# Can Vision-Language Models Assess Proxemic Risk from Egocentric Robot Images?

arXiv: https://arxiv.org/abs/2608.12515 · HTML: https://mlsanity.com/paper/2608.12515/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Embodied and 3D vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 14 of its 21 references that are in the map.
- First posted on arXiv 12 Aug 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Embodied and 3D vision-language models](https://mlsanity.com/mllm/embodied-and-3d-vision-language-models/)

## Builds on (most cited first, 14 of 14)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites](https://mlsanity.com/paper/2404.16821/) (2024)
- [Show and Tell: A Neural Image Caption Generator](https://mlsanity.com/paper/1411.4555/) (2014)
- [Instruction Tuning for Large Language Models: A Survey](https://mlsanity.com/paper/2308.10792/) (2023)
- [How Far are VLMs from Visual Spatial Intelligence? A Benchmark-Driven Perspective](https://mlsanity.com/paper/2509.18905/) (2025)
- [Chain of Thought Prompt Tuning in Vision Language Models](https://mlsanity.com/paper/2304.07919/) (2023)
- [The Format Tax](https://mlsanity.com/paper/2604.03616/) (2026)

## Abstract (arXiv)

Assessing proxemic danger from a robot's egocentric perspective is critical for safe embodied navigation in human environments and requires both visual and contextual reasoning. We evaluate three opensource vision-language models (VLMs) (textitInternVL, textitQwen-VL, and textitSmolVLM) on the classification of egocentric robot images into four danger levels, comparing three prompting strategies and two rounds of QLoRA fine-tuning against a stratified random baseline. Without fine-tuning, all models perform near the baseline, while fine-tuning yields only modest overall improvements. However, textitQwen-VL with an advanced prompt achieves substantially higher recall for high-danger cases than the other models. An analysis of person localization further shows that correct danger classification does not correspond to better spatial grounding, indicating that a model may produce a useful safety label without attending to the relevant region of the scene. These results show that current VLMs remain limited in fine-grained proxemic reasoning and spatial grounding, although targeted prompting and fine-tuning can improve high-danger detection in selected models.