# Co-Annotator: Expert-Distilled ViT and VLM for Visual and Documentation Guidance in Age-Related Macular Degeneration

arXiv: https://arxiv.org/abs/2608.30352 · HTML: https://mlsanity.com/paper/2608.30352/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 104 references that are in the map.
- First posted on arXiv 31 Aug 2026.

## Builds on (most cited first, 16 of 16)

- [Does the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performance](https://mlsanity.com/paper/2006.14779/) (2020)
- [Manipulating and Measuring Model Interpretability](https://mlsanity.com/paper/1802.07810/) (2018)
- [Questioning the AI: Informing Design Practices for Explainable AI User Experiences](https://mlsanity.com/paper/2001.02478/) (2020)
- [CoAuthor: Designing a Human-AI Collaborative Writing Dataset for Exploring Language Model Capabilities](https://mlsanity.com/paper/2201.06796/) (2022)
- [On Human Predictions with Explanations and Predictions of Machine Learning Models: A Case Study on Deception Detection](https://mlsanity.com/paper/1811.07901/) (2018)
- [Proxy Tasks and Subjective Measures Can Be Misleading in Evaluating Explainable AI Systems](https://mlsanity.com/paper/2001.08298/) (2020)
- [Human-Centered Tools for Coping with Imperfect Algorithms during Medical Decision-Making](https://mlsanity.com/paper/1902.02960/) (2019)
- [Appropriate Reliance on AI Advice: Conceptualization and the Effect of Explanations](https://mlsanity.com/paper/2302.02187/) (2023)
- [Beyond Expertise and Roles: A Framework to Characterize the Stakeholders of Interpretable Machine Learning and their Needs](https://mlsanity.com/paper/2101.09824/) (2021)
- [CheXplain: Enabling Physicians to Explore and UnderstandData-Driven, AI-Enabled Medical Imaging Analysis](https://mlsanity.com/paper/2001.05149/) (2020)
- [Designing AI for Trust and Collaboration in Time-Constrained Medical Decisions: A Sociotechnical Lens](https://mlsanity.com/paper/2102.00593/) (2021)
- [Who Goes First? Influences of Human-AI Workflow on Decision Making in Clinical Imaging](https://mlsanity.com/paper/2205.09696/) (2022)
- [Rethinking Human-AI Collaboration in Complex Medical Decision Making: A Case Study in Sepsis Diagnosis](https://mlsanity.com/paper/2309.12368/) (2023)
- [Eye-gaze Guided Multi-modal Alignment for Medical Representation Learning](https://mlsanity.com/paper/2403.12416/) (2024)
- [MedKnowts: Unified Documentation and Information Retrieval for Electronic Health Records](https://mlsanity.com/paper/2109.11451/) (2021)
- [Ontology-Constrained Generation of Domain-Specific Clinical Summaries](https://mlsanity.com/paper/2411.15666/) (2024)

## Abstract (arXiv)

Clinical AI often optimizes predictive performance without engaging how clinicians decide where to look and what to write. We present Co-Annotator, which distills expert gaze and dictation into two guidance components: a gaze-aligned Vision Transformer producing fixation-aligned areas of interest (AOIs), and an ontology-bounded vision-language model (VLM) that pre-fills editable biomarker summaries for retinal optical coherence tomography (OCT). We first collect expert gaze and dictations (US1) to train the models, significantly improving diagnostic accuracy and biomarker generation. We then deploy the system with ophthalmology residents: a controlled resident study (US2) confirmed each modality is safe and independently beneficial, with AOI guidance producing lasting perceptual efficiency gains through post-guidance carryover and VLM guidance more than doubling biomarker documentation breadth. In a combined deployment across two academic institutions (US3), providing both modalities simultaneously produced efficiency gains that substantially exceeded either modality alone: correct diagnoses per minute increased by 40% and comment editing time fell by 67%, without compromising diagnostic accuracy. Notably, neither modality improved efficiency during guidance in US2, which makes the in-guidance efficiency gain under combined guidance in US3 the more striking result. Expert-distilled multimodal guidance can remove two distinct clinical workflow bottlenecks at once (visual search overhead and documentation burden) without compromising the diagnostic accuracy clinicians already achieve.