Saved in:
Bibliographic Details
Main Authors: Zheng, Guangmin, Wang, Jin, Zhou, Xiaobing, Zhang, Xuejie
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2405.09848
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909205370241024
author Zheng, Guangmin
Wang, Jin
Zhou, Xiaobing
Zhang, Xuejie
author_facet Zheng, Guangmin
Wang, Jin
Zhou, Xiaobing
Zhang, Xuejie
contents Chain of thought (CoT) has proven useful for problems requiring complex reasoning. Many of these problems are both textual and multimodal. Given the inputs in different modalities, a model generates a rationale and then uses it to answer a question. Because of the hallucination issue, the generated soft negative rationales with high textual quality but illogical semantics do not always help improve answer accuracy. This study proposes a rationale generation method using soft negative sampling (SNSE-CoT) to mitigate hallucinations in multimodal CoT. Five methods were applied to generate soft negative samples that shared highly similar text but had different semantics from the original. Bidirectional margin loss (BML) was applied to introduce them into the traditional contrastive learning framework that involves only positive and negative samples. Extensive experiments on the ScienceQA dataset demonstrated the effectiveness of the proposed method. Code and data are released at https://github.com/zgMin/SNSE-CoT.
format Preprint
id arxiv_https___arxiv_org_abs_2405_09848
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Enhancing Semantics in Multimodal Chain of Thought via Soft Negative Sampling
Zheng, Guangmin
Wang, Jin
Zhou, Xiaobing
Zhang, Xuejie
Computation and Language
Artificial Intelligence
Chain of thought (CoT) has proven useful for problems requiring complex reasoning. Many of these problems are both textual and multimodal. Given the inputs in different modalities, a model generates a rationale and then uses it to answer a question. Because of the hallucination issue, the generated soft negative rationales with high textual quality but illogical semantics do not always help improve answer accuracy. This study proposes a rationale generation method using soft negative sampling (SNSE-CoT) to mitigate hallucinations in multimodal CoT. Five methods were applied to generate soft negative samples that shared highly similar text but had different semantics from the original. Bidirectional margin loss (BML) was applied to introduce them into the traditional contrastive learning framework that involves only positive and negative samples. Extensive experiments on the ScienceQA dataset demonstrated the effectiveness of the proposed method. Code and data are released at https://github.com/zgMin/SNSE-CoT.
title Enhancing Semantics in Multimodal Chain of Thought via Soft Negative Sampling
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2405.09848