# Visual Saliency Steering Distillation for Multimodal Chain-of-Thought Reasoning

arXiv: https://arxiv.org/abs/2607.22013 · HTML: https://mlsanity.com/paper/2607.22013/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 17 of its 18 references that are in the map.
- First posted on arXiv 24 Jul 2026.

## Placed in

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

## Builds on (most cited first, 17 of 17)

- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [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)
- [Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models](https://mlsanity.com/paper/2303.04671/) (2023)
- [UnifiedQA: Crossing Format Boundaries With a Single QA System](https://mlsanity.com/paper/2005.00700/) (2020)
- [Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models](https://mlsanity.com/paper/2304.09842/) (2023)
- [DDCoT: Duty-Distinct Chain-of-Thought Prompting for Multimodal Reasoning in Language Models](https://mlsanity.com/paper/2310.16436/) (2023)
- [M3CoT: A Novel Benchmark for Multi-Domain Multi-step Multi-modal Chain-of-Thought](https://mlsanity.com/paper/2405.16473/) (2024)
- [Cheap and Quick: Efficient Vision-Language Instruction Tuning for Large Language Models](https://mlsanity.com/paper/2305.15023/) (2023)
- [IdealGPT: Iteratively Decomposing Vision and Language Reasoning via Large Language Models](https://mlsanity.com/paper/2305.14985/) (2023)
- [Boosting the Power of Small Multimodal Reasoning Models to Match Larger Models with Self-Consistency Training](https://mlsanity.com/paper/2311.14109/) (2023)
- [Joint Multimodal Entity-Relation Extraction Based on Edge-enhanced Graph Alignment Network and Word-pair Relation Tagging](https://mlsanity.com/paper/2211.15028/) (2022)
- [Enhancing Human-like Multi-Modal Reasoning: A New Challenging Dataset and Comprehensive Framework](https://mlsanity.com/paper/2307.12626/) (2023)
- [Enhancing Semantics in Multimodal Chain of Thought via Soft Negative Sampling](https://mlsanity.com/paper/2405.09848/) (2024)
- [Vision-aware Multimodal Prompt Tuning for Uploadable Multi-source Few-shot Domain Adaptation](https://mlsanity.com/paper/2503.06106/) (2025)

## Abstract (arXiv)

Multimodal chain-of-thought (CoT) reasoning integrates visual and textual cues through step-by-step inference. In small models with limited token budgets, modality-interaction fusion often suppresses tiny cross-modal differences. In particular, multimodal CoT often struggles when different images pair with identical text or different texts pair with an identical image, making such inputs nearly indistinguishable after fusion. This study proposes Visual Saliency Steering Distillation (VSSD). VSSD leverages the attention maps of multimodal large language models to generate perturbed images that capture task-sensitive feature directions, and then applies singular value decomposition to extract dominant steering vectors to guide inter-layer distillation. Experiments on ScienceQA and M3CoT demonstrate that VSSD improves rationale generation and answer inference. The code is available at https://github.com/BGWH123/VSSD.