mlsanityData to 7 Oct 2026

Controlled Text Generation as Continuous Optimization with Multiple Constraints

arXiv 2108.01850 ↗First posted 4 Aug 2021cs.CLcited by 70

Not on the map. Why?

Abstract (arXiv)

As large-scale language model pretraining pushes the state-of-the-art in text generation, recent work has turned to controlling attributes of the text such models generate. While modifying the pretrained models via fine-tuning remains the popular approach, it incurs a significant computational cost and can be infeasible due to lack of appropriate data. As an alternative, we propose MuCoCO -- a flexible and modular algorithm for controllable inference from pretrained models. We formulate the decoding process as an optimization problem which allows for multiple attributes we aim to control to be easily incorporated as differentiable constraints to the optimization. By relaxing this discrete optimization to a continuous one, we make use of Lagrangian multipliers and gradient-descent based techniques to generate the desired text. We evaluate our approach on controllable machine translation and style transfer with multiple sentence-level attributes and observe significant improvements over baselines.

Key facts. Cited by 70 papers in the map (data through 7 Oct 2026). Builds on 59 of its 85 references that are in the map. First posted on arXiv 4 Aug 2021.

Builds on

59 of 85 references.

References · 59

  1. 2017Attention Is All You Needcited by 59,932
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  5. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  6. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
  7. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,462
  8. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,878
  9. 2017A Broad-Coverage Challenge Corpus for Sentence Understanding through Inferencecited by 2,665
  10. 2019HuggingFace's Transformers: State-of-the-art Natural Language Processingcited by 2,201
  11. 2019Fine-Tuning Language Models from Human Preferencescited by 2,124
  12. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,031
38 more
  1. 2016Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translationcited by 1,471
  2. 2017Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semanticscited by 1,345
  3. 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,235
  4. 2017SemEval-2017 Task 1: Semantic Textual Similarity - Multilingual and Cross-lingual Focused Evaluationcited by 998
  5. 2018Neural Network Acceptability Judgmentscited by 952
  6. 2017GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networkscited by 816
  7. 2018Multi-Task Learning as Multi-Objective Optimizationcited by 809
  8. 2019Plug and Play Language Models: A Simple Approach to Controlled Text Generationcited by 793
  9. 2019CTRL: A Conditional Transformer Language Model for Controllable Generationcited by 754
  10. 2019Universal Adversarial Triggers for Attacking and Analyzing NLPcited by 709
  11. 2020Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillationcited by 591
  12. 2019Defending Against Neural Fake Newscited by 553
  13. 2019Neural Text Generation with Unlikelihood Trainingcited by 489
  14. 2017Non-Autoregressive Neural Machine Translationcited by 457
  15. 2021DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Expertscited by 434
  16. 2020GeDi: Generative Discriminator Guided Sequence Generationcited by 377
  17. 2016SeqGAN: Sequence Generative Adversarial Nets with Policy Gradientcited by 375
  18. 2021FUDGE: Controlled Text Generation With Future Discriminatorscited by 374
  19. 2021Understanding Factuality in Abstractive Summarization with FRANK: A Benchmark for Factuality Metricscited by 285
  20. 2017Toward Controlled Generation of Textcited by 271
  21. 2017Style Transfer from Non-Parallel Text by Cross-Alignmentcited by 246
  22. 2017Lexically Constrained Decoding for Sequence Generation Using Grid Beam Searchcited by 234
  23. 2019Pareto Multi-Task Learningcited by 224
  24. 2018Dear Sir or Madam, May I introduce the GYAFC Dataset: Corpus, Benchmarks and Metrics for Formality Style Transfercited by 209
  25. 2018Marian: Fast Neural Machine Translation in C++cited by 202
  26. 2018Delete, Retrieve, Generate: A Simple Approach to Sentiment and Style Transfercited by 191
  27. 2020Reformulating Unsupervised Style Transfer as Paraphrase Generationcited by 181
  28. 2017Controlling Linguistic Style Aspects in Neural Language Generationcited by 147
  29. 2020NeuroLogic Decoding: (Un)supervised Neural Text Generation with Predicate Logic Constraintscited by 146
  30. 2018Style Transfer Through Back-Translationcited by 124
  31. 2019On NMT Search Errors and Model Errors: Cat Got Your Tongue?cited by 114
  32. 2019Beyond BLEU: Training Neural Machine Translation with Semantic Similaritycited by 84
  33. 2019Transforming Delete, Retrieve, Generate Approach for Controlled Text Style Transfercited by 81
  34. 2020Imitation Attacks and Defenses for Black-box Machine Translation Systemscited by 78
  35. 2020Back to the Future: Unsupervised Backprop-based Decoding for Counterfactual and Abductive Commonsense Reasoningcited by 65
  36. 2020PowerTransformer: Unsupervised Controllable Revision for Biased Language Correctioncited by 59
  37. 2017Classical Structured Prediction Losses for Sequence to Sequence Learningcited by 55
  38. 2020Controllable Pareto Multi-Task Learningcited by 51

50 most cited of 59 All 59 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 17; 2023: 19; 2024: 20; 2025: 9; 2026: 4.

70 papers.

Sort:

By area:

  1. 2022Generating Training Data with Language Models: Towards Zero-Shot Language Understandingcited by 199
  2. 2022SSD-LM: Semi-autoregressive Simplex-based Diffusion Language Model for Text Generation and Modular Controlcited by 191
  3. 2023Automatically Auditing Large Language Models via Discrete Optimizationcited by 187
  4. 2022COLD Decoding: Energy-based Constrained Text Generation with Langevin Dynamicscited by 159
  5. 2024Decoding-Time Language Model Alignment with Multiple Objectivescited by 90
  6. 2022Language Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Surveycited by 80
  7. 2023Sequential Monte Carlo Steering of Large Language Models using Probabilistic Programscited by 79
  8. 2023Word Embeddings Are Steers for Language Modelscited by 70
  9. 2022Mix and Match: Learning-free Controllable Text Generation using Energy Language Modelscited by 66
  10. 2021Prompt Waywardness: The Curious Case of Discretized Interpretation of Continuous Promptscited by 64
  11. 2022Gradient-Based Constrained Sampling from Language Modelscited by 64
  12. 2023Controlled Text Generation via Language Model Arithmeticcited by 51
38 more
  1. 2022Critic-Guided Decoding for Controlled Text Generationcited by 45
  2. 2025Syntactic and Semantic Control of Large Language Models via Sequential Monte Carlocited by 45
  3. 2022Plansformer: Generating Symbolic Plans using Transformerscited by 41
  4. 2022Tuning Language Models as Training Data Generators for Augmentation-Enhanced Few-Shot Learningcited by 36
  5. 2024Constrained Decoding for Secure Code Generationcited by 35
  6. 2022A Distributional Lens for Multi-Aspect Controllable Text Generationcited by 33
  7. 2023Inference-Time Policy Adapters (IPA): Tailoring Extreme-Scale LMs without Fine-tuningcited by 33
  8. 2024On the Challenges and Opportunities in Generative AIcited by 31
  9. 2022DisCup: Discriminator Cooperative Unlikelihood Prompt-tuning for Controllable Text Generationcited by 30
  10. 2022Composable Text Controls in Latent Space with ODEscited by 28
  11. 2022Controllable Text Generation via Probability Density Estimation in the Latent Spacecited by 27
  12. 2023BOLT: Fast Energy-based Controlled Text Generation with Tunable Biasescited by 24
  13. 2024LiSeCo: Linear Semantic Control for Language Generationcited by 20
  14. 2023ParaGuide: Guided Diffusion Paraphrasers for Plug-and-Play Textual Style Transfercited by 18
  15. 2022Socratic Pretraining: Question-Driven Pretraining for Controllable Summarizationcited by 17
  16. 2024Reinforcement Learning with Token-level Feedback for Controllable Text Generationcited by 16
  17. 2024TinyStyler: Efficient Few-Shot Text Style Transfer with Authorship Embeddingscited by 16
  18. 2022BioNLI: Generating a Biomedical NLI Dataset Using Lexico-semantic Constraints for Adversarial Examplescited by 15
  19. 2023Controlling Pre-trained Language Models for Grade-Specific Text Simplificationcited by 14
  20. 2023MacLaSa: Multi-Aspect Controllable Text Generation via Efficient Sampling from Compact Latent Spacecited by 12
  21. 2024Outcome-Constrained Large Language Models for Countering Hate Speechcited by 12
  22. 2024Analysing Zero-Shot Readability-Controlled Sentence Simplificationcited by 12
  23. 2024CLIBE: Detecting Dynamic Backdoors in Transformer-based NLP Modelscited by 11
  24. 2022An Extensible Plug-and-Play Method for Multi-Aspect Controllable Text Generationcited by 10
  25. 2023ECGBERT: Understanding Hidden Language of ECGs with Self-Supervised Representation Learningcited by 10
  26. 2023Air-Decoding: Attribute Distribution Reconstruction for Decoding-Time Controllable Text Generationcited by 10
  27. 2023Most Language Models can be Poets too: An AI Writing Assistant and Constrained Text Generation Studiocited by 9
  28. 2024Growing a Tail: Increasing Output Diversity in Large Language Modelscited by 9
  29. 2022Mitigating Toxic Degeneration with Empathetic Data: Exploring the Relationship Between Toxicity and Empathycited by 8
  30. 2025Inference-time Alignment in Continuous Spacecited by 7
  31. 2023Graph-Induced Syntactic-Semantic Spaces in Transformer-Based Variational AutoEncoderscited by 6
  32. 2023A Block Metropolis-Hastings Sampler for Controllable Energy-based Text Generationcited by 5
  33. 2023PlaSma: Making Small Language Models Better Procedural Knowledge Models for (Counterfactual) Planningcited by 4
  34. 2023Structured Voronoi Samplingcited by 4
  35. 2023P^3SUM: Preserving Author's Perspective in News Summarization with Diffusion Language Modelscited by 4
  36. 2024Language Rectified Flow: Advancing Diffusion Language Generation with Probabilistic Flowscited by 4
  37. 2025You Cannot Feed Two Birds with One Score: the Accuracy-Naturalness Tradeoff in Translationcited by 4
  38. 2026nabla-Reasoner: LLM Reasoning via Test-Time Gradient Descent in Latent Spacecited by 4

50 most cited of 70 All 70 (JSON)