Pith. sign in

REVIEW 13 cited by

HuatuoGPT, towards Taming Language Model to Be a Doctor

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2305.15075 v1 pith:7LF4ASBE submitted 2023-05-24 cs.CL cs.AI

classification cs.CLcs.AI
keywords datahuatuogptmodellanguagedistilledchatgptdoctorsevaluation
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

In this paper, we present HuatuoGPT, a large language model (LLM) for medical consultation. The core recipe of HuatuoGPT is to leverage both \textit{distilled data from ChatGPT} and \textit{real-world data from doctors} in the supervised fine-tuned stage. The responses of ChatGPT are usually detailed, well-presented and informative while it cannot perform like a doctor in many aspects, e.g. for integrative diagnosis. We argue that real-world data from doctors would be complementary to distilled data in the sense the former could tame a distilled language model to perform like doctors. To better leverage the strengths of both data, we train a reward model to align the language model with the merits that both data bring, following an RLAIF (reinforced learning from AI feedback) fashion. To evaluate and benchmark the models, we propose a comprehensive evaluation scheme (including automatic and manual metrics). Experimental results demonstrate that HuatuoGPT achieves state-of-the-art results in performing medical consultation among open-source LLMs in GPT-4 evaluation, human evaluation, and medical benchmark datasets. It is worth noting that by using additional real-world data and RLAIF, the distilled language model (i.e., HuatuoGPT) outperforms its teacher model ChatGPT in most cases. Our code, data, and models are publicly available at \url{https://github.com/FreedomIntelligence/HuatuoGPT}. The online demo is available at \url{https://www.HuatuoGPT.cn/}.

Discussion (0). Sign in to comment.

Forward citations

Cited by 13 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Enhancing Pathological VLMs with Cross-scale Reasoning

    cs.CV 2026-06 unverdicted novelty 7.0 of 10

    Presents Scale-VQA benchmark for cross-scale pathology VQA and RL-trained ScaleReasoner-R1 model that reaches SOTA on the new benchmark plus existing single-scale tasks.

  2. DrVD-Bench: Do Vision-Language Models Reason Like Human Doctors in Medical Image Diagnosis?

    cs.CV 2025-05 conditional novelty 7.0 of 10

    A new five-level medical imaging benchmark, DrVD-Bench, shows that vision-language models lose accuracy sharply as reasoning complexity grows and often diagnose without grounding in lesion evidence.

  3. PathScale-R1: Cross-scale Reasoning for Pathological Image Analysis

    cs.CV 2026-07 conditional novelty 6.0 of 10

    PathScale-R1 and PathScale-VQA raise cross-scale pathology VQA accuracy to 83% by training on expert multi-magnification paths with anti-shortcut curation and scale-aware RL.

  4. SCOPE and SCION: A Benchmark and an Auditable Reference Pipeline for Schema Induction and Fusion from Text

    cs.AI 2026-05 conditional novelty 6.0 of 10

    A 24-dataset benchmark for inducing schema graphs from raw text, plus an auditable LLM-based pipeline that reports the highest scores on the benchmark's four schema-similarity metrics.

  5. Toward a Vision-Language Foundation Model for Medical Data: Multimodal Dataset and Benchmarks for Vietnamese PET/CT Report Generation

    cs.CV 2025-09 conditional novelty 6.0 of 10

    A new Vietnamese PET/CT-report dataset improves medical VLM report generation and VQA, but clinical F1 scores remain modest.

  6. Toward Better EHR Reasoning in LLMs: Reinforcement Learning with Expert Attention Guidance

    cs.AI 2025-08 conditional novelty 6.0 of 10

    EAG-RL improves LLM performance on EHR mortality and readmission prediction by training on expert-generated reasoning traces and an attention-alignment RL reward.

  7. Leaps Beyond the Seen: Reinforced Reasoning Augmented Generation for Clinical Notes

    cs.CL 2025-06 conditional novelty 6.0 of 10

    ReinRAG uses reinforcement learning to select knowledge-graph reasoning paths, including deliberate leaps across semantic clusters, to help an LLM generate discharge instructions from sparse pre-admission clinical inf...

  8. Medical World Model: Generative Simulation of Tumor Evolution for Treatment Planning

    cs.CV 2025-06 conditional novelty 6.0 of 10

    MeWM combines a GPT-style policy, a diffusion tumor dynamics model, and a survival analysis heuristic to simulate post-treatment tumor appearance and select TACE treatment plans, improving physician F1-score by 13 points.

  9. Enhancing Visual Reliance in Text Generation: A Bayesian Perspective on Mitigating Hallucination in Large Vision-Language Models

    cs.CV 2025-05 conditional novelty 6.0 of 10

    EVRB is a three-part inference-time method that prunes ambiguous visual tokens, divides the model's output distribution by a text-only prior, and triggers early stopping to reduce hallucination in LVLMs.

  10. Gaokerena: A Small Persian Medical Language Model Family

    cs.CL 2026-08 conditional novelty 5.0 of 10

    Fine-tuned Persian medical language models reach 49-53% on translated medical MMLU, with datasets released, but the reasoning variant's gain depends on extra test-time compute and a verifier.

  11. LCDS: A Logic-Controlled Discharge Summary Generation System Supporting Source Attribution and Expert Review

    cs.CL 2025-07 conditional novelty 5.0 of 10

    A logic-controlled pipeline with source mapping and sentence-level attribution generates discharge summaries that score higher than a GPT-4o chain-of-thought baseline in this study.

  12. DoPI: Doctor-like Proactive Interrogation LLM for Traditional Chinese Medicine

    cs.AI 2025-07 reject novelty 5.0 of 10

    DoPI pairs a knowledge-graph-guided questioning model with a TCM expert model and claims 84.68% diagnostic accuracy, but the benchmark is built from the same symptom-disease rules that drive the system.

  13. Beyond Distillation: Pushing the Limits of Medical LLM Reasoning with Minimalist Rule-Based RL

    cs.CL 2025-05 conditional novelty 4.0 of 10

    Medical QA accuracy improves substantially through reinforcement learning with a binary correct-answer reward alone, without supervised fine-tuning on distilled reasoning traces.

Pith tools