# Enhancing Semantics in Multimodal Chain of Thought via Soft Negative Sampling

arXiv: https://arxiv.org/abs/2405.09848 · HTML: https://mlsanity.com/paper/2405.09848/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 10 of its 79 references that are in the map.
- First posted on arXiv 16 May 2024.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)

## Builds on (most cited first, 10 of 10)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Improved Baselines with Momentum Contrastive Learning](https://mlsanity.com/paper/2003.04297/) (2020)
- [LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention](https://mlsanity.com/paper/2303.16199/) (2023)
- [Multimodal Chain-of-Thought Reasoning in Language Models](https://mlsanity.com/paper/2302.00923/) (2023)
- [Automatic Chain of Thought Prompting in Large Language Models](https://mlsanity.com/paper/2210.03493/) (2022)
- [Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models](https://mlsanity.com/paper/2304.09842/) (2023)
- [Large Language Models Are Reasoning Teachers](https://mlsanity.com/paper/2212.10071/) (2022)
- [Teaching Small Language Models to Reason](https://mlsanity.com/paper/2212.08410/) (2022)
- [SNCSE: Contrastive Learning for Unsupervised Sentence Embedding with Soft Negative Samples](https://mlsanity.com/paper/2201.05979/) (2022)

## Built on by (most cited first, 8 of 8)

- [Multimodal Chain-of-Thought Reasoning: A Comprehensive Survey](https://mlsanity.com/paper/2503.12605/) (2025)
- [SceneAlign: Aligning Multimodal Reasoning to Scene Graphs in Complex Visual Scenes](https://mlsanity.com/paper/2601.05600/) (2026)
- [Beyond Spurious Signals: Debiasing Multimodal Large Language Models via Counterfactual Inference and Adaptive Expert Routing](https://mlsanity.com/paper/2509.15361/) (2025)
- [Towards Multimodal Metaphor Understanding: A Chinese Dataset and Model for Metaphor Mapping Identification](https://mlsanity.com/paper/2501.02434/) (2025)
- [EVL-MCoT: Enhanced Vision-Language Multi-CoT for Harmful Meme Detection](https://mlsanity.com/paper/2607.22016/) (2026)
- [Rationale-Enhanced Decoding for Multi-modal Chain-of-Thought](https://mlsanity.com/paper/2507.07685/) (2025)
- [Visual Saliency Steering Distillation for Multimodal Chain-of-Thought Reasoning](https://mlsanity.com/paper/2607.22013/) (2026)
- [DICA: Dual-Indicator Guided Contrastive Alignment in Multimodal Large Language Models](https://mlsanity.com/paper/2607.23944/) (2026)

## Abstract (arXiv)

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.