SurgPace: Complexity-Aware Progressive Distillation for Surgical Multimodal Large Language Models
cs.CV
Submitted: 2026-11-02
Updated: 2026-11-02
Code: https://github.com/hezhicheng2002/SurgPace
Terminology
Sources
- Surgical-VQLA: Transformer with Gated Vision-Language Embedding for Visual Question Localized-Answering in Robotic Surgery
- Towards Cross-Tokenizer Distillation: the Universal Logit Distillation Loss for LLMs
- VS-Assistant: Versatile Surgery Assistant on the Demand of Surgeons
- SurgLLM: A Versatile Large Multimodal Model with Spatial Focus and Temporal Awareness for Surgical Video Understanding
- Efficient Knowledge Distillation via Curriculum Extraction
- Surgery-R1: Advancing Surgical-VQLA with Reasoning Multimodal Large Language Model via Reinforcement Learning
- TinyLLaVA Factory: A Modularized Codebase for Small-scale Large Multimodal Models
- Hulu-Med: A Transparent Generalist Model towards Holistic Medical Vision-Language Understanding
- LLaVA-Surg: Towards Multimodal Surgical Assistant via Structured Surgical Video Learning
- Unlock the Power: Competitive Distillation for Multi-Modal Large Language Models
- MaTVLM: Hybrid Mamba-Transformer for Efficient Vision-Language Modeling
- Decoupled Weight Decay Regularization
- SUREON: A Benchmark and Vision-Language Model for Surgical Reasoning
- SurGo-R1: Benchmarking and Modeling Contextual Reasoning for Operative Zone in Surgical Video
- Systematic Evaluation of Large Vision-Language Models for Surgical Artificial Intelligence
- Exploring Curriculum Learning for Vision-Language Tasks: A Study on Small-Scale Multimodal Training
- GP-VLS: A general-purpose vision language model for surgery
- LLaVA-MoD: Making LLaVA Tiny via MoE Knowledge Distillation
- Surgical-LVLM: Learning to Adapt Large Vision-Language Model for Grounded Visual Question Answering in Robotic Surgery
- SurgVidLM: Towards Multi-grained Surgical Video Understanding with Large Language Model
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs