolution: Evaluation and Debiasing Methodscited by 657
  4. 2022Contrastive Decoding: Open-ended Text Generation as Optimizationcited by 598
  5. 2019The Woman Worked as a Babysitter: On Biases in Language Generationcited by 540
  6. 2020GeDi: Generative Discriminator Guided Sequence Generationcited by 377
  7. 2021FUDGE: Controlled Text Generation With Future Discriminatorscited by 374
  8. 2021Self-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLPcited by 336
  9. 2019Measurement and Fairnesscited by 299
  10. 2021Latent Hatred: A Benchmark for Understanding Implicit Hate Speechcited by 216
  11. 2021Detoxifying Language Models Risks Marginalizing Minority Voicescited by 106
  12. 2020Hate Speech Detection and Racial Bias Mitigation in Social Media based on BERT modelcited by 87
  13. 2020The Radicalization Risks of GPT-3 and Advanced Neural Language Modelscited by 76
  14. 2021Controllable Generation from Pre-trained Language Models via Inverse Promptingcited by 29
  15. 2022Why is constrained neural language generation particularly challenging?cited by 19

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 14; 2025: 11; 2026: 2.

29 papers.

Sort:

By area:

  1. 2023Steering Language Models With Activation Engineeringcited by 824
  2. 2024Tuning Language Models by Proxycited by 88
  3. 2023Controlled Text Generation via Language Model Arithmeticcited by 51
  4. 2025Parallel Scaling Law for Language Modelscited by 29
  5. 2025When One LLM Drools, Multi-LLM Collaboration Rulescited by 25
  6. 2024SafeInfer: Context Adaptive Decoding Time Safety Alignment for Large Language Modelscited by 22
  7. 2024Automatic Metrics in Natural Language Generation: A Survey of Current Evaluation Practicescited by 19
  8. 2024Controlled Text Generation for Large Language Model with Dynamic Attribute Graphscited by 16
  9. 2024Selective Prompt Anchoring for Code Generationcited by 16
  10. 2025G2: Guided Generation for Enhanced Output Diversity in LLMscited by 14
  11. 2024RSA-Control: A Pragmatics-Grounded Lightweight Controllable Text Generation Frameworkcited by 11
  12. 2025Measuring and Guiding Monosemanticitycited by 10
17 more
  1. 2024DESTEIN: Navigating Detoxification of Language Models via Universal Steering Pairs and Head-wise Activation Fusioncited by 9
  2. 2025Don't Throw Away Your Pretrained Modelcited by 9
  3. 2025Speculative Safety-Aware Decodingcited by 7
  4. 2024SCAR: Sparse Conditioned Autoencoders for Concept Detection and Steering in LLMscited by 6
  5. 2025JAM: Controllable and Responsible Text Generation via Causal Reasoning and Latent Vector Manipulationcited by 5
  6. 2024Questionable practices in machine learningcited by 3
  7. 2024VTechAGP: An Academic-to-General-Audience Text Paraphrase Dataset and Benchmark Modelscited by 3
  8. 2024LaSEr-Edit: Localized Span-level Error Editing with Energy-based Localizationcited by 2
  9. 2025Palette of Language Models: A Solver for Controlled Text Generationcited by 2
  10. 2024CPT: Consistent Proxy Tuning for Black-box Optimizationcited by 1
  11. 2025LaPA2: Length-Aware Prefix and Prompt Attention Augmentation for Long-Form Controllable Text Generationcited by 1
  12. 2026The Single-Multi Evolution Loop for Self-Improving Model Collaboration Systemscited by 1
  13. 2024LingGen: Scalable Multi-Attribute Linguistic Control via Power-Law Maskingcited by 0
  14. 2024Interweaving Memories of a Siamese Large Language Modelcited by 0
  15. 2025CLEAR: Contrasting Textual Feedback with Experts and Amateurs for Reasoningcited by 0
  16. 2025AgentCTG: Harnessing Multi-Agent Collaboration for Fine-Grained Precise Control in Text Generationcited by 0
  17. 2026Guidance Is Not a Hyperparameter: Learning Dynamic Control in Diffusion Language Modelscited by 0