ATGen: A Framework for Active Text Generation

2025/06/29 by Akim Tsvigun, Daniil Vasilev, Tsvigun, Akim +27
Computer Science · Medicine · #Artificial Intelligence (cs.AI) #Artificial Intelligence in Healthcare and Education #Computation and Language (cs.CL) #FOS: Computer and information sciences #Multimodal Machine Learning Applications #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2506.23342

openalex publication_date 2025/06/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Active learning (AL) has demonstrated remarkable potential in reducing the annotation effort required for training machine learning models. However, despite the surging popularity of natural language generation (NLG) tasks in recent years, the application of AL to NLG has been limited. In this paper, we introduce Active Text Generation (ATGen) - a comprehensive framework that bridges AL with text generation tasks, enabling the application of state-of-the-art AL strategies to NLG. Our framework simplifies AL-empowered annotation in NLG tasks using both human annotators and automatic annotation agents based on large language models (LLMs). The framework supports LLMs deployed as services, such as ChatGPT and Claude, or operated on-premises. Furthermore, ATGen provides a unified platform for smooth implementation and benchmarking of novel AL strategies tailored to NLG tasks. Finally, we present evaluation results for state-of-the-art AL strategies across diverse settings and multiple text generation tasks. We show that ATGen reduces both the effort of human annotators and costs associated with API calls to LLM-based annotation agents. The code of the framework is available on GitHub under the MIT license. The video presentation is available at http://atgen-video.nlpresearch.group

Citations

• Comparing the Efficacy of GPT-4 and Chat-GPT in Mental Health Care: A Blind Assessment of Large Language Models for Psychological Support
• To Label or Not to Label: Hybrid Active Learning for Neural Machine Translation
• DoRA: Weight-Decomposed Low-Rank Adaptation
• SGLang: Efficient Execution of Structured Language Model Programs
• Efficient Memory Management for Large Language Model Serving with PagedAttention
• AlignScore: Evaluating Factual Consistency with a Unified Alignment Function
• QLoRA: Efficient Finetuning of Quantized LLMs
• G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment
• Self-Instruct: Aligning Language Models with Self-Generated Instructions
• Unnatural Instructions: Tuning Language Models with (Almost) No Human Labor
• FAMIE: A Fast Active Learning Framework for Multilingual Information Extraction
• DeepAL: Deep Active Learning in Python
• Training Verifiers to Solve Math Word Problems
• Active Learning by Acquiring Contrastive Examples
• Datasets: A Community Library for Natural Language Processing
• Revisiting Uncertainty-based Query Strategies for Active Learning with Transformers
• BARTScore: Evaluating Generated Text as Text Generation
• Phrase-level Active Learning for Neural Machine Translation
• Active Learning for Sequence Tagging with Deep Pre-trained Models and Bayesian Uncertainty Estimates
• Cold-start Active Learning through Self-supervised Language Modeling
• Wat zei je? Detecting Out-of-Distribution Translations with Variational Transformers
• HuggingFace's Transformers: State-of-the-art Natural Language Processing
• This Email Could Save Your Life: Introducing the Task of Email Subject Line Generation
• BERTScore: Evaluating Text Generation with BERT
• Parameter-Efficient Transfer Learning for NLP
• Deep Active Learning for Named Entity Recognition
• TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for\n Reading Comprehension
• RACE: Large-scale ReAding Comprehension Dataset From Examinations

Related

• Active Learning for Natural Language Generation
• From Selection to Generation: A Survey of LLM-based Active Learning
• Active Learning for NLP with Large Language Models
• Applying LLMs to Active Learning: Towards Cost-Efficient Cross-Task Text Classification without Manually Labeled Data
• ALANNO: An Active Learning Annotation System for Mortals