| MEDITRON-70B: Scaling Medical Pretraining for Large Language Models | ✓ Link | 81.6 | Meditron-70B (CoT + SC) | 2023-11-27 |
| BioGPT: Generative Pre-trained Transformer for Biomedical Text Generation and Mining | ✓ Link | 81.0 | BioGPT-Large(1.5B) | 2022-10-19 |
| AdaSR: Adaptive Streaming Reasoning with Hierarchical Relative Policy Optimization | ✓ Link | 80.7 | AdaSR-HRPO (Qwen3-4B) | 2026-06-12 |
| RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMs | | 79.8 | RankRAG-llama3-70B (Zero-Shot) | 2024-07-02 |
| ClinFusion: A Vision-Centric Multimodal LLM System for Holistic Medical Understanding | ✓ Link | 79.7 | ClinFusion-32B + Agentic tools | 2026-07-27 |
| Towards Expert-Level Medical Question Answering with Large Language Models | ✓ Link | 79.2 | Med-PaLM 2 (5-shot) | 2023-05-16 |
| Large Language Models Encode Clinical Knowledge | ✓ Link | 79 | Flan-PaLM (540B, Few-shot) | 2022-12-26 |
| MetaRubric: Learning to Reward for Rubric-Based Reinforcement Learning | ✓ Link | 78.4 | MetaRubric (Qwen3-4B) | 2026-10-02 |
| Can large language models reason about medical questions? | ✓ Link | 78.2 | Codex 5-shot CoT | 2022-07-17 |
| BioGPT: Generative Pre-trained Transformer for Biomedical Text Generation and Mining | ✓ Link | 78.2 | BioGPT(345M) | 2022-10-19 |
| PubMedQA: A Dataset for Biomedical Research Question Answering | ✓ Link | 78.0 | Human Performance (single annotator) | 2019-09-13 |
| MetaGen Blended RAG: Higher Accuracy for Domain-Specific Q&A Without Fine-Tuning | ✓ Link | 77.9 | MetaGen Blended RAG (zero-shot) | 2025-05-23 |
| Meissa: Multi-modal Medical Agentic Intelligence | ✓ Link | 77.9 | Meissa (4B) | 2026-03-09 |
| Galactica: A Large Language Model for Science | ✓ Link | 77.6 | GAL 120B (zero-shot) | 2022-11-16 |
| Large Language Models Encode Clinical Knowledge | ✓ Link | 77.2 | Flan-PaLM (62B, Few-shot) | 2022-12-26 |
| MediSwift: Efficient Sparse Pre-trained Biomedical Language Models | | 76.8 | MediSwift-XL | 2024-03-01 |
| Evaluation of large language model performance on the Biomedical Language Understanding and Reasoning Benchmark | | 76.80 | Flan-T5-XXL | 2024-05-17 |
| BioMedGPT: Open Multimodal Generative Pre-trained Transformer for BioMedicine | ✓ Link | 76.1 | BioMedGPT-10B | 2023-08-18 |
| The Claude 3 Model Family: Opus, Sonnet, Haiku | | 75.8 | Claude 3 Opus (5-shot) | 2024-03-04 |
| Medic-AD: Towards Medical Vision-Language Model's Clinical Intelligence | ✓ Link | 75.6 | MEDIC-AD | 2026-03-28 |
| Large Language Models Encode Clinical Knowledge | ✓ Link | 75.2 | Flan-PaLM (540B, SC) | 2022-12-26 |
| Towards Expert-Level Medical Question Answering with Large Language Models | ✓ Link | 75.0 | Med-PaLM 2 (ER) | 2023-05-16 |
| The Claude 3 Model Family: Opus, Sonnet, Haiku | | 74.9 | Claude 3 Opus (zero-shot) | 2024-03-04 |
| Towards Expert-Level Medical Question Answering with Large Language Models | ✓ Link | 74.0 | Med-PaLM 2 (CoT + SC) | 2023-05-16 |
| Galactica: A Large Language Model for Science | ✓ Link | 73.6 | BLOOM (zero-shot) | 2022-11-16 |
| The CoT Collection: Improving Zero-shot and Few-shot Learning of Language Models via Chain-of-Thought Fine-Tuning | ✓ Link | 73.42 | CoT-T5-11B (1024 Shot) | 2023-05-23 |
| LinkBERT: Pretraining Language Models with Document Links | ✓ Link | 72.2 | BioLinkBERT (large) | 2022-03-29 |
| LinkBERT: Pretraining Language Models with Document Links | ✓ Link | 70.2 | BioLinkBERT (base) | 2022-03-29 |
| Galactica: A Large Language Model for Science | ✓ Link | 70.2 | OPT (zero-shot) | 2022-11-16 |
| Large Language Models Encode Clinical Knowledge | ✓ Link | 67.6 | Flan-PaLM (8B, Few-shot) | 2022-12-26 |
| MedGemma 1.5 Technical Report | ✓ Link | 67.6 | MedGemma 1.5 4B | 2026-04-06 |
| QuCo-RAG: Quantifying Uncertainty from the Pre-training Corpus for Dynamic Retrieval-Augmented Generation | ✓ Link | 66.4 | QuCo-RAG (OLMo-2-7B) | 2025-12-22 |
| BioELECTRA:Pretrained Biomedical text Encoder using Discriminators | ✓ Link | 64.2 | BioELECTRA uncased | 2021-06-11 |
| Eliciting Medical Reasoning with Knowledge-enhanced Data Synthesis: A Semi-Supervised Reinforcement Learning Approach | ✓ Link | 58.75 | Llama-3.1-8B-Instruct + MedSSR | 2026-04-13 |
| Large Language Models Encode Clinical Knowledge | ✓ Link | 57.8 | PaLM (62B, Few-shot) | 2022-12-26 |
| GPT-Neo: Large Scale Autoregressive Language Modeling with Mesh-Tensorflow | ✓ Link | 57.54 | GPT-Neo 2.7B | 2021-03-21 |
| Domain-Specific Language Model Pretraining for Biomedical Natural Language Processing | ✓ Link | 55.84 | PubMedBERT uncased | 2020-07-31 |
| Large Language Models Encode Clinical Knowledge | ✓ Link | 55 | PaLM (540B, Few-shot) | 2022-12-26 |
| Large Language Models Encode Clinical Knowledge | ✓ Link | 34 | PaLM (8B, Few-shot) | 2022-12-26 |