Saved in:
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2506.23342 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913917659250688 |
|---|---|
| author | Tsvigun, Akim Vasilev, Daniil Tsvigun, Ivan Lysenko, Ivan Bektleuov, Talgat Medvedev, Aleksandr Vinogradova, Uliana Severin, Nikita Mozikov, Mikhail Savchenko, Andrey Grigorev, Rostislav Kuleev, Ramil Zhdanov, Fedor Shelmanov, Artem Makarov, Ilya |
| author_facet | Tsvigun, Akim Vasilev, Daniil Tsvigun, Ivan Lysenko, Ivan Bektleuov, Talgat Medvedev, Aleksandr Vinogradova, Uliana Severin, Nikita Mozikov, Mikhail Savchenko, Andrey Grigorev, Rostislav Kuleev, Ramil Zhdanov, Fedor Shelmanov, Artem Makarov, Ilya |
| contents | 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 |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_23342 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ATGen: A Framework for Active Text Generation Tsvigun, Akim Vasilev, Daniil Tsvigun, Ivan Lysenko, Ivan Bektleuov, Talgat Medvedev, Aleksandr Vinogradova, Uliana Severin, Nikita Mozikov, Mikhail Savchenko, Andrey Grigorev, Rostislav Kuleev, Ramil Zhdanov, Fedor Shelmanov, Artem Makarov, Ilya Computation and Language Artificial Intelligence 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 |
| title | ATGen: A Framework for Active Text Generation |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2506.23342 |