Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2305.11789 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866929227679399936 |
|---|---|
| author | Kaneko, Masahiro Neubig, Graham Okazaki, Naoaki |
| author_facet | Kaneko, Masahiro Neubig, Graham Okazaki, Naoaki |
| contents | Humans work together to solve common problems by having discussions, explaining, and agreeing or disagreeing with each other. Similarly, if a system can have discussions with humans when solving tasks, it can improve the system's performance and reliability. In previous research on explainability, it has only been possible for the system to make predictions and for humans to ask questions about them rather than having a mutual exchange of opinions. This research aims to create a dataset and computational framework for systems that discuss and refine their predictions through dialogue. Through experiments, we show that the proposed system can have beneficial discussions with humans improving the accuracy by up to 25 points in the natural language inference task. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2305_11789 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Solving NLP Problems through Human-System Collaboration: A Discussion-based Approach Kaneko, Masahiro Neubig, Graham Okazaki, Naoaki Computation and Language Humans work together to solve common problems by having discussions, explaining, and agreeing or disagreeing with each other. Similarly, if a system can have discussions with humans when solving tasks, it can improve the system's performance and reliability. In previous research on explainability, it has only been possible for the system to make predictions and for humans to ask questions about them rather than having a mutual exchange of opinions. This research aims to create a dataset and computational framework for systems that discuss and refine their predictions through dialogue. Through experiments, we show that the proposed system can have beneficial discussions with humans improving the accuracy by up to 25 points in the natural language inference task. |
| title | Solving NLP Problems through Human-System Collaboration: A Discussion-based Approach |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2305.11789 |