Summary of Stoc-tot: Stochastic Tree-of-thought with Constrained Decoding For Complex Reasoning in Multi-hop Question Answering, by Zhenyu Bi et al.
STOC-TOT: Stochastic Tree-of-Thought with Constrained Decoding for Complex Reasoning in Multi-Hop Question Answering
by Zhenyu Bi, Daniel Hajialigol, Zhongkai Sun, Jie Hao, Xuan Wang
First submitted to arxiv on: 4 Jul 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel approach for multi-hop question answering (MHQA) called STOC-TOT, which leverages stochastic tree-of-thought reasoning prompting methods with constrained decoding. The authors integrate evidence retrieval with reasoning prompts, such as chain-of-thought reasoning, to enhance the performance of MHQA under zero-shot settings. They construct a tree-like reasoning structure by breaking down original questions into smaller sub-questions and prompt the model to provide probability estimations for each reasoning path. The framework outperforms other reasoning prompts on two MHQA datasets using five large language models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about making computers better at answering tricky questions that require searching multiple sources of information. Currently, big language models can do this by getting hints from humans, but this paper wants to see if we can make them do it without any help. The authors came up with a new way of giving the computer hints called STOC-TOT, which helps the model think more clearly and avoids making mistakes. |
Keywords
» Artificial intelligence » Probability » Prompt » Prompting » Question answering » Zero shot
