Self-consistent Reasoning For Solving Math Word Problems

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xiong, Jing, Wan, Zhongwei, Hu, Xiping, Yang, Min, Li, Chengming
Format: Preprint
Published: 2022
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929246116511744
author Xiong, Jing
Wan, Zhongwei
Hu, Xiping
Yang, Min
Li, Chengming
author_facet Xiong, Jing
Wan, Zhongwei
Hu, Xiping
Yang, Min
Li, Chengming
contents Math word problems (MWPs) is a task that automatically derives solution expression from a giving math problems in text. The previous studies suffer from spurious correlations between input text and output expression. To mitigate this issue, we propose a self-consistent reasoning framework called SCR, which attempts to adopt a pruning strategy to correct the output distribution shift so as to implicitly fix those spurious correlative samples. Specifically, we firstly obtain a sub-network by pruning a roberta2tree model, for the sake to use the gap on output distribution between the original roberta2tree model and the pruned sub-network to expose spurious correlative samples. Then, we calibrate the output distribution shift by applying symmetric Kullback-Leibler divergence to alleviate spurious correlations. In addition, SCR generates equivalent expressions, thereby, capturing the original text's logic rather than relying on hints from original text. Extensive experiments on two large-scale benchmarks demonstrate that our model substantially outperforms the strong baseline methods.
format Preprint
id arxiv_https___arxiv_org_abs_2210_15373
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Self-consistent Reasoning For Solving Math Word Problems
Xiong, Jing
Wan, Zhongwei
Hu, Xiping
Yang, Min
Li, Chengming
Computation and Language
Math word problems (MWPs) is a task that automatically derives solution expression from a giving math problems in text. The previous studies suffer from spurious correlations between input text and output expression. To mitigate this issue, we propose a self-consistent reasoning framework called SCR, which attempts to adopt a pruning strategy to correct the output distribution shift so as to implicitly fix those spurious correlative samples. Specifically, we firstly obtain a sub-network by pruning a roberta2tree model, for the sake to use the gap on output distribution between the original roberta2tree model and the pruned sub-network to expose spurious correlative samples. Then, we calibrate the output distribution shift by applying symmetric Kullback-Leibler divergence to alleviate spurious correlations. In addition, SCR generates equivalent expressions, thereby, capturing the original text's logic rather than relying on hints from original text. Extensive experiments on two large-scale benchmarks demonstrate that our model substantially outperforms the strong baseline methods.
title Self-consistent Reasoning For Solving Math Word Problems
topic Computation and Language
url https://arxiv.org/abs/2210.15373