Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2402.09954 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916129445773312 |
|---|---|
| author | Pu, Jiashu Wan, Yajing Zhang, Yuru Chen, Jing Cheng, Ling Shao, Qian Chang, Yongzhu Lv, Tangjie Zhang, Rongsheng |
| author_facet | Pu, Jiashu Wan, Yajing Zhang, Yuru Chen, Jing Cheng, Ling Shao, Qian Chang, Yongzhu Lv, Tangjie Zhang, Rongsheng |
| contents | Previous in-context learning (ICL) research has focused on tasks such as classification, machine translation, text2table, etc., while studies on whether ICL can improve human-like dialogue generation are scarce. Our work fills this gap by systematically investigating the ICL capabilities of large language models (LLMs) in persona-based dialogue generation, conducting extensive experiments on high-quality real human Chinese dialogue datasets. From experimental results, we draw three conclusions: 1) adjusting prompt instructions is the most direct, effective, and economical way to improve generation quality; 2) randomly retrieving demonstrations (demos) achieves the best results, possibly due to the greater diversity and the amount of effective information; counter-intuitively, retrieving demos with a context identical to the query performs the worst; 3) even when we destroy the multi-turn associations and single-turn semantics in the demos, increasing the number of demos still improves dialogue performance, proving that LLMs can learn from corrupted dialogue demos. Previous explanations of the ICL mechanism, such as $n$-gram induction head, cannot fully account for this phenomenon. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_09954 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Crafting a Good Prompt or Providing Exemplary Dialogues? A Study of In-Context Learning for Persona-based Dialogue Generation Pu, Jiashu Wan, Yajing Zhang, Yuru Chen, Jing Cheng, Ling Shao, Qian Chang, Yongzhu Lv, Tangjie Zhang, Rongsheng Computation and Language Machine Learning Previous in-context learning (ICL) research has focused on tasks such as classification, machine translation, text2table, etc., while studies on whether ICL can improve human-like dialogue generation are scarce. Our work fills this gap by systematically investigating the ICL capabilities of large language models (LLMs) in persona-based dialogue generation, conducting extensive experiments on high-quality real human Chinese dialogue datasets. From experimental results, we draw three conclusions: 1) adjusting prompt instructions is the most direct, effective, and economical way to improve generation quality; 2) randomly retrieving demonstrations (demos) achieves the best results, possibly due to the greater diversity and the amount of effective information; counter-intuitively, retrieving demos with a context identical to the query performs the worst; 3) even when we destroy the multi-turn associations and single-turn semantics in the demos, increasing the number of demos still improves dialogue performance, proving that LLMs can learn from corrupted dialogue demos. Previous explanations of the ICL mechanism, such as $n$-gram induction head, cannot fully account for this phenomenon. |
| title | Crafting a Good Prompt or Providing Exemplary Dialogues? A Study of In-Context Learning for Persona-based Dialogue Generation |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2402.09954 |