# DisCup: Discriminator Cooperative Unlikelihood Prompt-tuning for Controllable Text Generation

arXiv: https://arxiv.org/abs/2210.09551 · HTML: https://mlsanity.com/paper/2210.09551/

## Key facts

- Cited by 30 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 26 references that are in the map.
- First posted on arXiv 18 Oct 2022.

## Builds on (most cited first, 22 of 22)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [Don't Stop Pretraining: Adapt Language Models to Domains and Tasks](https://mlsanity.com/paper/2004.10964/) (2020)
- [RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models](https://mlsanity.com/paper/2009.11462/) (2020)
- [Plug and Play Language Models: A Simple Approach to Controlled Text Generation](https://mlsanity.com/paper/1912.02164/) (2019)
- [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://mlsanity.com/paper/1909.05858/) (2019)
- [Neural Text Generation with Unlikelihood Training](https://mlsanity.com/paper/1908.04319/) (2019)
- [DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts](https://mlsanity.com/paper/2105.03023/) (2021)
- [GeDi: Generative Discriminator Guided Sequence Generation](https://mlsanity.com/paper/2009.06367/) (2020)
- [FUDGE: Controlled Text Generation With Future Discriminators](https://mlsanity.com/paper/2104.05218/) (2021)
- [A Survey of Controllable Text Generation using Transformer-based Pre-trained Language Models](https://mlsanity.com/paper/2201.05337/) (2022)
- [A Distributional Approach to Controlled Text Generation](https://mlsanity.com/paper/2012.11635/) (2020)
- [Controllable Natural Language Generation with Contrastive Prefixes](https://mlsanity.com/paper/2202.13257/) (2022)
- [Controlled Text Generation as Continuous Optimization with Multiple Constraints](https://mlsanity.com/paper/2108.01850/) (2021)
- [CoCon: A Self-Supervised Approach for Controlled Text Generation](https://mlsanity.com/paper/2006.03535/) (2020)
- [Data Boost: Text Data Augmentation Through Reinforcement Learning Guided Conditional Generation](https://mlsanity.com/paper/2012.02952/) (2020)
- [DIRECTOR: Generator-Classifiers For Supervised Language Modeling](https://mlsanity.com/paper/2206.07694/) (2022)
- [Controllable Generation from Pre-trained Language Models via Inverse Prompting](https://mlsanity.com/paper/2103.10685/) (2021)
- [Attribute Alignment: Controlling Text Generation from Pre-trained Language Models](https://mlsanity.com/paper/2103.11070/) (2021)
- [Tailor: A Prompt-Based Approach to Attribute-Based Controlled Text Generation](https://mlsanity.com/paper/2204.13362/) (2022)
- [Plug-and-Blend: A Framework for Controllable Story Generation with Blended Control Codes](https://mlsanity.com/paper/2104.04039/) (2021)

## Built on by (most cited first, 30 of 30)

- [Controlled Decoding from Language Models](https://mlsanity.com/paper/2310.17022/) (2023)
- [ChatGPT vs Human-authored Text: Insights into Controllable Text Summarization and Sentence Style Transfer](https://mlsanity.com/paper/2306.07799/) (2023)
- [Tag-LLM: Repurposing General-Purpose LLMs for Specialized Domains](https://mlsanity.com/paper/2402.05140/) (2024)
- [Controllable Text Generation via Probability Density Estimation in the Latent Space](https://mlsanity.com/paper/2212.08307/) (2022)
- [Benchmarking Large Language Models on Controllable Generation under Diversified Instructions](https://mlsanity.com/paper/2401.00690/) (2024)
- [BOLT: Fast Energy-based Controlled Text Generation with Tunable Biases](https://mlsanity.com/paper/2305.12018/) (2023)
- [Generating Summaries with Controllable Readability Levels](https://mlsanity.com/paper/2310.10623/) (2023)
- [Mixture of Soft Prompts for Controllable Data Generation](https://mlsanity.com/paper/2303.01580/) (2023)
- [Controlled Text Generation for Large Language Model with Dynamic Attribute Graphs](https://mlsanity.com/paper/2402.11218/) (2024)
- [Reinforcement Learning with Token-level Feedback for Controllable Text Generation](https://mlsanity.com/paper/2403.11558/) (2024)
- [RSA-Control: A Pragmatics-Grounded Lightweight Controllable Text Generation Framework](https://mlsanity.com/paper/2410.19109/) (2024)
- [Air-Decoding: Attribute Distribution Reconstruction for Decoding-Time Controllable Text Generation](https://mlsanity.com/paper/2310.14892/) (2023)
- [An Invariant Learning Characterization of Controlled Text Generation](https://mlsanity.com/paper/2306.00198/) (2023)
- [Less Likely Brainstorming: Using Language Models to Generate Alternative Hypotheses](https://mlsanity.com/paper/2305.19339/) (2023)
- [PoetryDiffusion: Towards Joint Semantic and Metrical Manipulation in Poetry Generation](https://mlsanity.com/paper/2306.08456/) (2023)
- [Controllable Text Generation in the Instruction-Tuning Era](https://mlsanity.com/paper/2405.01490/) (2024)
- [Guaranteed Generation from Large Language Models](https://mlsanity.com/paper/2410.06716/) (2024)
- [FreeCtrl: Constructing Control Centers with Feedforward Layers for Learning-Free Controllable Text Generation](https://mlsanity.com/paper/2406.09688/) (2024)
- [Advancing Precise Outline-Conditioned Text Generation with Task Duality and Explicit Outline Control](https://mlsanity.com/paper/2305.14459/) (2023)
- [Enhancing Court View Generation with Knowledge Injection and Guidance](https://mlsanity.com/paper/2403.04366/) (2024)
- [Controllable Text Generation with Residual Memory Transformer](https://mlsanity.com/paper/2309.16231/) (2023)
- [GTA: Gated Toxicity Avoidance for LM Performance Preservation](https://mlsanity.com/paper/2312.06122/) (2023)
- [LiFi: Lightweight Controlled Text Generation with Fine-Grained Control Codes](https://mlsanity.com/paper/2402.06930/) (2024)
- [Successor Features for Efficient Multisubject Controlled Text Generation](https://mlsanity.com/paper/2311.04921/) (2023)
- [LaPA2: Length-Aware Prefix and Prompt Attention Augmentation for Long-Form Controllable Text Generation](https://mlsanity.com/paper/2508.04047/) (2025)
- [Funny or Persuasive, but Not Both: Evaluating Fine-Grained Multi-Concept Control in LLMs](https://mlsanity.com/paper/2601.18483/) (2026)
- [Towards Lightweight, Adaptive and Attribute-Aware Multi-Aspect Controllable Text Generation with Large Language Models](https://mlsanity.com/paper/2502.13474/) (2025)
- [Cognitive-Level Adaptive Generation via Capability-Aware Retrieval and Style Adaptation](https://mlsanity.com/paper/2509.19336/) (2025)
- [A Comparative Study of Controlled Text Generation Systems Using Level-Playing-Field Evaluation Principles](https://mlsanity.com/paper/2605.12395/) (2026)
- [Beyond Tokens: A Survey on Decoding Methods for Large Language and Vision-Language Models](https://mlsanity.com/paper/2608.14797/) (2026)

## Abstract (arXiv)

Prompt learning with immensely large Casual Language Models (CLMs) has been shown promising for attribute-controllable text generation (CTG). However, vanilla prompt tuning tends to imitate training corpus characteristics beyond the control attributes, resulting in a poor generalization ability. Moreover, it is less able to capture the relationship between different attributes, further limiting the control performance. In this paper, we propose a new CTG approach, namely DisCup, which incorporates the attribute knowledge of discriminator to optimize the control-prompts, steering a frozen CLM to produce attribute-specific texts. Specifically, the frozen CLM model, capable of producing multitudinous texts, is first used to generate the next-token candidates based on the context, so as to ensure the diversity of tokens to be predicted. Then, we leverage an attribute-discriminator to select desired/undesired tokens from those candidates, providing the inter-attribute knowledge. Finally, we bridge the above two traits by an unlikelihood objective for prompt-tuning. Extensive experimental results show that DisCup can achieve a new state-of-the-art control performance while maintaining an efficient and high-quality text generation, only relying on around 10 virtual tokens.