# When to Trust the Answer: Question-Aligned Semantic Nearest Neighbor Entropy for Safer Surgical VQA

arXiv: https://arxiv.org/abs/2511.01458 · HTML: https://mlsanity.com/paper/2511.01458/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 23 references that are in the map.
- First posted on arXiv 3 Nov 2025.

## Builds on (most cited first, 21 of 21)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://mlsanity.com/paper/2006.03654/) (2020)
- [M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation](https://mlsanity.com/paper/2402.03216/) (2024)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [MedGemma Technical Report](https://mlsanity.com/paper/2507.05201/) (2025)
- [A Survey on Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2402.00253/) (2024)
- [Don't Just Assume; Look and Answer: Overcoming Priors for Visual Question Answering](https://mlsanity.com/paper/1712.00377/) (2017)
- [Woodpecker: Hallucination Correction for Multimodal Large Language Models](https://mlsanity.com/paper/2310.16045/) (2023)
- [Multi-Modal Hallucination Control by Visual Information Grounding](https://mlsanity.com/paper/2403.14003/) (2024)
- [Understanding Adversarial Attacks on Deep Learning Based Medical Image Analysis Systems](https://mlsanity.com/paper/1907.10456/) (2019)
- [A Survey on Uncertainty Quantification of Large Language Models: Taxonomy, Open Research Challenges, and Future Directions](https://mlsanity.com/paper/2412.05563/) (2024)
- [VL-Uncertainty: Detecting Hallucination in Large Vision-Language Model via Uncertainty Estimation](https://mlsanity.com/paper/2411.11919/) (2024)
- [Surgical-VQA: Visual Question Answering in Surgical Scenes using Transformer](https://mlsanity.com/paper/2206.11053/) (2022)
- [SurgicalGPT: End-to-End Language-Vision GPT for Visual Question Answering in Surgery](https://mlsanity.com/paper/2304.09974/) (2023)
- [Beyond Semantic Entropy: Boosting LLM Uncertainty Quantification with Pairwise Semantic Similarity](https://mlsanity.com/paper/2506.00245/) (2025)
- [PitVQA: Image-grounded Text Embedding LLM for Visual Question Answering in Pituitary Surgery](https://mlsanity.com/paper/2405.13949/) (2024)
- [Reference-free Hallucination Detection for Large Vision-Language Models](https://mlsanity.com/paper/2408.05767/) (2024)
- [A comprehensive taxonomy of hallucinations in Large Language Models](https://mlsanity.com/paper/2508.01781/) (2025)
- [DHCP: Detecting Hallucinations by Cross-modal Attention Pattern in Large Vision-Language Models](https://mlsanity.com/paper/2411.18659/) (2024)
- [PitVQA++: Vector Matrix-Low-Rank Adaptation for Open-Ended Visual Question Answering in Pituitary Surgery](https://mlsanity.com/paper/2502.14149/) (2025)

## Built on by (most cited first, 3 of 3)

- [TemporalDoRA: Temporal PEFT for Robust Surgical Video Question Answering](https://mlsanity.com/paper/2603.09696/) (2026)
- [Training LLMs with Reinforcement Learning over Digital Twin Representations for Reasoning-Intensive Surgical VideoQA](https://mlsanity.com/paper/2606.17279/) (2026)
- [PROVE: Proof-guided Regime-aware Operator Verification for Hallucination Detection in Medical Visual Question Answering](https://mlsanity.com/paper/2609.29604/) (2026)

## Abstract (arXiv)

Safety and reliability are critical for deploying visual question answering (VQA) systems in surgery, where incorrect or ambiguous responses can cause patient harm. A key limitation of existing uncertainty estimation methods, such as Semantic Nearest Neighbor Entropy (SNNE), is that they do not explicitly account for the conditioning question. As a result, they may assign high confidence to answers that are semantically consistent yet misaligned with the clinical question, especially under variation in question phrasing. We propose Question-Aligned Semantic Nearest Neighbor Entropy (QA-SNNE), a black-box uncertainty estimator that incorporates question-answer alignment into semantic entropy through bilateral gating. QA-SNNE measures uncertainty by weighting pairwise semantic similarities among sampled answers according to their relevance to the question, using embedding-based, entailment-based, or cross-encoder alignment strategies. To assess robustness to language variation, we construct an out-of-template rephrased version of a benchmark surgical VQA dataset, where only the question wording is modified while images and ground-truth answers remain unchanged. We evaluate QA-SNNE on five VQA models across two benchmark surgical VQA datasets in both zero-shot and parameter-efficient fine-tuned (PEFT) settings, including out-of-template questions. QA-SNNE improves AUROC on EndoVis18-VQA for two of three zero-shot models in-template (e.g., +15% for Llama3.2 and +21% for Qwen2.5) and achieves up to +8% AUROC improvement under out-of-template rephrasing, with mixed results on external validation. Overall, QA-SNNE provides a practical, model-agnostic safeguard for surgical VQA by linking semantic uncertainty to question relevance.