Enabling Natural Zero-Shot Prompting on Encoder Models via Statement-Tuning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866910653447405568 |
|---|---|
| author | Elshabrawy, Ahmed Huang, Yongxin Gurevych, Iryna Aji, Alham Fikri |
| author_facet | Elshabrawy, Ahmed Huang, Yongxin Gurevych, Iryna Aji, Alham Fikri |
| contents | While Large Language Models (LLMs) exhibit remarkable capabilities in zero-shot and few-shot scenarios, they often require computationally prohibitive sizes. Conversely, smaller Masked Language Models (MLMs) like BERT and RoBERTa achieve state-of-the-art results through fine-tuning but struggle with extending to few-shot and zero-shot settings due to their architectural constraints. Hence, we propose Statement-Tuning, a technique that models discriminative tasks as a set of finite statements and trains an encoder model to discriminate between the potential statements to determine the label. We do Statement-Tuning on multiple tasks to enable cross-task generalization. Experimental results demonstrate that Statement-Tuning achieves competitive performance compared to state-of-the-art LLMs with significantly fewer parameters. Moreover, the study investigates the impact of several design choices on few-shot and zero-shot generalization, revealing that Statement-Tuning can achieve strong performance with modest training data and benefits from task and statement diversity for unseen task generalizability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_12897 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Enabling Natural Zero-Shot Prompting on Encoder Models via Statement-Tuning Elshabrawy, Ahmed Huang, Yongxin Gurevych, Iryna Aji, Alham Fikri Computation and Language While Large Language Models (LLMs) exhibit remarkable capabilities in zero-shot and few-shot scenarios, they often require computationally prohibitive sizes. Conversely, smaller Masked Language Models (MLMs) like BERT and RoBERTa achieve state-of-the-art results through fine-tuning but struggle with extending to few-shot and zero-shot settings due to their architectural constraints. Hence, we propose Statement-Tuning, a technique that models discriminative tasks as a set of finite statements and trains an encoder model to discriminate between the potential statements to determine the label. We do Statement-Tuning on multiple tasks to enable cross-task generalization. Experimental results demonstrate that Statement-Tuning achieves competitive performance compared to state-of-the-art LLMs with significantly fewer parameters. Moreover, the study investigates the impact of several design choices on few-shot and zero-shot generalization, revealing that Statement-Tuning can achieve strong performance with modest training data and benefits from task and statement diversity for unseen task generalizability. |
| title | Enabling Natural Zero-Shot Prompting on Encoder Models via Statement-Tuning |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2404.12897 |