Pith Number

pith:TXZZOX6H

pith:2025:TXZZOX6HIQM2T2SRD4LEUKCH7Q

not attested not anchored not stored refs resolved

Lingshu: A Generalist Foundation Model for Unified Multimodal Medical Understanding and Reasoning

Chaojun Wang, Chaoqun Liu, Chenghao Xiao, Deli Zhao, Guizhen Chen, Hao Zhang, Hou Pong Chan, Jianyu Wang, Jie Tan, Junao Shen, LASA Team, Long Li, Mahani Aljunied, Ruifeng Yuan, Tingyang Xu, Weiwen Xu, Yu Rong, Yu Sun, Zhaodonghui Li

Lingshu, a medical multimodal model, outperforms open-source peers on visual QA, text QA, and report generation after targeted data curation and staged training.

arxiv:2506.07044 v4 · 2025-06-08 · cs.CL · cs.AI · cs.CV

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{TXZZOX6HIQM2T2SRD4LEUKCH7Q}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

The results show that Lingshu consistently outperforms the existing open-source multimodal models on most tasks across multimodal QA, text-based QA, and medical report generation.

C2weakest assumption

That the synthesized captions, VQA pairs, and reasoning samples produced by the data curation procedure are accurate and free of hallucinations or factual errors that would propagate into the model.

C3one line summary

Lingshu is a medical-specialized multimodal LLM that outperforms prior open-source models on multimodal QA, text QA, and report generation after training on a large curated dataset of medical knowledge.

References

95 extracted · 95 resolved · 6 Pith anchors

[1] DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning 2024 · doi:10.1016/j.artmed.2024.103001

[2] Meddr: Diagnosis-guided bootstrapping for large-scale medical vision-language learning.arXiv preprint arXiv:2404.15127 2003

[3] Maira-1: A specialisedlargemultimodalmodelforradiologyreportgeneration.arXivpreprintarXiv:2311.13668 2024

[4] https://doi.org/10.1609/aaai.v33i01.3301590. Saahil Jain, Ashwin Agrawal, Adriel Saporta, Steven Truong, Du Nguyen Duong Nguyen Duong, Tan Bui, Pierre Chambon, Yuhao Zhang, Matthew Lungren, Andrew Ng, 2021 · doi:10.1609/aaai.v33i01.3301590

[5] URLhttps://www.nature.com/articles/s41597-019-0322-0 2076 · doi:10.1038/s41597-019-0322-0

Formal links

3 machine-checked theorem links

Cited by

38 papers in Pith

MedExpMem: Adapting Experience Memory for Differential Diagnosis

DDX-TRACE: A Benchmark for Medical Diagnostic Trajectories in VLMs

Look-Closer-Then-Diagnose: Confidence-Aware Ultrasound VQA via Active Zooming

Seizure-Semiology-Suite (S3): A Clinically Multimodal Dataset, Benchmark, and Models for Seizure Semiology Understanding

JMed48k: A Multi-Profession Japanese Medical Licensing Benchmark for Vision-Language Model Evaluation

Receipt and verification

First computed	2026-05-17T23:38:52.713597Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

9df3975fc74419a9ea511f164a2847fc16168471c8c5f85693b29d1d14cd6bef

Aliases

arxiv: 2506.07044 · arxiv_version: 2506.07044v4 · doi: 10.48550/arxiv.2506.07044 · pith_short_12: TXZZOX6HIQM2 · pith_short_16: TXZZOX6HIQM2T2SR · pith_short_8: TXZZOX6H

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/TXZZOX6HIQM2T2SRD4LEUKCH7Q \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 9df3975fc74419a9ea511f164a2847fc16168471c8c5f85693b29d1d14cd6bef

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "226cc92a595c243d9368b89129df002f3da8ed9c2b3bb3629f3562c4024a07dd",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.CV"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2025-06-08T08:47:30Z",
    "title_canon_sha256": "6b6ef6f5d5401cee547e3c2a4f28813c735544e11425063d383145f42c3a8f48"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2506.07044",
    "kind": "arxiv",
    "version": 4
  }
}