# Controllable Text Generation via Probability Density Estimation in the Latent Space

arXiv: https://arxiv.org/abs/2212.08307 · HTML: https://mlsanity.com/paper/2212.08307/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- Builds on 41 of its 54 references that are in the map.
- First posted on arXiv 16 Dec 2022.

## Builds on (most cited first, 30 of 41)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [Character-level Convolutional Networks for Text Classification](https://mlsanity.com/paper/1509.01626/) (2015)
- [Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing](https://mlsanity.com/paper/2107.13586/) (2021)
- [Fine-Tuning Language Models from Human Preferences](https://mlsanity.com/paper/1909.08593/) (2019)
- [Density estimation using Real NVP](https://mlsanity.com/paper/1605.08803/) (2016)
- [Glow: Generative Flow with Invertible 1x1 Convolutions](https://mlsanity.com/paper/1807.03039/) (2018)
- [A Diversity-Promoting Objective Function for Neural Conversation Models](https://mlsanity.com/paper/1510.03055/) (2015)
- [NICE: Non-linear Independent Components Estimation](https://mlsanity.com/paper/1410.8516/) (2014)
- [Diffusion-LM Improves Controllable Text Generation](https://mlsanity.com/paper/2205.14217/) (2022)
- [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)
- [Masked Autoregressive Flow for Density Estimation](https://mlsanity.com/paper/1705.07057/) (2017)
- [Improving Variational Inference with Inverse Autoregressive Flow](https://mlsanity.com/paper/1606.04934/) (2016)
- [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)
- [Toward Controlled Generation of Text](https://mlsanity.com/paper/1703.00955/) (2017)
- [Quark: Controllable Text Generation with Reinforced Unlearning](https://mlsanity.com/paper/2205.13636/) (2022)
- [COLD Decoding: Energy-based Constrained Text Generation with Langevin Dynamics](https://mlsanity.com/paper/2202.11705/) (2022)
- [Controlling Linguistic Style Aspects in Neural Language Generation](https://mlsanity.com/paper/1707.02633/) (2017)
- [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)
- [FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow](https://mlsanity.com/paper/1909.02480/) (2019)
- [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)
- [Discrete Flows: Invertible Generative Models of Discrete Data](https://mlsanity.com/paper/1905.10347/) (2019)

## Built on by (most cited first, 27 of 27)

- [A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions](https://mlsanity.com/paper/2311.05232/) (2023)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [Automatic Metrics in Natural Language Generation: A Survey of Current Evaluation Practices](https://mlsanity.com/paper/2408.09169/) (2024)
- [Reinforcement Learning with Token-level Feedback for Controllable Text Generation](https://mlsanity.com/paper/2403.11558/) (2024)
- [Has this Fact been Edited? Detecting Knowledge Edits in Language Models](https://mlsanity.com/paper/2405.02765/) (2024)
- [MacLaSa: Multi-Aspect Controllable Text Generation via Efficient Sampling from Compact Latent Space](https://mlsanity.com/paper/2305.12785/) (2023)
- [DiffLM: Controllable Synthetic Data Generation via Diffusion Language Models](https://mlsanity.com/paper/2411.03250/) (2024)
- [A Survey on Responsible Generative AI: What to Generate and What Not](https://mlsanity.com/paper/2404.05783/) (2024)
- [Learning Disentangled Semantic Spaces of Explanations via Invertible Neural Networks](https://mlsanity.com/paper/2305.01713/) (2023)
- [FreeCtrl: Constructing Control Centers with Feedforward Layers for Learning-Free Controllable Text Generation](https://mlsanity.com/paper/2406.09688/) (2024)
- [LiFi: Lightweight Controlled Text Generation with Fine-Grained Control Codes](https://mlsanity.com/paper/2402.06930/) (2024)
- [Multi-Aspect Controllable Text Generation with Disentangled Counterfactual Augmentation](https://mlsanity.com/paper/2405.19958/) (2024)
- [LangVAE and LangSpace: Building and Probing for Language Model VAEs](https://mlsanity.com/paper/2505.00004/) (2025)
- [C3TG: Conflict-aware, Composite, and Collaborative Controlled Text Generation](https://mlsanity.com/paper/2511.09292/) (2025)
- [Linguistically-Controlled Paraphrase Generation](https://mlsanity.com/paper/2410.24199/) (2024)
- [Funny or Persuasive, but Not Both: Evaluating Fine-Grained Multi-Concept Control in LLMs](https://mlsanity.com/paper/2601.18483/) (2026)
- [Reproducing the Metric-Based Evaluation of a Set of Controllable Text Generation Techniques](https://mlsanity.com/paper/2405.07875/) (2024)
- [Towards Lightweight, Adaptive and Attribute-Aware Multi-Aspect Controllable Text Generation with Large Language Models](https://mlsanity.com/paper/2502.13474/) (2025)
- [Bridging Compositional and Distributional Semantics: A Survey on Latent Semantic Geometry via AutoEncoder](https://mlsanity.com/paper/2506.20083/) (2025)
- [That's Deprecated! Understanding, Detecting, and Steering Knowledge Conflicts in Language Models for Code Generation](https://mlsanity.com/paper/2510.19116/) (2025)
- [OLLM: Options-based Large Language Models](https://mlsanity.com/paper/2604.19087/) (2026)
- [Output Composability of QLoRA PEFT Modules for Plug-and-Play Attribute-Controlled Text Generation](https://mlsanity.com/paper/2605.12345/) (2026)
- [A Comparative Study of Controlled Text Generation Systems Using Level-Playing-Field Evaluation Principles](https://mlsanity.com/paper/2605.12395/) (2026)
- [Latent Heuristic Search: Continuous Optimization for Automated Algorithm Design](https://mlsanity.com/paper/2605.17137/) (2026)
- [Uncovering Latent Reasoning Strategies in Language Models](https://mlsanity.com/paper/2607.17674/) (2026)
- [The Illusion of Control: Why Bare Classifier Inversion Silently Fails in Concept-Bottleneck Text Generation](https://mlsanity.com/paper/2608.22956/) (2026)
- [Unlocking Latent Personalization in LLMs](https://mlsanity.com/paper/2609.33182/) (2026)

## Abstract (arXiv)

Previous work on controllable text generation has explored the idea of control from the latent space, such as optimizing a representation with attribute-related classifiers or sampling a representation from relevant discrete samples. However, they are not effective enough in modeling both the latent space and the control, leaving controlled text with low quality and diversity. In this work, we propose a novel control framework using probability density estimation in the latent space. Our method utilizes an invertible transformation function, the Normalizing Flow, that maps the complex distributions in the latent space to simple Gaussian distributions in the prior space. Thus, we can perform sophisticated and flexible control in the prior space and feed the control effects back into the latent space owing to the one-one-mapping property of invertible transformations. Experiments on single-attribute controls and multi-attribute control reveal that our method outperforms several strong baselines on attribute relevance and text quality and achieves the SOTA. Further analysis of control strength adjustment demonstrates the flexibility of our control strategy.