pith. sign in
Pith Number

pith:XFMCH2AC

pith:2024:XFMCH2AC4HGE6PVO3NJHRFWJ3C
not attested not anchored not stored refs resolved

MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark

Botao Yu, Ge Zhang, Graham Neubig, Huan Sun, Kai Zhang, Shengbang Tong, Tianyu Zheng, Wenhu Chen, Xiang Yue, Yuansheng Ni, Yubo Wang, Yu Su, Yuxuan Sun

Multimodal models show 17 to 27 percent lower accuracy on MMMU-Pro than on MMMU.

arxiv:2409.02813 v3 · 2024-09-04 · cs.CL · cs.CV

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{XFMCH2AC4HGE6PVO3NJHRFWJ3C}

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

Results show that model performance is substantially lower on MMMU-Pro than on MMMU, ranging from 16.8% to 26.9% across models.

C2weakest assumption

The assumption that questions solvable by text-only models require no visual understanding and that embedding questions in images cleanly tests seamless visual-textual integration without introducing new confounds or biases.

C3one line summary

MMMU-Pro is a stricter multimodal benchmark that removes text-only solvable questions, augments options, and requires reading text from images, yielding substantially lower model scores of 16.8-26.9%.

References

73 extracted · 73 resolved · 24 Pith anchors

[1] Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone 2024 · arXiv:2404.14219
[2] Jean-Baptiste Alayrac, Jeff Donahue, Pauline Luc, Antoine Miech, Iain Barr, Yana Hasson, Karel Lenc, Arthur Mensch, Katherine Millican, Malcolm Reynolds, et al. 2022. Flamingo: a visual language model 2022
[3] Anthropic. 2024. https://www.anthropic.com/news/claude-3-5-sonnet Claude 3.5 sonnet. https://www.anthropic.com/news/claude-3-5-sonnet 2024
[4] Lawrence Zitnick, and Devi Parikh 2015 · doi:10.1109/iccv.2015.279
[5] OpenFlamingo: An Open-Source Framework for Training Large Autoregressive Vision-Language Models 2023 · arXiv:2308.01390

Cited by

39 papers in Pith

Receipt and verification
First computed 2026-05-18T03:12:35.515185Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

b95823e802e1cc4f3eaedb527896c9d8ad685b09e125b553c5dd36f781fada47

Aliases

arxiv: 2409.02813 · arxiv_version: 2409.02813v3 · doi: 10.48550/arxiv.2409.02813 · pith_short_12: XFMCH2AC4HGE · pith_short_16: XFMCH2AC4HGE6PVO · pith_short_8: XFMCH2AC
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/XFMCH2AC4HGE6PVO3NJHRFWJ3C \
  | 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: b95823e802e1cc4f3eaedb527896c9d8ad685b09e125b553c5dd36f781fada47
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "03fddb64137f7efda7b707f3951553b297797400c7bc0d5b2d748cd0b5e97744",
    "cross_cats_sorted": [
      "cs.CV"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2024-09-04T15:31:26Z",
    "title_canon_sha256": "65f7013ef12de1a9b7bc90dab017a73fc9dd8b5e2b271edadf5a123aaa6ccbc0"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2409.02813",
    "kind": "arxiv",
    "version": 3
  }
}