# Can DeepSeek Reason Like a Surgeon? An Empirical Evaluation for Vision-Language Understanding in Robotic-Assisted Surgery

arXiv: https://arxiv.org/abs/2503.23130 · HTML: https://mlsanity.com/paper/2503.23130/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning.
- Builds on 33 of its 43 references that are in the map.
- First posted on arXiv 29 Mar 2025.

## Placed in

- LLM reasoning / [Clinical and biomedical applications](https://mlsanity.com/lr/clinical-and-biomedical-applications/)

## Builds on (most cited first, 30 of 33)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling](https://mlsanity.com/paper/2501.17811/) (2025)
- [DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding](https://mlsanity.com/paper/2412.10302/) (2024)
- [Making LLaMA SEE and Draw with SEED Tokenizer](https://mlsanity.com/paper/2310.01218/) (2023)
- [MedVLM-R1: Incentivizing Medical Reasoning Capability of Vision-Language Models (VLMs) via Reinforcement Learning](https://mlsanity.com/paper/2502.19634/) (2025)
- [Med-R1: Reinforcement Learning for Generalizable Medical Reasoning in Vision-Language Models](https://mlsanity.com/paper/2503.13939/) (2025)
- [2018 Robotic Scene Segmentation Challenge](https://mlsanity.com/paper/2001.11190/) (2020)
- [Learning Multi-modal Representations by Watching Hundreds of Surgical Video Lectures](https://mlsanity.com/paper/2307.15220/) (2023)
- [Surgical-VQA: Visual Question Answering in Surgical Scenes using Transformer](https://mlsanity.com/paper/2206.11053/) (2022)
- [SAR-RARP50: Segmentation of surgical instrumentation and Action Recognition on Robot-Assisted Radical Prostatectomy Challenge](https://mlsanity.com/paper/2401.00496/) (2023)
- [SurgicalGPT: End-to-End Language-Vision GPT for Visual Question Answering in Surgery](https://mlsanity.com/paper/2304.09974/) (2023)
- [Surgical-VQLA: Transformer with Gated Vision-Language Embedding for Visual Question Localized-Answering in Robotic Surgery](https://mlsanity.com/paper/2305.11692/) (2023)
- [A Transformer-based representation-learning model with unified processing of multimodal input for clinical diagnostics](https://mlsanity.com/paper/2306.00864/) (2023)
- [Endo-4DGS: Endoscopic Monocular Scene Reconstruction with 4D Gaussian Splatting](https://mlsanity.com/paper/2401.16416/) (2024)
- [Advancing Surgical VQA with Scene Graph Knowledge](https://mlsanity.com/paper/2312.10251/) (2023)
- [Surgical-LVLM: Learning to Adapt Large Vision-Language Model for Grounded Visual Question Answering in Robotic Surgery](https://mlsanity.com/paper/2405.10948/) (2024)
- [EndoChat: Grounded Multimodal Large Language Model for Endoscopic Surgery](https://mlsanity.com/paper/2501.11347/) (2025)
- [LLaVA-Surg: Towards Multimodal Surgical Assistant via Structured Surgical Video Learning](https://mlsanity.com/paper/2408.07981/) (2024)
- [CAT-ViL: Co-Attention Gated Vision-Language Embedding for Visual Question Localized-Answering in Robotic Surgery](https://mlsanity.com/paper/2307.05182/) (2023)
- [Surgical-VQLA++: Adversarial Contrastive Learning for Calibrated Robust Visual Question-Localized Answering in Robotic Surgery](https://mlsanity.com/paper/2408.04958/) (2024)
- [Revisiting Distillation for Continual Learning on Visual Question Localized-Answering in Robotic Surgery](https://mlsanity.com/paper/2307.12045/) (2023)
- [CoPESD: A Multi-Level Surgical Motion Dataset for Training Large Vision-Language Models to Co-Pilot Endoscopic Submucosal Dissection](https://mlsanity.com/paper/2410.07540/) (2024)
- [Dynamic Interactive Relation Capturing via Scene Graph Learning for Robotic Surgical Report Generation](https://mlsanity.com/paper/2306.02651/) (2023)
- [ASI-Seg: Audio-Driven Surgical Instrument Segmentation with Surgeon Intention Understanding](https://mlsanity.com/paper/2407.19435/) (2024)
- [VS-Assistant: Versatile Surgery Assistant on the Demand of Surgeons](https://mlsanity.com/paper/2405.08272/) (2024)

## Abstract (arXiv)

The DeepSeek models have shown exceptional performance in general scene understanding, question-answering (QA), and text generation tasks, owing to their efficient training paradigm and strong reasoning capabilities. In this study, we investigate the dialogue capabilities of the DeepSeek model in robotic surgery scenarios, focusing on tasks such as Single Phrase QA, Visual QA, and Detailed Description. The Single Phrase QA tasks further include sub-tasks such as surgical instrument recognition, action understanding, and spatial position analysis. We conduct extensive evaluations using publicly available datasets, including EndoVis18 and CholecT50, along with their corresponding dialogue data. Our empirical study shows that, compared to existing general-purpose multimodal large language models, DeepSeek-VL2 performs better on complex understanding tasks in surgical scenes. Additionally, although DeepSeek-V3 is purely a language model, we find that when image tokens are directly inputted, the model demonstrates better performance on single-sentence QA tasks. However, overall, the DeepSeek models still fall short of meeting the clinical requirements for understanding surgical scenes. Under general prompts, DeepSeek models lack the ability to effectively analyze global surgical concepts and fail to provide detailed insights into surgical scenarios. Based on our observations, we argue that the DeepSeek models are not ready for vision-language tasks in surgical contexts without fine-tuning on surgery-specific datasets.