# MedOmni-45deg: A Safety-Performance Benchmark for Reasoning-Oriented LLMs in Medicine

arXiv: https://arxiv.org/abs/2508.16213 · HTML: https://mlsanity.com/paper/2508.16213/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning.
- Builds on 26 of its 50 references that are in the map.
- First posted on arXiv 22 Aug 2025.

## Placed in

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

## Builds on (most cited first, 26 of 26)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams](https://mlsanity.com/paper/2009.13081/) (2020)
- [PubMedQA: A Dataset for Biomedical Research Question Answering](https://mlsanity.com/paper/1909.06146/) (2019)
- [Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement](https://mlsanity.com/paper/2409.12122/) (2024)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?](https://mlsanity.com/paper/2004.03685/) (2020)
- [Reasoning Models Don't Always Say What They Think](https://mlsanity.com/paper/2505.05410/) (2025)
- [TrustLLM: Trustworthiness in Large Language Models](https://mlsanity.com/paper/2401.05561/) (2024)
- [HuatuoGPT-o1, Towards Medical Complex Reasoning with LLMs](https://mlsanity.com/paper/2412.18925/) (2024)
- [Helping or Herding? Reward Model Ensembles Mitigate but do not Eliminate Reward Hacking](https://mlsanity.com/paper/2312.09244/) (2023)
- [Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs](https://mlsanity.com/paper/2502.17424/) (2025)
- [SycEval: Evaluating LLM Sycophancy](https://mlsanity.com/paper/2502.08177/) (2025)
- [Beyond Functional Correctness: Exploring Hallucinations in LLM-Generated Code](https://mlsanity.com/paper/2404.00971/) (2024)
- [Thought Crime: Backdoors and Emergent Misalignment in Reasoning Models](https://mlsanity.com/paper/2506.13206/) (2025)
- [A Survey of Useful LLM Evaluation](https://mlsanity.com/paper/2406.00936/) (2024)
- [MEDIC: Comprehensive Evaluation of Leading Indicators for LLM Safety and Utility in Clinical Applications](https://mlsanity.com/paper/2409.07314/) (2024)
- [Evaluating LLM -- Generated Multimodal Diagnosis from Medical Images and Symptom Analysis](https://mlsanity.com/paper/2402.01730/) (2024)
- [A Comprehensive Survey on Evaluating Large Language Model Applications in the Medical Industry](https://mlsanity.com/paper/2404.15777/) (2024)
- [Improving Large Language Model (LLM) fidelity through context-aware grounding: A systematic approach to reliability and veracity](https://mlsanity.com/paper/2408.04023/) (2024)
- [Personalizing Prostate Cancer Education for Patients Using an EHR-Integrated LLM Agent](https://mlsanity.com/paper/2409.19100/) (2024)
- [Research-Oriented Human-Centric Evaluation for Foundation Models](https://mlsanity.com/paper/2506.01793/) (2025)

## Built on by (most cited first, 6 of 6)

- [ReasonBENCH: Benchmarking the (In)Stability of LLM Reasoning](https://mlsanity.com/paper/2512.07795/) (2025)
- [Faithful or Just Plausible? Evaluating the Faithfulness of Closed-Source LLMs in Medical Reasoning](https://mlsanity.com/paper/2603.13988/) (2026)
- [ROFI: A Deep Learning-Based Ophthalmic Sign-Preserving and Reversible Patient Face Anonymizer](https://mlsanity.com/paper/2510.11073/) (2025)
- [PriceSeer: Evaluating Large Language Models in Real-Time Stock Prediction](https://mlsanity.com/paper/2601.06088/) (2025)
- [A3: Towards Advertising Aesthetic Assessment](https://mlsanity.com/paper/2603.24037/) (2026)
- [Untangling the Mechanisms of Misleading Context in Medical Question Answering](https://mlsanity.com/paper/2609.02754/) (2026)

## Abstract (arXiv)

With the increasing use of large language models (LLMs) in medical decision-support, it is essential to evaluate not only their final answers but also the reliability of their reasoning. Two key risks are Chain-of-Thought (CoT) faithfulness -- whether reasoning aligns with responses and medical facts -- and sycophancy, where models follow misleading cues over correctness. Existing benchmarks often collapse such vulnerabilities into single accuracy scores. To address this, we introduce MedOmni-45 Degrees, a benchmark and workflow designed to quantify safety-performance trade-offs under manipulative hint conditions. It contains 1,804 reasoning-focused medical questions across six specialties and three task types, including 500 from MedMCQA. Each question is paired with seven manipulative hint types and a no-hint baseline, producing about 27K inputs. We evaluate seven LLMs spanning open- vs. closed-source, general-purpose vs. medical, and base vs. reasoning-enhanced models, totaling over 189K inferences. Three metrics -- Accuracy, CoT-Faithfulness, and Anti-Sycophancy -- are combined into a composite score visualized with a 45 Degrees plot. Results show a consistent safety-performance trade-off, with no model surpassing the diagonal. The open-source QwQ-32B performs closest (43.81 Degrees), balancing safety and accuracy but not leading in both. MedOmni-45 Degrees thus provides a focused benchmark for exposing reasoning vulnerabilities in medical LLMs and guiding safer model development.