Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.18320 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915211442651136 |
|---|---|
| author | Jing, Dong Fei, Nanyi Lu, Zhiwu |
| author_facet | Jing, Dong Fei, Nanyi Lu, Zhiwu |
| contents | In the realm of Large Multi-modal Models (LMMs), the instruction quality during the visual instruction tuning stage significantly influences the performance of modality alignment. In this paper, we assess the instruction quality from a unique perspective termed \textbf{Writing Manner}, which encompasses the selection of vocabulary, grammar and sentence structure to convey specific semantics. We argue that there exists a substantial writing manner gap between the visual instructions and the base Large Language Models (LLMs) within LMMs. This gap forces the pre-trained base LLMs to deviate from their original writing styles, leading to capability degradation of both base LLMs and LMMs. To bridge the writing manner gap while preserving the original semantics, we propose directly leveraging the base LLM to align the writing manner of soft-format visual instructions with that of the base LLM itself, resulting in novel LLM-aligned instructions. The manual writing manner evaluation results demonstrate that our approach successfully minimizes the writing manner gap. By utilizing LLM-aligned instructions, the baseline models LLaVA-7B and QwenVL demonstrate enhanced resistance to hallucinations and non-trivial comprehensive improvements across all $15$ visual and language benchmarks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_18320 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Bridging Writing Manner Gap in Visual Instruction Tuning by Creating LLM-aligned Instructions Jing, Dong Fei, Nanyi Lu, Zhiwu Artificial Intelligence Computation and Language In the realm of Large Multi-modal Models (LMMs), the instruction quality during the visual instruction tuning stage significantly influences the performance of modality alignment. In this paper, we assess the instruction quality from a unique perspective termed \textbf{Writing Manner}, which encompasses the selection of vocabulary, grammar and sentence structure to convey specific semantics. We argue that there exists a substantial writing manner gap between the visual instructions and the base Large Language Models (LLMs) within LMMs. This gap forces the pre-trained base LLMs to deviate from their original writing styles, leading to capability degradation of both base LLMs and LMMs. To bridge the writing manner gap while preserving the original semantics, we propose directly leveraging the base LLM to align the writing manner of soft-format visual instructions with that of the base LLM itself, resulting in novel LLM-aligned instructions. The manual writing manner evaluation results demonstrate that our approach successfully minimizes the writing manner gap. By utilizing LLM-aligned instructions, the baseline models LLaVA-7B and QwenVL demonstrate enhanced resistance to hallucinations and non-trivial comprehensive improvements across all $15$ visual and language benchmarks. |
| title | Bridging Writing Manner Gap in Visual Instruction Tuning by Creating LLM-aligned Instructions |
| topic | Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2503.18320 |