LLM Attributor: Interactive Visual Attribution for LLM Generation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909156914495488 |
|---|---|
| author | Lee, Seongmin Wang, Zijie J. Chakravarthy, Aishwarya Helbling, Alec Peng, ShengYun Phute, Mansi Chau, Duen Horng Kahng, Minsuk |
| author_facet | Lee, Seongmin Wang, Zijie J. Chakravarthy, Aishwarya Helbling, Alec Peng, ShengYun Phute, Mansi Chau, Duen Horng Kahng, Minsuk |
| contents | While large language models (LLMs) have shown remarkable capability to generate convincing text across diverse domains, concerns around its potential risks have highlighted the importance of understanding the rationale behind text generation. We present LLM Attributor, a Python library that provides interactive visualizations for training data attribution of an LLM's text generation. Our library offers a new way to quickly attribute an LLM's text generation to training data points to inspect model behaviors, enhance its trustworthiness, and compare model-generated text with user-provided text. We describe the visual and interactive design of our tool and highlight usage scenarios for LLaMA2 models fine-tuned with two different datasets: online articles about recent disasters and finance-related question-answer pairs. Thanks to LLM Attributor's broad support for computational notebooks, users can easily integrate it into their workflow to interactively visualize attributions of their models. For easier access and extensibility, we open-source LLM Attributor at https://github.com/poloclub/ LLM-Attribution. The video demo is available at https://youtu.be/mIG2MDQKQxM. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_01361 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | LLM Attributor: Interactive Visual Attribution for LLM Generation Lee, Seongmin Wang, Zijie J. Chakravarthy, Aishwarya Helbling, Alec Peng, ShengYun Phute, Mansi Chau, Duen Horng Kahng, Minsuk Computation and Language Artificial Intelligence Human-Computer Interaction Machine Learning While large language models (LLMs) have shown remarkable capability to generate convincing text across diverse domains, concerns around its potential risks have highlighted the importance of understanding the rationale behind text generation. We present LLM Attributor, a Python library that provides interactive visualizations for training data attribution of an LLM's text generation. Our library offers a new way to quickly attribute an LLM's text generation to training data points to inspect model behaviors, enhance its trustworthiness, and compare model-generated text with user-provided text. We describe the visual and interactive design of our tool and highlight usage scenarios for LLaMA2 models fine-tuned with two different datasets: online articles about recent disasters and finance-related question-answer pairs. Thanks to LLM Attributor's broad support for computational notebooks, users can easily integrate it into their workflow to interactively visualize attributions of their models. For easier access and extensibility, we open-source LLM Attributor at https://github.com/poloclub/ LLM-Attribution. The video demo is available at https://youtu.be/mIG2MDQKQxM. |
| title | LLM Attributor: Interactive Visual Attribution for LLM Generation |
| topic | Computation and Language Artificial Intelligence Human-Computer Interaction Machine Learning |
| url | https://arxiv.org/abs/2404.01361 |