MediX-R1: Open Ended Medical Reinforcement Learning
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: AI Radio. Generated commentary on the latest Artificial Intelligence papers.
Tom: We're looking at the paper: "MediX-R1: Open Ended Medical Reinforcement Learning".
Jane: Here's the gist of the paper. MediX-R1 introduces an open-ended Reinforcement Learning framework for medical multimodal large language models that enables clinically grounded, free-form answers beyond multiple-choice formats The paper presents MediX-R1, an open-ended RL framework that fine-tunes a baseline vision-language backbone with Group Based RL and a composite reward tailored for medical reasoning This framework is significant because it provides a principled approach for training medical MLLMs with open-ended RL that integrates semantic correctness with structural and modality constraints Framework Overview MediX-R1 fine-tunes a baseline multimodal backbone for open-ended medical reasoning using Group Based RL Given an image I and question q, the vision encoder produces visual tokens that are fused with text tokens and fed to the LLM policy πθ The model generates structured outputs of the form:
modality tag: free-form clinical reasoningfinal concise answer.
Tom: This design encourages the model to generate accurate and interpretable reasoning paths.
Jane: How the training signal is built The composite reward is defined as r = wfmt Rformat + wllm Rllm + wemb Rembed + wmod Rmodality.
Tom: The specific coefficients were selected through a staged procedure, fixing the format reward weight to 0.10 and allocating the remaining 0.90 mass to task-facing signals.
Paper summary: Tom: So if we boil down what MediX-R1 is about: it’s taking those complex ideas—the composite reward structure and the unified evaluation framework—and making them practical for open ended medical reasoning.
Jane: Yeah, the paper by Sahal Shaji Mullappilly et al., titled "MediX-R1: Open Ended Medical Reinforcement Learning," shows how you can use reinforcement learning to train these multimodal models to give answers that aren't just simple picks, but actual free form clinical thoughts.
Lu: It’s about using a multi faceted reward signal, incorporating semantic checks from embeddings and LLM judges alongside structural requirements like the format reward. That gives the model discipline in generating those structured outputs we see in examples.
Meng: From a deployment standpoint, I'm looking at how they handle things at scale, but what this means is that if you want to deploy an AI for medical reports, you need these kinds of signals to ensure it’s not just guessing the right words.
Tom: Exactly. The implication is that we can move away from relying too much on brittle string matching metrics for evaluating complex medical reasoning tasks and use a more principled approach instead.
Jane: It moves us toward models that are better at handling real world uncertainty, like paraphrasing a question or changing the required output format, while still maintaining high levels of semantic correctness.
Lu: The authors also commit to releasing their training inference code, configurations, checkpoints, curated datasets and RL evaluation prompt templates under a CC BY NC SA four point zero license which makes it more accessible for others to build on this.
Meng: That’s good because transparency is key when you're working in sensitive areas like medicine. They also pointed out risks around demographic and dataset bias, suggesting future work should include fairness analyses.
Tom: So the title itself suggests a lot—open ended RL. It shows that open ended reinforcement learning can be a practical tool for making medical MLLMs more reliable and trustworthy in their output.
Conclusion: Tom: So we’ve looked at how MediX-R1 uses reinforcement learning to train these medical multimodal models for open ended answers, now we need to talk about what that title actually means.
Jane: It’s about taking those complex medical image and text questions and letting the AI reason through them freely, instead of just picking one pre written answer.
Lu: The authors are doing this by combining a few different signals for their training, like a format reward, an embedding based reward, and an llm based accuracy reward.
Meng: From where I sit at the startup level it looks like they’ve built a system that tries to make the AI not just get the right answer but get it in a way that looks clinically sound.
Tom: Yeah, so if you hear "Open Ended Medical Reinforcement Learning" you have to think about how this moves away from multiple choice tests and toward actual doctor-like explanations.
Jane: Right. It’s trying to make the AI generate these structured outputs, like it first says what kind of test it’s doing, then shows its reasoning step by step, and finally gives a concise answer.
Lu: The main point is that this multi signal design stabilizes the training process and helps reduce those weird reward hacking problems where you trick the model into giving you an easy answer instead of a good one.
Tom: It seems like they’ve found a way to use structure and multiple checks to get better performance than just using one simple accuracy check.
Jane: And they used this framework on real medical data, showing it performs well in those challenging image plus text tasks compared to other models we’ve seen.
Meng: I’m still checking how stable it is when you run it on high traffic because I don't trust benchmarks that aren't running on real user traffic.
Tom: Well that brings us to the evaluation part next, where they use a unified reference based llm as judge to check if the answers are actually good.
Jane: That’s the next big thing—moving away from just counting word overlap and using an instruction tuned model to actually judge if the reasoning makes sense clinically.
Sahal Shaji Mullappilly, Mohammed Irfan Kurpath, Omair Mohamed, Mohamed Zidan, Fahad Khan, Salman Khan, Rao Anwer, Hisham Cholakkal
Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) · Jubilee Mission Medical College and Research Institute, JJM Medical College
cs.CV
Submitted: 2026-02-26
Updated: 2026-11-15
Comments: Accepted to EMNLP 2026 Findings
DOI: 10.18653/2026.findings-emnlp.2133
Code: https://github.com/abhinand5/MedEmbed
License: http://creativecommons.org/licenses/by-nc-sa/4.0/
Importance score: 89/100
The gist: MediX-R1 introduces an open-ended Reinforcement Learning framework for medical multimodal large language models that enables clinically grounded…
Key concepts
- MediX-R1 Framework
- This is an open-ended RL system that fine-tunes a multimodal backbone for medical reasoning. It uses Group Based RL and a composite reward to train the model to produce detailed, free-form clinical answers instead of just selecting options. The goal is to achieve clinically relevant, interpretable outputs.
- Composite Reward Function
- The training signal is a combination of four rewards: format, LLM accuracy (checking semantic correctness), embedding similarity (checking meaning against a reference), and modality recognition. These weights are carefully tuned to ensure the model learns not just correct answers, but also the right structure and modality.
- Reference-based LLM-as-Judge Evaluation
- Instead of brittle string matching, this evaluation uses an instruction-tuned LLM (like vLLM) to judge generated answers against a reference using templates like BASE or MIMIC. This captures complex semantic correctness and reasoning adequacy across different medical tasks.
- Reward Hacking Mitigation
- The composite reward design, including penalties for format mismatches and modality errors, helps prevent 'reward hacking' where the model cheats to maximize the score without actually improving reasoning. Specific techniques like embedding gating further stabilize optimization.
Terminology
Summary
MediX-R1 introduces an open-ended Reinforcement Learning framework for medical multimodal large language models that enables clinically grounded, free-form answers beyond multiple-choice formats The paper presents MediX-R1, an open-ended RL framework that fine-tunes a baseline vision-language backbone with Group Based RL and a composite reward tailored for medical reasoning This framework is significant because it provides a principled approach for training medical MLLMs with open-ended RL that integrates semantic correctness with structural and modality constraints
Framework Overview
MediX-R1 fine-tunes a baseline multimodal backbone for open-ended medical reasoning using Group Based RL Given an image I and question q, the vision encoder produces visual tokens that are fused with text tokens and fed to the LLM policy πθ The model generates structured outputs of the form: [modality tag]free-form clinical reasoningfinal concise answer. This design encourages the model to generate accurate and interpretable reasoning paths. p. 2 This reward is composed of: a) an LLM-based reward for evaluating the overall
quality and correctness of the output; b) an embedding-based reward to ensure semantic alignment; c) a format reward to enforce the
desired output structure (<think> and <answer> blocks); and d) a modality reward to ensure the response is grounded in the specified
imaging modality. This reward-guided approach encourages the model to generate accurate and interpretable reasoning paths.Where the paper says it
How the training signal is built
The composite reward is defined as r = wfmt Rformat + wllm Rllm + wemb Rembed + wmod Rmodality. The specific coefficients were selected through a staged procedure, fixing the format reward weight to 0. p. 3 Reward Design
We define a composite reward:
r = wfmt Rformat + wllm Rllm + wemb Rembed + wmod Rmodality
with coefficients chosen via a simple, staged ablation procedure. In all experiments we keep a small, fixed weight on the
format reward to enforce parseable outputs, and distribute
the remaining weight across semantic correctness (LLM
judge / embeddings) and modality grounding. Moreover, using a single reward
signal can induce instability or reward hacking, especially
when the signal is noisy or overly permissive. Hence, it is
desirable to have a principled approach for training medical MLLMs with open-ended RL that integrates semantic
correctness with structural and modality constraints, while
remaining data- and compute-efficient. 10 in all experiments
and allocated the remaining 0. 90 mass to the task-facing
signals. We present MediX-R1, an open-ended medical RL framework that fine-tunes a baseline multimodal backbone with
Group Based RL (GRPO/GSPO/DAPO) using a composite
reward tailored for clinical reasoning. Our design combines: (1) an LLM-based accuracy reward that enforces
a strict YES/NO decision on semantic correctness, (2) a
medical embedding-based semantic reward that captures
paraphrases and terminology variants, (3) a lightweight format reward that elicits interpretable reasoning traces, and
(4) a modality recognition reward that discourages crossmodality hallucinations by requiring explicit modality tags. This captures correctness and robustness to paraphrasing while keeping the signal discrete. Embedding-Based Reward (Rembed) To further encourage
semantic alignment, we compute cosine similarity between
the predicted answer and the reference using a medical
embedding model MedEmbed-large (Balachandran, 2024).Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the gains come from
The gains stem from this multi-signal design which stabilizes optimization and supplies informative feedback where traditional verifiable or MCQ-only rewards fall short. The LLM judge is the strongest single signal for text correctness, and embeddings complement it by reducing false negatives from paraphrases. p. 6 5, which shows reduced
volatility and fewer signs of reward hacking, these results
suggest that the composite reward both improves aggregate
performance and stabilizes optimization. Key takeaways: (i) The LLM judge is the strongest single
signal for text correctness, and embeddings complement it
by reducing false negatives from paraphrases. (ii) Default
string matching degrades substantially on open-ended multimodal evaluations. (iii) Modality recognition is important
for VLM tasks and drives substantial gains in image+text
tasks; the full composite delivers the best overall results. The composite design
(format + LLM judge + embeddings + modality recognition) stabilizes optimization and reduces reward hacking
(Fig. 5), translating into the best average performance in
Table 2.Where the paper says it
Where the paper says it
Where the paper says it
Evaluation Methodology
To measure progress, a unified, three-stage Reference-based LLM-as-judge evaluation framework is proposed which supports both text-only and image+text tasks under a common protocol. This pipeline involves Generation via vLLM inference, Evaluation using Reference-based LLMas-judge with BASE and MIMIC templates, and Scoring through aggregation of judgment outputs. p. 3 0
Figure 3. Evaluation Framework
Our three-stage evaluation pipeline:
(1) Generation via vLLM inference
on the model under test, (2) Evaluation using Reference-based LLMas-judge with BASE and MIMIC
templates, and (3) Scoring through
aggregation of judgment outputs. By replacing brittle string-overlap metrics
with instruction-tuned judges served via vLLM (Kwon et al. ,
2023), our evaluation captures semantic correctness, reasoning adequacy, and contextual alignment, and scales from
short-form QA to long-form report generation.Where the paper says it
Where the paper says it
Performance and Human Validation
MediX-R1 achieves the highest average score across all benchmarks, demonstrating state-of-the-art performance among open models. Human expert preference studies further corroborate its clinical answer quality, with MediX-R1 being selected as the best response in 72. p. 8 Empirically,
MediX-R1 achieves strong results across diverse medical
benchmarks and shows improved stability and resistance to
reward hacking compared to single-signal RL variants. Human expert preference studies further corroborate its clinical
answer quality, while qualitative examples illustrate faithful grounding and interpretable reasoning traces. MediX-R1: Open Ended Medical Reinforcement Learning
The results demonstrate a strong preference for MediX-R1,
which was selected as the best response in 72. 7% of the
cases. Empirically,
MediX-R1 achieves strong results across diverse medical
benchmarks and shows improved stability and resistance to
reward hacking compared to single-signal RL variants. Human expert preference studies further corroborate its clinical
answer quality, while qualitative examples illustrate faithful grounding and interpretable reasoning traces. Additional details on human
expert evaluation is available in Sec. Evaluation on Real World Clinical Data
To further assess the generalization ability of our model,
we conducted additional evaluation on MedPix 2. The evaluation on MedPix 2. 0 demonstrates that our model,
MediX-R1, consistently outperforms other medical visionlanguage models.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Mitigation of Hacking and Bias
The composite objective mitigates reward hacking by penalizing mismatches in content or modality recognition. Specific mitigation strategies include embedding gating to set Rembed=0 for low-quality answers, and structural control enforced by Rformat. p. 7 ”
Mitigation in MediX-R1 To curb these failures, MediX-R1
employs a composite reward and input/output constraints:
(i) Composite objective: Rllm + Rembed + Rmodality (with
shared Rformat) reduces reliance on any single brittle signal
and penalizes mismatches in content or modality recognition (Table 4). (ii) Embedding gating: set Rembed=0 for
answers below a minimum character/word length, with high
punctuation or non-alphanumeric ratio; strip punctuation
before embedding; calibrate the similarity threshold. 5, which shows reduced
volatility and fewer signs of reward hacking, these results
suggest that the composite reward both improves aggregate
performance and stabilizes optimization. Key takeaways: (i) The LLM judge is the strongest single
signal for text correctness, and embeddings complement it
by reducing false negatives from paraphrases. While we used only publicly available, de-identified datasets under their respective
licenses and did not conduct a prospective human-subjects
study, residual privacy risks can remain; downstream users
should perform auditing prior to redistribution or deployment. We also highlight risks of demographic and dataset
bias and potential amplification of health disparities; future
work should include fairness analyses where legally and
ethically permissible, uncertainty calibration, bias-aware
reward shaping, and clinician-in-the-loop evaluation.Where the paper says it
Where the paper says it
Where the paper says it
Conclusion
Together, the framework demonstrates that carefully composed, structure-aware rewards plus standardized LLM-judge evaluation provide a practical path to scalable and interpretable medical multimodal RL. MediX-R1 achieves strong results across diverse medical benchmarks and shows improved stability and resistance to reward hacking compared to single-signal RL variants. p. 8 A unified vLLM-based evaluation pipeline enables consistent, paraphrase-robust scoring
across both text-only and image+text tasks. Empirically,
MediX-R1 achieves strong results across diverse medical
benchmarks and shows improved stability and resistance to
reward hacking compared to single-signal RL variants. Altogether, the framework demonstrates that carefully composed, structure-aware rewards plus standardized
LLM-judge evaluation provide a practical path to scalable
and interpretable medical multimodal RL fine-tuning. Impact Statement
This work advances methods for open-ended RL of medical
MLLMs, with the goal of improving semantic correctness
under paraphrase, format compliance, and modality grounding in medical question answering and report-style tasks. Misuse risks include self-diagnosis,
generation of misleading medical narratives, and adversarial
prompting with malicious intent. While we used only publicly available, de-identified datasets under their respective
licenses and did not conduct a prospective human-subjects
study, residual privacy risks can remain; downstream users
should perform auditing prior to redistribution or deployment. To support responsible use and scrutiny, we commit to
releasing training/inference code, configurations, checkpoints, curated datasets, and RL/evaluation prompt templates, alongside a model card and clear usage restrictions,
under a CC-BY-NC-SA 4. We also disclose our
use of generative AI tools: assisted coding was limited to
boilerplate scaffolding and minor refactors with all algorithmic logic authored and reviewed manually; writing-support
models were used for grammar and style, while all technical claims and results were verified by the authors. To support responsible use and scrutiny, we commit to
releasing training/inference code, configurations, checkpoints, curated datasets, and RL/evaluation prompt templates, alongside a model card and clear usage restrictions,
under a CC-BY-NC-SA 4. We also disclose our
use of generative AI tools: assisted coding was limited to
boilerplate scaffolding and minor refactors with all algorithmic logic authored and reviewed manually; writing-support
models were used for grammar and style, while all technical claims and results were verified by the authors.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Acknowledgments
This work was partially supported with NVIDIA Academic Grant 2025 and MBZUAI-IITD Research Collaboration Seed Grant. The authors disclose their use of generative AI tools, stating that assisted coding was limited to boilerplate scaffolding and minor refactors with all algorithmic logic authored and reviewed manually. p. 9 To support responsible use and scrutiny, we commit to
releasing training/inference code, configurations, checkpoints, curated datasets, and RL/evaluation prompt templates, alongside a model card and clear usage restrictions,
under a CC-BY-NC-SA 4. We also disclose our
use of generative AI tools: assisted coding was limited to
boilerplate scaffolding and minor refactors with all algorithmic logic authored and reviewed manually; writing-support
models were used for grammar and style, while all technical claims and results were verified by the authors. At the same time, the societal and ethical risks are nontrivial. MediX-R1 is a research prototype and is not intended for clinical or commercial deployment.Where the paper says it
Where the paper says it
References
Balachandran, A. Medembed: Medical-focused embedding models, 2024. p. 9 References
Balachandran, A. Medembed: Medical-focused embedding models, 2024. References
Balachandran, A. Medembed: Medical-focused embedding models, 2024. Medembed: Medical-focused embedding models, 2024. URL https://github. Medembed: Medical-focused embedding models, 2024. URL https://github. URL https://github. com/
abhinand5/MedEmbed. URL https://github. com/
abhinand5/MedEmbed. com/
abhinand5/MedEmbed. , and Wang, B. MediX-R1: Open Ended Medical Reinforcement Learning
Sahal Shaji Mullappilly * 1 Mohammed Irfan Kurpath * 1 Omair Mohamed 2 Mohamed Zidan 3
Fahad Khan 1 Salman Khan 1 Rao Anwer 1 Hisham Cholakkal 1
1Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI)
2
Jubilee Mission Medical College and Research Institute, 3
JJM Medical College
40%
45%
50%
55%
60%
65%
70%
75%
Av
era
g
e
A
c
c
ura
c
y (%)
10K 50K 200K 1M 5M 30M
Training Dataset Size
Params: ~2B ~8B ~30B = params size
Data Available Data Not Available
MediX-R1 MedGemma MedMO HuatuoGPT BiMediX2 MedVLM
Less data,
higher accuracy
MedGemma 27B
68. 4%
MedMO 8B
62. org/abs/2412. , Ouyang, R. , and Wang, B. Huatuogpt-vision, towards injecting medical visual knowledge into multimodal llms
at scale, 2024b. org/abs/
2406. , Kumar, K. org/abs/
2406. , Kumar, K. , Kumar, K. , Dukre, A. , and
Razzak, I. Medmo: Grounding and understanding multimodal large language model for medical images, 2026. , 2018), PathVQA (He et al. ,
2020), PMC-VQA (Zhang et al. , Zhang, R. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement
learning. 12948, 2025. , Adams, L. 12948, 2025. , Adams, L. , and Bressem, ¨
K. Medalpaca–an open-source collection of medical
conversational ai models and training data. , and Bressem, ¨
K. Medalpaca–an open-source collection of medical
conversational ai models and training data. , 2018), PathVQA (He et al. ,
2020), PMC-VQA (Zhang et al. , 2018), PathVQA (He et al. ,
2020), PMC-VQA (Zhang et al. , and Xie, P. Pathvqa:
30000+ questions for medical visual question answering, 2020. The
evaluation includes standard medical QA, multiple-choice,
and open-ended report generation, as well as visual question
answering and clinical image interpretation. The datasets
used for evaluation are as follows:
LLM (text-only) benchmarks: MMLU-Clinical, MMLUBio, MMLU-Med, MMLU-Genetics, MMLU-ProfMed,
MMLU-Anat (Hendrycks et al. Hendrycks, D. , Burns, C. Hendrycks, D. , Burns, C. , and Steinhardt, J. Measuring massive multitask language understanding. , and Steinhardt, J. Measuring massive multitask language understanding. 0 (Siragusa et al. , 2025), a publicly available real-world clinical
VQA dataset derived from the original MedPix (Henigman
& Kennedy, 2025) database maintained by the U. 0 (Siragusa et al. , 2025), a publicly available real-world clinical
VQA dataset derived from the original MedPix (Henigman
& Kennedy, 2025) database maintained by the U. 0 (Siragusa et al. , 2025), a publicly available real-world clinical
VQA dataset derived from the original MedPix (Henigman
& Kennedy, 2025) database maintained by the U. 0 (Siragusa et al. , 2025), a publicly available real-world clinical
VQA dataset derived from the original MedPix (Henigman
& Kennedy, 2025) database maintained by the U. and Kennedy, B. Medpix®: database of medical images, teaching cases, and clinical topics. and Kennedy, B. Medpix®: database of medical images, teaching cases, and clinical topics. Medical
Reference Services Quarterly, 44(3):328–333, 2025. , Oufattole, N. , and
Szolovits, P. What disease does this patient have? , and
Szolovits, P. What disease does this patient have? What disease does this patient have? a
large-scale open domain question answering dataset from
medical exams. What disease does this patient have? a
large-scale open domain question answering dataset from
medical exams. , 2023), PubMedQA (Jin et al. , 2019), MIMIC-CXRSummarization (Johnson et al. , 2023), PubMedQA (Jin et al. , 2019), MIMIC-CXRSummarization (Johnson et al. Johnson, A. , Pollard, T. Johnson, A. , Pollard, T. , and Mark, R. Mimic-iii, a freely accessible critical care database. , and Mark, R. Mimic-iii, a freely accessible critical care database. , and Stoica, I. Efficient
memory management for large language model serving
with pagedattention. , and Stoica, I. Efficient
memory management for large language model serving
with pagedattention. In Proceedings of the ACM SIGOPS
29th Symposium on Operating Systems Principles, 2023. , Gayen, S. In Proceedings of the ACM SIGOPS
29th Symposium on Operating Systems Principles, 2023. , Gayen, S. , and DemnerFushman, D. A dataset of clinically generated visual
questions and answers about radiology images. , and DemnerFushman, D. A dataset of clinically generated visual
questions and answers about radiology images. Moreover, we
have conducted experiments on different baseline VLMs,
including Qwen2. 5-VL, Qwen3-VL (Team, 2025), and
SmolVLM2 (Marafioti et al. Moreover, we
have conducted experiments on different baseline VLMs,
including Qwen2. 5-VL, Qwen3-VL (Team, 2025), and
SmolVLM2 (Marafioti et al. Hence, it is
desirable to have a principled approach for training medical MLLMs with open-ended RL that integrates semantic
correctness with structural and modality constraints, while
remaining data- and compute-efficient. We present MediX-R1, an open-ended medical RL framework that fine-tunes a baseline multimodal backbone with
Group Based RL (GRPO/GSPO/DAPO) using a composite
reward tailored for clinical reasoning. Hence, it is
desirable to have a principled approach for training medical MLLMs with open-ended RL that integrates semantic
correctness with structural and modality constraints, while
remaining data- and compute-efficient. We present MediX-R1, an open-ended medical RL framework that fine-tunes a baseline multimodal backbone with
Group Based RL (GRPO/GSPO/DAPO) using a composite
reward tailored for clinical reasoning. MediX-R1: Open Ended Medical Reinforcement Learning
Sahal Shaji Mullappilly * 1 Mohammed Irfan Kurpath * 1 Omair Mohamed 2 Mohamed Zidan 3
Fahad Khan 1 Salman Khan 1 Rao Anwer 1 Hisham Cholakkal 1
1Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI)
2
Jubilee Mission Medical College and Research Institute, 3
JJM Medical College
40%
45%
50%
55%
60%
65%
70%
75%
Av
era
g
e
A
c
c
ura
c
y (%)
10K 50K 200K 1M 5M 30M
Training Dataset Size
Params: ~2B ~8B ~30B = params size
Data Available Data Not Available
MediX-R1 MedGemma MedMO HuatuoGPT BiMediX2 MedVLM
Less data,
higher accuracy
MedGemma 27B
68. 4%
MedMO 8B
62. MediX-R1: Open Ended Medical Reinforcement Learning
Sahal Shaji Mullappilly * 1 Mohammed Irfan Kurpath * 1 Omair Mohamed 2 Mohamed Zidan 3
Fahad Khan 1 Salman Khan 1 Rao Anwer 1 Hisham Cholakkal 1
1Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI)
2
Jubilee Mission Medical College and Research Institute, 3
JJM Medical College
40%
45%
50%
55%
60%
65%
70%
75%
Av
era
g
e
A
c
c
ura
c
y (%)
10K 50K 200K 1M 5M 30M
Training Dataset Size
Params: ~2B ~8B ~30B = params size
Data Available Data Not Available
MediX-R1 MedGemma MedMO HuatuoGPT BiMediX2 MedVLM
Less data,
higher accuracy
MedGemma 27B
68. 4%
MedMO 8B
62. MediX-R1: Open Ended Medical Reinforcement Learning
Sahal Shaji Mullappilly * 1 Mohammed Irfan Kurpath * 1 Omair Mohamed 2 Mohamed Zidan 3
Fahad Khan 1 Salman Khan 1 Rao Anwer 1 Hisham Cholakkal 1
1Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI)
2
Jubilee Mission Medical College and Research Institute, 3
JJM Medical College
40%
45%
50%
55%
60%
65%
70%
75%
Av
era
g
e
A
c
c
ura
c
y (%)
10K 50K 200K 1M 5M 30M
Training Dataset Size
Params: ~2B ~8B ~30B = params size
Data Available Data Not Available
MediX-R1 MedGemma MedMO HuatuoGPT BiMediX2 MedVLM
Less data,
higher accuracy
MedGemma 27B
68. 4%
MedMO 8B
62. MediX-R1: Open Ended Medical Reinforcement Learning
Sahal Shaji Mullappilly * 1 Mohammed Irfan Kurpath * 1 Omair Mohamed 2 Mohamed Zidan 3
Fahad Khan 1 Salman Khan 1 Rao Anwer 1 Hisham Cholakkal 1
1Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI)
2
Jubilee Mission Medical College and Research Institute, 3
JJM Medical College
40%
45%
50%
55%
60%
65%
70%
75%
Av
era
g
e
A
c
c
ura
c
y (%)
10K 50K 200K 1M 5M 30M
Training Dataset Size
Params: ~2B ~8B ~30B = params size
Data Available Data Not Available
MediX-R1 MedGemma MedMO HuatuoGPT BiMediX2 MedVLM
Less data,
higher accuracy
MedGemma 27B
68. 4%
MedMO 8B
62. MediX-R1: Open Ended Medical Reinforcement Learning
Model Diverse Medical Single-Stage Interpretable Open-Ended Annotation-Free Composite
Modalities RL Reasoning Responses Reasoning RL Reward
MedVLM-R1 (Pan et al. , 2025) ✗ ✓ ✓ ✗ ✓ ✗
BiMediX2 (Mullappilly et al. MediX-R1: Open Ended Medical Reinforcement Learning
Model Diverse Medical Single-Stage Interpretable Open-Ended Annotation-Free Composite
Modalities RL Reasoning Responses Reasoning RL Reward
MedVLM-R1 (Pan et al. , 2025) ✗ ✓ ✓ ✗ ✓ ✗
BiMediX2 (Mullappilly et al. What disease does this patient have? a
large-scale open domain question answering dataset from
medical exams. What disease does this patient have? a
large-scale open domain question answering dataset from
medical exams. , Zhang, R. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement
learning. , Zhang, R. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement
learning. MediX-R1: Open Ended Medical Reinforcement Learning
Sahal Shaji Mullappilly * 1 Mohammed Irfan Kurpath * 1 Omair Mohamed 2 Mohamed Zidan 3
Fahad Khan 1 Salman Khan 1 Rao Anwer 1 Hisham Cholakkal 1
1Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI)
2
Jubilee Mission Medical College and Research Institute, 3
JJM Medical College
40%
45%
50%
55%
60%
65%
70%
75%
Av
era
g
e
A
c
c
ura
c
y (%)
10K 50K 200K 1M 5M 30M
Training Dataset Size
Params: ~2B ~8B ~30B = params size
Data Available Data Not Available
MediX-R1 MedGemma MedMO HuatuoGPT BiMediX2 MedVLM
Less data,
higher accuracy
MedGemma 27B
68. 4%
MedMO 8B
62. MediX-R1: Open Ended Medical Reinforcement Learning
Sahal Shaji Mullappilly * 1 Mohammed Irfan Kurpath * 1 Omair Mohamed 2 Mohamed Zidan 3
Fahad Khan 1 Salman Khan 1 Rao Anwer 1 Hisham Cholakkal 1
1Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI)
2
Jubilee Mission Medical College and Research Institute, 3
JJM Medical College
40%
45%
50%
55%
60%
65%
70%
75%
Av
era
g
e
A
c
c
ura
c
y (%)
10K 50K 200K 1M 5M 30M
Training Dataset Size
Params: ~2B ~8B ~30B = params size
Data Available Data Not Available
MediX-R1 MedGemma MedMO HuatuoGPT BiMediX2 MedVLM
Less data,
higher accuracy
MedGemma 27B
68. 4%
MedMO 8B
62. The
outputs were anonymized and labeled as Model A, Model
B, Model C and Model D with no identifiers provided to
the reviewers. Medical experts were asked to evaluate the
responses against the provided ground truth descriptions
for each question. The
outputs were anonymized and labeled as Model A, Model
B, Model C and Model D with no identifiers provided to
the reviewers. Medical experts were asked to evaluate the
responses against the provided ground truth descriptions
for each question. , 2024) ✓ ✗ ✗ ✓ ✗ ✗
HuatuoGPT-V (Chen et al. , 2024b) ✓ ✗ ✗ ✓ ✗ ✗
MedGemma (Sellergren et al. , 2024) ✓ ✗ ✗ ✓ ✗ ✗
HuatuoGPT-V (Chen et al. , 2024b) ✓ ✗ ✗ ✓ ✗ ✗
MedGemma (Sellergren et al. To support responsible use and scrutiny, we commit to
releasing training/inference code, configurations, checkpoints, curated datasets, and RL/evaluation prompt templates, alongside a model card and clear usage restrictions,
under a CC-BY-NC-SA 4. We also disclose our
use of generative AI tools: assisted coding was limited to
boilerplate scaffolding and minor refactors with all algorithmic logic authored and reviewed manually; writing-support
models were used for grammar and style, while all technical claims and results were verified by the authors. To support responsible use and scrutiny, we commit to
releasing training/inference code, configurations, checkpoints, curated datasets, and RL/evaluation prompt templates, alongside a model card and clear usage restrictions,
under a CC-BY-NC-SA 4. We also disclose our
use of generative AI tools: assisted coding was limited to
boilerplate scaffolding and minor refactors with all algorithmic logic authored and reviewed manually; writing-support
models were used for grammar and style, while all technical claims and results were verified by the authors. Abstract
We introduce MediX-R1, an open-ended Reinforcement Learning (RL) framework for medical
multimodal large language models (MLLMs) that
enables clinically grounded, free-form answers
beyond multiple-choice formats. MediX-R1 finetunes a baseline vision-language backbone with
Group Based RL and a composite reward tailored
for medical reasoning: an LLM-based accuracy
reward that judges semantic correctness with a
strict YES/NO decision, a medical embeddingbased semantic reward to capture paraphrases
and terminology variants, and lightweight format
and modality rewards that enforce interpretable
* Equal contribution. Abstract
We introduce MediX-R1, an open-ended Reinforcement Learning (RL) framework for medical
multimodal large language models (MLLMs) that
enables clinically grounded, free-form answers
beyond multiple-choice formats. MediX-R1 finetunes a baseline vision-language backbone with
Group Based RL and a composite reward tailored
for medical reasoning: an LLM-based accuracy
reward that judges semantic correctness with a
strict YES/NO decision, a medical embeddingbased semantic reward to capture paraphrases
and terminology variants, and lightweight format
and modality rewards that enforce interpretable
* Equal contribution. Evaluation on Real World Clinical Data
To further assess the generalization ability of our model,
we conducted additional evaluation on MedPix 2. 0 (Siragusa et al. Evaluation on Real World Clinical Data
To further assess the generalization ability of our model,
we conducted additional evaluation on MedPix 2. 0 (Siragusa et al. Additional details on human
expert evaluation is available in Sec. Evaluation on Real World Clinical Data
To further assess the generalization ability of our model,
we conducted additional evaluation on MedPix 2. Additional details on human
expert evaluation is available in Sec. Evaluation on Real World Clinical Data
To further assess the generalization ability of our model,
we conducted additional evaluation on MedPix 2. , and
Razzak, I. Medmo: Grounding and understanding multimodal large language model for medical images, 2026. , and
Razzak, I. Medmo: Grounding and understanding multimodal large language model for medical images, 2026. To support responsible use and scrutiny, we commit to
releasing training/inference code, configurations, checkpoints, curated datasets, and RL/evaluation prompt templates, alongside a model card and clear usage restrictions,
under a CC-BY-NC-SA 4. We also disclose our
use of generative AI tools: assisted coding was limited to
boilerplate scaffolding and minor refactors with all algorithmic logic authored and reviewed manually; writing-support
models were used for grammar and style, while all technical claims and results were verified by the authors. To support responsible use and scrutiny, we commit to
releasing training/inference code, configurations, checkpoints, curated datasets, and RL/evaluation prompt templates, alongside a model card and clear usage restrictions,
under a CC-BY-NC-SA 4. We also disclose our
use of generative AI tools: assisted coding was limited to
boilerplate scaffolding and minor refactors with all algorithmic logic authored and reviewed manually; writing-support
models were used for grammar and style, while all technical claims and results were verified by the authors. To support responsible use and scrutiny, we commit to
releasing training/inference code, configurations, checkpoints, curated datasets, and RL/evaluation prompt templates, alongside a model card and clear usage restrictions,
under a CC-BY-NC-SA 4. We also disclose our
use of generative AI tools: assisted coding was limited to
boilerplate scaffolding and minor refactors with all algorithmic logic authored and reviewed manually; writing-support
models were used for grammar and style, while all technical claims and results were verified by the authors. To support responsible use and scrutiny, we commit to
releasing training/inference code, configurations, checkpoints, curated datasets, and RL/evaluation prompt templates, alongside a model card and clear usage restrictions,
under a CC-BY-NC-SA 4. We also disclose our
use of generative AI tools: assisted coding was limited to
boilerplate scaffolding and minor refactors with all algorithmic logic authored and reviewed manually; writing-support
models were used for grammar and style, while all technical claims and results were verified by the authors. To support responsible use and scrutiny, we commit to
releasing training/inference code, configurations, checkpoints, curated datasets, and RL/evaluation prompt templates, alongside a model card and clear usage restrictions,
under a CC-BY-NC-SA 4. We also disclose our
use of generative AI tools: assisted coding was limited to
boilerplate scaffolding and minor refactors with all algorithmic logic authored and reviewed manually; writing-support
models were used for grammar and style, while all technical claims and results were verified by the authors. To support responsible use and scrutiny, we commit to
releasing training/inference code, configurations, checkpoints, curated datasets, and RL/evaluation prompt templates, alongside a model card and clear usage restrictions,
under a CC-BY-NC-SA 4. We also disclose our
use of generative AI tools: assisted coding was limited to
boilerplate scaffolding and minor refactors with all algorithmic logic authored and reviewed manually; writing-support
models were used for grammar and style, while all technical claims and results were verified by the authors. , 2018), PathVQA (He et al. ,
2020), PMC-VQA (Zhang et al. , 2018), PathVQA (He et al. ,
2020), PMC-VQA (Zhang et al. , 2018), PathVQA (He et al. ,
2020), PMC-VQA (Zhang et al. , 2018), PathVQA (He et al. ,
2020), PMC-VQA (Zhang et al. These results highlight the effectiveness of our open-ended
RL training and composite reward design, which enable
MediX-R1 to generate accurate, semantically aligned, and
clinically grounded responses beyond the capabilities of
models trained only with supervised or MCQ objectives. We additionally report results on the Massive Multidiscipline Multimodal Understanding and Reasoning
(MMMU) benchmark (Yue et al. These results highlight the effectiveness of our open-ended
RL training and composite reward design, which enable
MediX-R1 to generate accurate, semantically aligned, and
clinically grounded responses beyond the capabilities of
models trained only with supervised or MCQ objectives. We additionally report results on the Massive Multidiscipline Multimodal Understanding and Reasoning
(MMMU) benchmark (Yue et al. These results highlight the effectiveness of our open-ended
RL training and composite reward design, which enable
MediX-R1 to generate accurate, semantically aligned, and
clinically grounded responses beyond the capabilities of
models trained only with supervised or MCQ objectives. We additionally report results on the Massive Multidiscipline Multimodal Understanding and Reasoning
(MMMU) benchmark (Yue et al. These results highlight the effectiveness of our open-ended
RL training and composite reward design, which enable
MediX-R1 to generate accurate, semantically aligned, and
clinically grounded responses beyond the capabilities of
models trained only with supervised or MCQ objectives. We additionally report results on the Massive Multidiscipline Multimodal Understanding and Reasoning
(MMMU) benchmark (Yue et al. Embedding
Reward
LLM
Reward
Modality
Reward
Format
Reward
think
answer
MRI
XRAY
CT
MediX-R1
GRPO DAPO GSPO
Figure 2. MediX-R1: Overall Architecture The MediX-R1 reinforcement learning framework for open-ended medical reasoning. Embedding
Reward
LLM
Reward
Modality
Reward
Format
Reward
think
answer
MRI
XRAY
CT
MediX-R1
GRPO DAPO GSPO
Figure 2. MediX-R1: Overall Architecture The MediX-R1 reinforcement learning framework for open-ended medical reasoning. State-of-the-art Comparisons
We evaluate MediX-R1 on a comprehensive suite of medical language and vision-language benchmarks, covering
both text-only (LLM) and image+text (VLM) tasks. The
evaluation includes standard medical QA, multiple-choice,
and open-ended report generation, as well as visual question
answering and clinical image interpretation. State-of-the-art Comparisons
We evaluate MediX-R1 on a comprehensive suite of medical language and vision-language benchmarks, covering
both text-only (LLM) and image+text (VLM) tasks. The
evaluation includes standard medical QA, multiple-choice,
and open-ended report generation, as well as visual question
answering and clinical image interpretation. An
input of a medical image and a natural language question is processed by MediX-R1. The model’s policy is trained using Group Based
RL, which leverages a multi-faceted reward signal. An
input of a medical image and a natural language question is processed by MediX-R1. The model’s policy is trained using Group Based
RL, which leverages a multi-faceted reward signal. Hence, it is
desirable to have a principled approach for training medical MLLMs with open-ended RL that integrates semantic
correctness with structural and modality constraints, while
remaining data- and compute-efficient. We present MediX-R1, an open-ended medical RL framework that fine-tunes a baseline multimodal backbone with
Group Based RL (GRPO/GSPO/DAPO) using a composite
reward tailored for clinical reasoning. Hence, it is
desirable to have a principled approach for training medical MLLMs with open-ended RL that integrates semantic
correctness with structural and modality constraints, while
remaining data- and compute-efficient. We present MediX-R1, an open-ended medical RL framework that fine-tunes a baseline multimodal backbone with
Group Based RL (GRPO/GSPO/DAPO) using a composite
reward tailored for clinical reasoning.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Appendix
A.1 Training Data and Modality Distribution
We trained MediX-R1 on 51335 multimodal medical instruction samples spanning 16 modality tags. Table 8 details the Modality Breakdown and Source Dataset composition of these samples. p. 4 For each dataset, we follow the evaluation protocol described in the previous section, using Reference-based LLMas-judge scoring for both short-form and long-form responses. Table 2 summarizes performance on our unified
benchmark suite across several open-source medical models,
including MedVLM-R1 (2B), BiMediX2 (8B), HuatuoGPTV (7B), MedGemma (4B/27B), MedMO (8B) (Deria et al.Where the paper says it
A.4 Reward Function Source Code
The Format reward function checks for the presence of the required structural pattern. The LLM-based accuracy reward compares the predicted answer against the ground truth using an external helper. p. 4 ” with ground truth “Low perfusion” a
model response that includes hidden reasoning and the final
answer “Low blood flow or less perfusion” is judged correct
and assigned a score of 1. The judge compares predicted
answers against references, accounting for paraphrase and
clinically equivalent phrasing. Embedding-Based Reward (Rembed) To further encourage
semantic alignment, we compute cosine similarity between
the predicted answer and the reference using a medical
embedding model MedEmbed-large (Balachandran, 2024). We convert it to a binary reward via a threshold (default
0. This stabilizes training
and improves interpretability of the medical reasoning. Modality recognition Reward (Rmodality) We encourage
explicit grounding to the imaging modality by requiring
the model to emit the predicted modality tag before the
</think> block (case-insensitive).Where the paper says it
Where the paper says it
Where the paper says it
A.7 Reinforcement Learning Training Prompt
The RL training prompt enforces an explicit modality tag, structured reasoning in tags, and a concise final answer in tags. This structure aligns with the format reward (Rformat) and modality reward (Rmodality) in our composite objective. p. 7 0) against the reference “Lung. ”
Mitigation in MediX-R1 To curb these failures, MediX-R1
employs a composite reward and input/output constraints:
(i) Composite objective: Rllm + Rembed + Rmodality (with
shared Rformat) reduces reliance on any single brittle signal
and penalizes mismatches in content or modality recognition (Table 4). 1Rformat + 0. For the combined semantic reward, we evaluated a small set of intuitive splits between
Rllm and Rembed while keeping wfmt fixed; performance was
similar across these choices on our validation benchmark,
and we selected the configuration that slightly favored the
LLM judge. We present MediX-R1, an open-ended medical RL framework that fine-tunes a baseline multimodal backbone with
Group Based RL (GRPO/GSPO/DAPO) using a composite
reward tailored for clinical reasoning. Our design combines: (1) an LLM-based accuracy reward that enforces
a strict YES/NO decision on semantic correctness, (2) a
medical embedding-based semantic reward that captures
paraphrases and terminology variants, (3) a lightweight format reward that elicits interpretable reasoning traces, and
(4) a modality recognition reward that discourages crossmodality hallucinations by requiring explicit modality tags. 0) against the reference “Lung. ”
Mitigation in MediX-R1 To curb these failures, MediX-R1
employs a composite reward and input/output constraints:
(i) Composite objective: Rllm + Rembed + Rmodality (with
shared Rformat) reduces reliance on any single brittle signal
and penalizes mismatches in content or modality recognition (Table 4).Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Impact Statement
This work advances methods for open-ended RL of medical MLLMs, with the goal of improving semantic correctness under paraphrase, format compliance, and modality grounding in medical question answering and report-style tasks. If used appropriately, the proposed composite-reward RL and unified Referencebased LLM-as-judge evaluation framework may reduce reliance on brittle string-match metrics. p. 8 If
used appropriately (e. , as a research and education aid),
the proposed composite-reward RL and unified Referencebased LLM-as-judge evaluation framework may reduce reliance on brittle string-match metrics, enable more realistic
benchmarking of free-form clinical responses, and improve
transparency through structured outputs (modality tags and
separated reasoning/answer blocks). At the same time, the societal and ethical risks are nontrivial. MediX-R1 is a research prototype and is not intended for clinical or commercial deployment.Where the paper says it
Where the paper says it
Balachandran, A. Medembed: Medical-focused embedding models, 2024. URL https://github.com/abhinand5/MedEmbed. Chen, J., Cai, Z., Ji, K., Wang, X.: towards medical complex reasoning with llms Chen, J. p. 9 , and Wang, B. Huatuogpt-o1, towards medical
complex reasoning with llms, 2024a. , and Wang, B. Huatuogpt-vision, towards injecting medical visual knowledge into multimodal llms
at scale, 2024b. , and
Razzak, I. Medmo: Grounding and understanding multimodal large language model for medical images, 2026. , Zhang, R. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement
learning. , and Xie, P. Pathvqa:
30000+ questions for medical visual question answering, 2020.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Open-ended reasoning capability in medical MLLMs can be achieved by fine-tuning a baseline vision-language backbone with Group Based RL using a composite reward tailored for medical reasoning. This enables models to generate
clinically grounded, free-form answers beyond multiple-choice formats.
-
The system can produce structured outputs that enforce interpretability through the use of and blocks, which are stabilized by a lightweight format reward that
elicits interpretable reasoning traces.
-
The model can perform robust semantic alignment across complex medical phrasing by utilizing an embedding-based semantic reward, which captures
paraphrases and terminology variants.
-
Modality grounding can be enforced to prevent hallucinations by requiring explicit modality tags before the thinking block, which
discourages cross-modality hallucinations by requiring explicit modality tags.
-
A unified evaluation framework allows for measuring progress across both text-only and image+text tasks using a
Reference-based LLM-as-judge,
which capturessemantic correctness, reasoning, and contextual alignment
in a scalable manner. -
The system can generate detailed clinical reports by utilizing an instruction template that forces the model to first output a modality tag, then its thinking process in tags, and finally the answer in tags.
Abstract
In this work, we introduce MediX-R1, an open-ended Reinforcement Learning (RL) framework for medical vision language models (VLMs) that enables free-form answers beyond multiple-choice formats. MediX-R1 fine-tunes a baseline vision-language backbone with Group Based RL and a composite reward tailored for medical reasoning: an LLM-based accuracy reward that judges semantic correctness with a strict YES/NO decision, a medical embedding-based semantic reward to capture paraphrases and terminology variants, and lightweight format and modality rewards that enforce interpretable reasoning and modality recognition. This multi-signal design provides stable, informative feedback for open-ended outputs where traditional verifiable or MCQ-only rewards fall short. To evaluate performance, we employ a unified evaluation framework for both text-only and multimodal (image+text) tasks that uses a Reference-based LLM-as-judge in place of brittle string-overlap metrics, capturing semantic correctness, clinical accuracy, and contextual alignment. Our results show that open-ended RL with comprehensive reward signals can effectively strengthen medical reasoning in multimodal models. Despite using only about51 K instruction examples, MediX-R1 achieves strong performance across standard medical LLM (text-only) and VLM (image + text) benchmarks, outperforming existing open-weight medical VLMs, with large gains particularly on open-ended clinical tasks. These improvements remain consistent across multiple RL algorithms and VLM backbones and are further supported by blinded evaluations with medical experts. Our trained models, curated datasets and source code are available at https://medix.cvmbzuai.com
Sources
- HuatuoGPT-o1, Towards Medical Complex Reasoning with LLMs
- HuatuoGPT-Vision, Towards Injecting Medical Visual Knowledge into Multimodal LLMs at Scale
- MedMO: Grounding and Understanding Multimodal Large Language Model for Medical Images
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- MedAlpaca -- An Open-Source Collection of Medical Conversational AI Models and Training Data
- Measuring Massive Multitask Language Understanding
- PubMedQA: A Dataset for Biomedical Research Question Answering
- SmolVLM: Redefining small and efficient multimodal models
- BiMediX2: Bio-Medical EXpert LMM for Diverse Medical Modalities
- MedGemma Technical Report
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Qwen3 Technical Report
- MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI
- PMC-VQA: Visual Instruction Tuning for Medical Visual Question Answering
- Group Sequence Policy Optimization
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs