Yizhou Chi

EECS Department, University of California, Berkeley

Technical Report No. UCB/EECS-2024-55

May 7, 2024

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2024/EECS-2024-55.pdf

We present THOUGHTSCULPT, a general reasoning and search method for tasks with outputs that can be decomposed into components. THOUGHTSCULPT explores a search tree of potential solutions using Monte Carlo Tree Search (MCTS), building solutions one action at a time and evaluating according to any domain-specific heuristic, which in practice is often simply an LLM evaluator. Critically, our action space includes revision actions: THOUGHTSCULPT may choose to revise part of its previous output rather than continuing to build the rest of its output. Empirically, THOUGHTSCULPT outperforms state-of-the-art reasoning methods across three challenging tasks: Story Outline Improvement (up to +30% interestingness), Mini-Crosswords Solving (up to +16% word success rate), and Constrained Generation (up to +10% concept coverage).

Advisors: Daniel Klein

---

BibTeX citation:

@mastersthesis{Chi:EECS-2024-55,
    Author= {Chi, Yizhou},
    Title= {THOUGHTSCULPT: Reasoning with Intermediate Revision and Search},
    School= {EECS Department, University of California, Berkeley},
    Year= {2024},
    Month= {May},
    Url= {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2024/EECS-2024-55.html},
    Number= {UCB/EECS-2024-55},
    Abstract= {We present THOUGHTSCULPT, a general reasoning and search method for tasks with outputs that can be decomposed into components. THOUGHTSCULPT explores a search tree of potential solutions using Monte Carlo Tree Search (MCTS), building solutions one action at a time and evaluating according to any domain-specific heuristic, which in practice is often simply an LLM evaluator. Critically, our action space includes revision actions: THOUGHTSCULPT may choose to revise part of its previous output rather than continuing to build the rest of its output. Empirically, THOUGHTSCULPT outperforms state-of-the-art reasoning methods across three challenging tasks: Story Outline Improvement (up to +30% interestingness), Mini-Crosswords Solving (up to +16% word success rate), and Constrained Generation (up to +10% concept coverage).},
}

EndNote citation:

%0 Thesis
%A Chi, Yizhou 
%T THOUGHTSCULPT: Reasoning with Intermediate Revision and Search
%I EECS Department, University of California, Berkeley
%D 2024
%8 May 7
%@ UCB/EECS-2024-55
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2024/EECS-2024-55.html
%F Chi:EECS-2024-55