Think&Cite: Improving Attributed Text Generation with Self-Guided Tree Search and Progress Reward Modeling
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_ | 1866916801689944064 |
|---|---|
| author | Li, Junyi Ng, Hwee Tou |
| author_facet | Li, Junyi Ng, Hwee Tou |
| contents | Despite their outstanding capabilities, large language models (LLMs) are prone to hallucination and producing factually incorrect information. This challenge has spurred efforts in attributed text generation, which prompts LLMs to generate content with supporting evidence. In this paper, we propose a novel framework, called Think&Cite, and formulate attributed text generation as a multi-step reasoning problem integrated with search. Specifically, we propose Self-Guided Monte Carlo Tree Search (SG-MCTS), which capitalizes on the self-reflection capability of LLMs to reason about the intermediate states of MCTS for guiding the tree expansion process. To provide reliable and comprehensive feedback, we introduce Progress Reward Modeling to measure the progress of tree search from the root to the current state from two aspects, i.e., generation and attribution progress. We conduct extensive experiments on three datasets and the results show that our approach significantly outperforms baseline approaches. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_14860 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Think&Cite: Improving Attributed Text Generation with Self-Guided Tree Search and Progress Reward Modeling Li, Junyi Ng, Hwee Tou Computation and Language Despite their outstanding capabilities, large language models (LLMs) are prone to hallucination and producing factually incorrect information. This challenge has spurred efforts in attributed text generation, which prompts LLMs to generate content with supporting evidence. In this paper, we propose a novel framework, called Think&Cite, and formulate attributed text generation as a multi-step reasoning problem integrated with search. Specifically, we propose Self-Guided Monte Carlo Tree Search (SG-MCTS), which capitalizes on the self-reflection capability of LLMs to reason about the intermediate states of MCTS for guiding the tree expansion process. To provide reliable and comprehensive feedback, we introduce Progress Reward Modeling to measure the progress of tree search from the root to the current state from two aspects, i.e., generation and attribution progress. We conduct extensive experiments on three datasets and the results show that our approach significantly outperforms baseline approaches. |
| title | Think&Cite: Improving Attributed Text Generation with Self-Guided Tree Search and Progress Reward Modeling |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2412.14860 |