Skip to yearly menu bar Skip to main content


Poster

TroVE: Inducing Verifiable and Efficient Toolboxes for Solving Programmatic Tasks

Zhiruo Wang · Graham Neubig · Daniel Fried

Hall C 4-9 #615
[ ] [ Project Page ] [ Paper PDF ]
[ Slides [ Poster
Thu 25 Jul 4:30 a.m. PDT — 6 a.m. PDT

Abstract:

Language models (LMs) can solve tasks such as answering questions about tables or images by writing programs. However, using primitive functions often leads to verbose and error-prone programs, and higher-level functions require expert design. To enable better solutions without human labor, we ask code LMs to curate reusable high-level functions, and use them to write solutions. We present TROVE, a training-free method of inducing a verifiable and efficient toolbox of functions, by generating via using, growing, and periodically trimming the toolbox. On 11 datasets from math, table question answering, and image reasoning tasks, TROVE consistently yields simpler solutions with higher accuracy than baselines using CodeLLaMa and previous methods using GPT, while using 79-98% smaller toolboxes. TROVE further enables 31% faster and 13% more accurate human verification than baselines. With the same pipeline, it creates diverse functions for varied tasks and datasets, providing insights into their individual characteristics.

Chat is not available.