CTRL: A Conditional Transformer Language Model for Controllable Generation

Abstract

Large-scale language models show promising text generation capabilities, but users cannot easily control particular aspects of the generated text. We release CTRL, a 1.63 billion-parameter conditional transformer language model, trained to condition on control codes that govern style, content, and task-specific behavior. Control codes were derived from structure that naturally co-occurs with raw text, preserving the advantages of unsupervised learning while providing more explicit control over text generation. These codes also allow CTRL to predict which parts of the training data are most likely given a sequence. This provides a potential method for analyzing large amounts of data via model-based source attribution. We have released multiple full-sized, pretrained versions of CTRL at https://github.com/salesforce/ctrl.

References in corpus (14)

• Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
• Natural Language Processing (almost) from Scratch
• Cross-lingual Language Model Pretraining
• Layer Normalization
• Generating Long Sequences with Sparse Transformers
• Regularizing and Optimizing LSTM Language Models
• One Model To Learn Them All
• Neural Text Generation with Unlikelihood Training
• Tying Word Vectors and Word Classifiers: A Loss Framework for Language Modeling
• NewsQA: A Machine Comprehension Dataset
• Pretrained AI Models: Performativity, Mobility, and Change
• ELI5: Long Form Question Answering
• Answers Unite! Unsupervised Metrics for Reinforced Summarization Models
• SumQE: a BERT-based Summary Quality Estimation Model

Cited by in corpus (182)

• Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
• On the Opportunities and Risks of Foundation Models
• Pre-trained Models for Natural Language Processing: A Survey
• Evaluating Large Language Models Trained on Code
• Multilingual Denoising Pre-training for Neural Machine Translation
• Decision Transformer: Reinforcement Learning via Sequence Modeling
• Prefix-Tuning: Optimizing Continuous Prompts for Generation
• Towards a Human-like Open-Domain Chatbot
• TweepFake: about Detecting Deepfake Tweets
• A Simple Language Model for Task-Oriented Dialogue
• Plug and Play Language Models: A Simple Approach to Controlled Text Generation
• Data Augmentation using Pre-trained Transformer Models
• Large Language Models for Code: Security Hardening and Adversarial Testing
• SOLOIST: Building Task Bots at Scale with Transfer Learning and Machine Teaching
• Recipes for Safety in Open-domain Chatbots
• It's Morphin' Time! Combating Linguistic Discrimination with Inflectional Perturbations
• Survey of different Large Language Model Architectures: Trends, Benchmarks, and Challenges
• Few-Shot Text Generation with Pattern-Exploiting Training
• Abstractive Text Summarization: State of the Art, Challenges, and Improvements
• Few-Shot Named Entity Recognition: A Comprehensive Study
• Pop Music Transformer: Beat-based Modeling and Generation of Expressive Pop Piano Compositions
• Hash Layers For Large Sparse Models
• exBERT: A Visual Analysis Tool to Explore Learned Representations in Transformers Models
• Transformer-based Map Matching Model with Limited Ground-Truth Data using Transfer-Learning Approach
• Explaining Question Answering Models through Text Generation
• PlotMachines: Outline-Conditioned Generation with Dynamic Plot State Tracking
• Transformer-based Conditional Variational Autoencoder for Controllable Story Generation
• Wordcraft: a Human-AI Collaborative Editor for Story Writing
• ProGen: Language Modeling for Protein Generation
• Controlling Style in Generated Dialogue
• A Distributional Approach to Controlled Text Generation
• Few-shot Natural Language Generation for Task-Oriented Dialog
• Clinical Insights: A Comprehensive Review of Language Models in Medicine
• Artificial Text Detection via Examining the Topology of Attention Maps
• Controllable Semantic Parsing via Retrieval Augmentation
• GenNI: Human-AI Collaboration for Data-Backed Text Generation
• Efficient Retrieval Augmented Generation from Unstructured Knowledge for Task-Oriented Dialog
• Synthetic Data Augmentation for Zero-Shot Cross-Lingual Question Answering
• On Feature Normalization and Data Augmentation
• scb-mt-en-th-2020: A Large English-Thai Parallel Corpus
• Applying Transformer-based Text Summarization for Keyphrase Generation
• Data Movement Is All You Need: A Case Study on Optimizing Transformers
• Controllable Data Generation by Deep Learning: A Review
• Safety Analysis in the Era of Large Language Models: A Case Study of STPA using ChatGPT
• Dungeons and Dragons as a Dialog Challenge for Artificial Intelligence
• Neural Language Generation: Formulation, Methods, and Evaluation
• Adversarial Robustness of Neural-Statistical Features in Detection of Generative Transformers
• Say As You Wish: Fine-grained Control of Image Caption Generation with Abstract Scene Graphs
• Polyjuice: Generating Counterfactuals for Explaining, Evaluating, and Improving Models
• Compound Word Transformer: Learning to Compose Full-Song Music over Dynamic Directed Hypergraphs
• Residual Energy-Based Models for Text
• C5T5: Controllable Generation of Organic Molecules with Transformers
• Back to the Future: Unsupervised Backprop-based Decoding for Counterfactual and Abductive Commonsense Reasoning
• CoCon: A Self-Supervised Approach for Controlled Text Generation
• Anticipating Safety Issues in E2E Conversational AI: Framework and Tooling
• PLATO: Pre-trained Dialogue Generation Model with Discrete Latent Variable
• Finding Experts in Transformer Models
• SceneFormer: Indoor Scene Generation with Transformers
• Better Distractions: Transformer-based Distractor Generation and Multiple Choice Question Filtering
• AraGPT2: Pre-Trained Transformer for Arabic Language Generation
• DSGPT: Domain-Specific Generative Pre-Training of Transformers for Text Generation in E-commerce Title and Review Summarization
• Style Example-Guided Text Generation using Generative Adversarial Transformers
• Transferring Inductive Biases through Knowledge Distillation
• Findings of the The RuATD Shared Task 2022 on Artificial Text Detection in Russian
• Mirostat: A Neural Text Decoding Algorithm that Directly Controls Perplexity
• Deep Extrapolation for Attribute-Enhanced Generation
• MEGATRON-CNTRL: Controllable Story Generation with External Knowledge Using Large-Scale Language Models
• Ratatouille: A tool for Novel Recipe Generation
• Towards Controllable Biases in Language Generation
• DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts
• Zero-Shot Fine-Grained Style Transfer: Leveraging Distributed Continuous Style Representations to Transfer To Unseen Styles
• Detoxifying Language Models Risks Marginalizing Minority Voices
• TLDR: Token Loss Dynamic Reweighting for Reducing Repetitive Utterance Generation
• Optimus: Organizing Sentences via Pre-trained Modeling of a Latent Space
• WangchanBERTa: Pretraining transformer-based Thai Language Models
• Unsupervised Contextual Paraphrase Generation using Lexical Control and Reinforcement Learning
• Unsupervised Editing for Counterfactual Stories
• POINTER: Constrained Progressive Text Generation via Insertion-based Generative Pre-training
• PCAE: A Framework of Plug-in Conditional Auto-Encoder for Controllable Text Generation
• ETC-NLG: End-to-end Topic-Conditioned Natural Language Generation
• CBAG: Conditional Biomedical Abstract Generation
• Conditioned Natural Language Generation using only Unconditioned Language Model: An Exploration
• The Offense-Defense Balance of Scientific Knowledge: Does Publishing AI Research Reduce Misuse?
• Improved Natural Language Generation via Loss Truncation
• Keyphrase Generation with Cross-Document Attention
• RADDLE: An Evaluation Benchmark and Analysis Platform for Robust Task-oriented Dialog Systems
• Multimodal Conditionality for Natural Language Generation
• You are right. I am ALARMED -- But by Climate Change Counter Movement
• Conditional Text Generation for Harmonious Human-Machine Interaction
• The Rumour Mill: Making the Spread of Misinformation Explicit and Tangible
• PopMAG: Pop Music Accompaniment Generation
• DAGA: Data Augmentation with a Generation Approach for Low-resource Tagging Tasks
• Controllable Abstractive Dialogue Summarization with Sketch Supervision
• Human Language Modeling
• SideControl: Controlled Open-domain Dialogue Generation via Additive Side Networks
• Decentralized Attribution of Generative Models
• Template Guided Text Generation for Task-Oriented Dialogue
• What's New? Summarizing Contributions in Scientific Literature
• GeDi: Generative Discriminator Guided Sequence Generation
• Which *BERT? A Survey Organizing Contextualized Encoders
• KGPT: Knowledge-Grounded Pre-Training for Data-to-Text Generation
• Language Model Augmented Relevance Score
• Composed Variational Natural Language Generation for Few-shot Intents
• RSTGen: Imbuing Fine-Grained Interpretable Control into Long-FormText Generators
• Few-shot Sequence Learning with Transformers
• Outline to Story: Fine-grained Controllable Story Generation from Cascaded Events
• Conditional Augmentation for Aspect Term Extraction via Masked Sequence-to-Sequence Generation
• Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Model
• Neural Rule-Execution Tracking Machine For Transformer-Based Text Generation
• PatentTransformer-2: Controlling Patent Text Generation by Structural Metadata
• What Changes Can Large-scale Language Models Bring? Intensive Study on HyperCLOVA: Billions-scale Korean Generative Pretrained Transformers
• Reverse Engineering Configurations of Neural Text Generation Models
• Of Human Criteria and Automatic Metrics: A Benchmark of the Evaluation of Story Generation
• Nano: Nested Human-in-the-Loop Reward Learning for Few-shot Language Model Control
• Pre-train and Plug-in: Flexible Conditional Text Generation with Variational Auto-Encoders
• Conditional Neural Generation using Sub-Aspect Functions for Extractive News Summarization
• Staircase Attention for Recurrent Processing of Sequences
• PAIR: Planning and Iterative Refinement in Pre-trained Transformers for Long Text Generation
• Improving Diversity of Neural Text Generation via Inverse Probability Weighting
• You Can Do Better! If You Elaborate the Reason When Making Prediction
• Collocation2Text: Controllable Text Generation from Guide Phrases in Russian
• Adapting a Language Model for Controlled Affective Text Generation
• Argumentative Text Generation in Economic Domain
• Which Discriminator for Cooperative Text Generation?
• Melody Structure Transfer Network: Generating Music with Separable Self-Attention
• Investigation of Sentiment Controllable Chatbot
• Towards Generating Citation Sentences for Multiple References with Intent Control
• DocNLI: A Large-scale Dataset for Document-level Natural Language Inference
• Controlled Text Generation as Continuous Optimization with Multiple Constraints
• Language Model Evaluation in Open-ended Text Generation
• Changing the Mind of Transformers for Topically-Controllable Language Generation
• Generating Synthetic Text Data to Evaluate Causal Inference Methods
• COBE: Contextualized Object Embeddings from Narrated Instructional Video
• Directed Beam Search: Plug-and-Play Lexically Constrained Language Generation
• Zero-Shot Controlled Generation with Encoder-Decoder Transformers
• Increasing Faithfulness in Knowledge-Grounded Dialogue with Controllable Features
• Generative Models for Security: Attacks, Defenses, and Opportunities
• Self-conditioning pre-trained language models
• Towards Universality in Multilingual Text Rewriting
• Emergent Properties of Finetuned Language Representation Models
• Controllable Dialogue Generation with Disentangled Multi-grained Style Specification and Attribute Consistency Reward
• IGA : An Intent-Guided Authoring Assistant
• Plug-and-Blend: A Framework for Controllable Story Generation with Blended Control Codes
• Graphmax for Text Generation
• Goal-driven text descriptions for images
• Math Word Problem Generation with Mathematical Consistency and Problem Context Constraints
• Plan ahead: Self-Supervised Text Planning for Paragraph Completion Task
• TextSETTR: Few-Shot Text Style Extraction and Tunable Targeted Restyling
• Global-aware Beam Search for Neural Abstractive Summarization
• A New Approach to Overgenerating and Scoring Abstractive Summaries
• Metadata Shaping: Natural Language Annotations for the Tail
• Efficient Attribute Injection for Pretrained Language Models
• AI-generated text boundary detection with RoFT
• Classification as Decoder: Trading Flexibility for Control in Medical Dialogue
• Linguacodus: A Synergistic Framework for Transformative Code Generation in Machine Learning Pipelines
• Automatically Exposing Problems with Neural Dialog Models
• Facilitating the Communication of Politeness through Fine-Grained Paraphrasing
• Attribute Alignment: Controlling Text Generation from Pre-trained Language Models
• InFillmore: Frame-Guided Language Generation with Bidirectional Context
• Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding
• SHARE: a System for Hierarchical Assistive Recipe Editing
• Fingerprinting Fine-tuned Language Models in the Wild
• Towards Explainable Fact Checking
• WakaVT: A Sequential Variational Transformer for Waka Generation
• "TL;DR:" Out-of-Context Adversarial Text Summarization and Hashtag Recommendation
• DYPLOC: Dynamic Planning of Content Using Mixed Language Models for Text Generation
• Estimating Subjective Crowd-Evaluations as an Additional Objective to Improve Natural Language Generation
• Semantic Frame Forecast
• Logic-Consistency Text Generation from Semantic Parses
• Bag-of-Vectors Autoencoders for Unsupervised Conditional Text Generation
• LogicalFactChecker: Leveraging Logical Operations for Fact Checking with Graph Module Network
• MEDCOD: A Medically-Accurate, Emotive, Diverse, and Controllable Dialog System
• Realistic simulation of users for IT systems in cyber ranges
• CO-STAR: Conceptualisation of Stereotypes for Analysis and Reasoning
• Domain Conditional Predictors for Domain Adaptation
• Generating abstractive summaries of Lithuanian news articles using a transformer model
• Examining the rhetorical capacities of neural language models
• Multi-Pair Text Style Transfer on Unbalanced Data
• CFL: Causally Fair Language Models Through Token-level Attribute Controlled Generation
• Beyond [CLS] through Ranking by Generation
• Style Control for Schema-Guided Natural Language Generation
• Char2Subword: Extending the Subword Embedding Space Using Robust Character Compositionality