Awakening Augmented Generation: Learning to Awaken Internal Knowledge of Large Language Models for Question Answering
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916522293723136 |
|---|---|
| author | Liao, Huanxuan He, Shizhu Xu, Yao Zhang, Yuanzhe Liu, Kang Liu, Shengping Zhao, Jun |
| author_facet | Liao, Huanxuan He, Shizhu Xu, Yao Zhang, Yuanzhe Liu, Kang Liu, Shengping Zhao, Jun |
| contents | Retrieval-Augmented-Generation and Generation-Augmented-Generation have been proposed to enhance the knowledge required for question answering with Large Language Models (LLMs) by leveraging richer context. However, the former relies on external resources, and both require incorporating explicit documents into the context, which increases execution costs and susceptibility to noise data during inference. Recent works indicate that LLMs model rich knowledge, but it is often not effectively activated and awakened. Inspired by this, we propose a novel knowledge-augmented framework, $\textbf{Awakening-Augmented-Generation}$ (AAG), which mimics the human ability to answer questions using only thinking and recalling to compensate for knowledge gaps, thereby awaking relevant knowledge in LLMs without relying on external resources. AAG consists of two key components for awakening richer context. Explicit awakening fine-tunes a context generator to create a synthetic, compressed document that functions as symbolic context. Implicit awakening utilizes a hypernetwork to generate adapters based on the question and synthetic document, which are inserted into LLMs to serve as parameter context. Experimental results on three datasets demonstrate that AAG exhibits significant advantages in both open-domain and closed-book settings, as well as in out-of-distribution generalization. Our code will be available at \url{https://github.com/Xnhyacinth/IAG}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_15268 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Awakening Augmented Generation: Learning to Awaken Internal Knowledge of Large Language Models for Question Answering Liao, Huanxuan He, Shizhu Xu, Yao Zhang, Yuanzhe Liu, Kang Liu, Shengping Zhao, Jun Computation and Language Retrieval-Augmented-Generation and Generation-Augmented-Generation have been proposed to enhance the knowledge required for question answering with Large Language Models (LLMs) by leveraging richer context. However, the former relies on external resources, and both require incorporating explicit documents into the context, which increases execution costs and susceptibility to noise data during inference. Recent works indicate that LLMs model rich knowledge, but it is often not effectively activated and awakened. Inspired by this, we propose a novel knowledge-augmented framework, $\textbf{Awakening-Augmented-Generation}$ (AAG), which mimics the human ability to answer questions using only thinking and recalling to compensate for knowledge gaps, thereby awaking relevant knowledge in LLMs without relying on external resources. AAG consists of two key components for awakening richer context. Explicit awakening fine-tunes a context generator to create a synthetic, compressed document that functions as symbolic context. Implicit awakening utilizes a hypernetwork to generate adapters based on the question and synthetic document, which are inserted into LLMs to serve as parameter context. Experimental results on three datasets demonstrate that AAG exhibits significant advantages in both open-domain and closed-book settings, as well as in out-of-distribution generalization. Our code will be available at \url{https://github.com/Xnhyacinth/IAG}. |
| title | Awakening Augmented Generation: Learning to Awaken Internal Knowledge of Large Language Models for Question Answering |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2403.15268 |