# Quantized Vision-Language Models for Damage Assessment: A Comparative Study of LLaVA-1.5-7B Quantization Levels

arXiv: https://arxiv.org/abs/2603.26770 · HTML: https://mlsanity.com/paper/2603.26770/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Chart understanding and anomaly detection, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 13 of its 38 references that are in the map.
- First posted on arXiv 24 Mar 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Chart understanding and anomaly detection](https://mlsanity.com/mllm/chart-understanding-and-anomaly-detection/)

## Builds on (most cited first, 13 of 13)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models](https://mlsanity.com/paper/2304.10592/) (2023)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [CogVLM: Visual Expert for Pretrained Language Models](https://mlsanity.com/paper/2311.03079/) (2023)
- [Multi-stage Bridge Inspection System: Integrating Foundation Models with Location Anonymization](https://mlsanity.com/paper/2601.17254/) (2026)
- [Adapting Methods for Domain-Specific Japanese Small LMs: Scale, Architecture, and Quantization](https://mlsanity.com/paper/2603.18037/) (2026)

## Built on by (most cited first, 2 of 2)

- [Fine-Tuning Vision-Language Models for Understanding Current Damage and Scoring Priority with Quality Guard Agent](https://mlsanity.com/paper/2605.27452/) (2026)
- [Encoding Invisible Causation for Bridge Diagnostic Agents: Triple-Guided Retrieval-Augmented Fine-Tuning with QLoRA](https://mlsanity.com/paper/2607.21680/) (2026)

## Abstract (arXiv)

Bridge infrastructure inspection is a critical but labor-intensive task requiring expert assessment of structural damage such as rebar exposure, cracking, and corrosion. This paper presents a comprehensive study of quantized Vision-Language Models (VLMs) for automated bridge damage assessment, focusing on the trade-offs between description quality, inference speed, and resource requirements. We develop an end-to-end pipeline combining LLaVA-1.5-7B for visual damage analysis, structured JSON extraction, and rule-based priority scoring. To enable deployment on consumer-grade GPUs, we conduct a systematic comparison of three quantization levels: Q4KM, Q5KM, and Q80 across 254 rebar exposure images. We introduce a 5-point quality evaluation framework assessing damage type recognition, severity classification. Our results demonstrate that Q5KM achieves the optimal balance: quality score 3.18±1.35/5.0, inference time 5.67s/image, and 0.56 quality/sec efficiency -- 8.5% higher quality than Q4KM with only 4.5% speed reduction, while matching Q80's quality with 25% faster inference. Statistical analysis reveals Q5KM exhibits the weakest text-quality correlation (-0.148), indicating consistent performance regardless of description length.