You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

SurgHint

SurgHint is a surgical visual question answering system developed for the FRAME and SEGMENT tracks of the ORena FOCUS Challenge. This repository provides the detector checkpoints and Qwen LoRA adapters for answering questions about foreign objects in surgical images and video clips.

Code repository

Team Details

VLM_MAXXING — Orhun Utku Aydin, Frank te Nijenhuis, Dietmar Frey.

Released weights and licences

Track	Checkpoint	Licence
FRAME	Plain-DETR detector head, fine-tuned from Meta's DINOv3 COCO2017 checkpoint	DINOv3 License, unchanged
FRAME	Qwen3.5-9B LoRA adapter	CC BY-NC-SA 4.0 for the adaptation
SEGMENT	Fine-tuned RF-DETR-L checkpoint	CC BY-NC-SA 4.0 for the adaptation
SEGMENT	Qwen3.6-27B LoRA adapter	CC BY-NC-SA 4.0 for the adaptation

Sign in and request access on this model page to download the checkpoints. Qwen base weights and FRAME's frozen DINOv3 backbone are not included; please obtain them separately under their upstream licences and access terms.

SurgHint inference uses greedy decoding (do_sample=False) with thinking disabled. The inference code overrides the sampling defaults stored in generation_config.json.

Upstream Qwen and RF-DETR materials retain their Apache-2.0 licences and notices. Full licence texts and modification notices accompany the checkpoints.

Training-data sources include LapChole, HeiCo, Surgical Gauze, SISVSE, hSDB and GynSurg. These datasets are not included, and checkpoint access grants no rights to them. The checkpoint licences do not override applicable training-data licences or access agreements.

These models are intended for noncommercial research and reproducibility.

Downloads last month: -; Downloads are not tracked for this model. How to track