pith. sign in
Pith Number

pith:GQ37DA3B

pith:2025:GQ37DA3BEJQ5ZGO6DKY3C3CIWC
not attested not anchored not stored refs resolved

JointAVBench: A Benchmark for Joint Audio-Visual Reasoning Evaluation

Jianghan Chao, Jianzhang Gao, Liyun Ru, Ruihua Song, Wenhui Tan, Yuchong Sun

Even the best Omni-LLMs reach only 65.3 percent average accuracy on a benchmark that demands strict joint audio-visual reasoning in videos.

arxiv:2512.12772 v2 · 2025-12-14 · cs.MM · cs.CV

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{GQ37DA3BEJQ5ZGO6DKY3C3CIWC}

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

even the best-performing Omni-LLM achieves an average accuracy of only 65.3%, outperforming uni-modal baselines but revealing substantial room for improvement, especially in cross-scene reasoning.

C2weakest assumption

The automated pipeline using vision-LLMs, audio-LLMs, and general LLMs produces questions and answers that strictly require joint audio-visual understanding without introducing biases or answer leakage from the generation process itself.

C3one line summary

JointAVBench is a benchmark for joint audio-visual reasoning that shows leading Omni-LLMs reach only 65.3% accuracy, with particular weakness in cross-scene tasks.

References

53 extracted · 53 resolved · 0 Pith anchors

[1] achieves optimal performance in identifying potential hallucinations. During the general check, we utilize only the QA pair and its explanation to filter out unqualified QA pairs. This stage includes 2024
[2] Only include details that are clearly visible in the video
[3] Focus on their most significant movements, gestures, and interactions
[4] Include the sequence of events and the pacing of the scene to convey how it unfolds over time
[5] Only describe emotions that are clearly expressed through visible actions or expressions

Cited by

5 papers in Pith

Receipt and verification
First computed 2026-05-17T23:39:00.521344Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

3437f183612261dc99de1ab1b16c48b0a44c4604c1a8637a855d0b16143662e5

Aliases

arxiv: 2512.12772 · arxiv_version: 2512.12772v2 · doi: 10.48550/arxiv.2512.12772 · pith_short_12: GQ37DA3BEJQ5 · pith_short_16: GQ37DA3BEJQ5ZGO6 · pith_short_8: GQ37DA3B
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/GQ37DA3BEJQ5ZGO6DKY3C3CIWC \
  | 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: 3437f183612261dc99de1ab1b16c48b0a44c4604c1a8637a855d0b16143662e5
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "b9b2e3d389fd0cabfe9f758af7ae6591daa7eddc251f5e1570e98b50af9b80a1",
    "cross_cats_sorted": [
      "cs.CV"
    ],
    "license": "http://creativecommons.org/licenses/by-nc-sa/4.0/",
    "primary_cat": "cs.MM",
    "submitted_at": "2025-12-14T17:23:21Z",
    "title_canon_sha256": "33a5624ddaf1623adcbeecca9e9824644a2f5f8dada4f0eb79edcebfff59965a"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2512.12772",
    "kind": "arxiv",
    "version": 2
  }
}