Pith Number

pith:SUUZZ533

pith:2026:SUUZZ533FBXSSFHLJZGXK2NXDB

not attested not anchored not stored refs pending

Universal Skeleton Understanding via Differentiable Rendering and MLLMs

Kai-Kuang Ma, Mengyuan Liu, Peiming Li, Xinshun Wang, Yang Tang, Ziyi Wang

Differentiable rendering converts arbitrary skeleton sequences into images that MLLMs can process directly.

arxiv:2603.18003 v4 · 2026-03-18 · 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{SUUZZ533FBXSSFHLJZGXK2NXDB}

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

SkeletonLLM achieves universal skeleton understanding by translating arbitrary skeleton sequences into the MLLM's native visual modality via DrAction, with cooperative training enabling strong generalization in open-vocabulary action recognition and extension to motion captioning and QA across heterogeneous formats.

C2weakest assumption

That converting skeleton kinematics into compact image sequences via differentiable rendering preserves all task-relevant information without significant loss and that MLLM gradients can meaningfully guide the renderer to produce informative visual tokens.

C3one line summary

SkeletonLLM translates arbitrary skeleton sequences into visual image sequences via a differentiable renderer DrAction, allowing MLLMs to perform open-vocabulary action recognition, captioning, and QA across heterogeneous skeleton formats.

Formal links

2 machine-checked theorem links

Cited by

2 papers in Pith

Lens Privacy Sealing: A New Benchmark and Method for Physical Privacy-Preserving Action Recognition

Receipt and verification

First computed	2026-05-20T01:05:11.461264Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

95299cf77b286f2914eb4e4d7569b71843ff85f1a54a34495bfb0d9bd31f3677

Aliases

arxiv: 2603.18003 · arxiv_version: 2603.18003v4 · doi: 10.48550/arxiv.2603.18003 · pith_short_12: SUUZZ533FBXS · pith_short_16: SUUZZ533FBXSSFHL · pith_short_8: SUUZZ533

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/SUUZZ533FBXSSFHLJZGXK2NXDB \
  | 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: 95299cf77b286f2914eb4e4d7569b71843ff85f1a54a34495bfb0d9bd31f3677

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "3b9be5fcea6414be906fca4d985389f559572a957794e3559b6b5a5d406063cd",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by-nc-nd/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-03-18T17:59:12Z",
    "title_canon_sha256": "6ee668eae94df421a3bb42bb3de036911ca2050d51e070adadb41b5de605df38"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2603.18003",
    "kind": "arxiv",
    "version": 4
  }
}