Pith Number

pith:E25GRCME

pith:2026:E25GRCMEVIVYKWIEQV37DKGIBP

not attested not anchored not stored refs pending

Divide and Conquer: Decoupled Representation Alignment for Multimodal World Models

Bin Xia, Dingkang Liang, Guangmo Yi, Jianlou Si, Jun Huang, Junyuan Xiao, Qiang Lyu, Shurui Shi, Tongtong Su, Wenming Yang, Xin Zhou, Yixuan Ye

Decoupling modality features in a diffusion model allows each to align with a separate foundation model expert for improved multi-modal video generation.

arxiv:2605.01896 v2 · 2026-05-03 · 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{E25GRCMEVIVYKWIEQV37DKGIBP}

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

This design enables joint optimization, fully exploiting priors from multiple foundation models. Extensive experiments demonstrate that our method significantly outperforms baselines in visual quality and long-term consistency.

C2weakest assumption

Foundation models trained on different modality spaces naturally capture distinct domain-specific priors, acting as complementary experts, and that the proposed decoupling regularization will maintain this complementarity without introducing new conflicts during joint optimization.

C3one line summary

M²-REPA decouples modality-specific features inside a diffusion model and aligns each to its matching expert foundation model via an alignment loss plus a decoupling regularizer, yielding better visual quality and long-term consistency in multi-modal video generation.

Cited by

1 paper in Pith

PhysRAG: Enhancing Physics-Awareness in Video Generation via Retrieval-Augmented Generation

Receipt and verification

First computed	2026-07-02T01:17:32.037792Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

26ba688984aa2b8559048577f1a8c80bf6317aec6c146228e1bf54ec6a9a7857

Aliases

arxiv: 2605.01896 · arxiv_version: 2605.01896v2 · doi: 10.48550/arxiv.2605.01896 · pith_short_12: E25GRCMEVIVY · pith_short_16: E25GRCMEVIVYKWIE · pith_short_8: E25GRCME

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/E25GRCMEVIVYKWIEQV37DKGIBP \
  | 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: 26ba688984aa2b8559048577f1a8c80bf6317aec6c146228e1bf54ec6a9a7857

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "b7181e8bcd07ac9db4ee641caa78ccfccb5c06f9ea71f102d4f9c609e077a1fc",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-05-03T14:22:22Z",
    "title_canon_sha256": "c5197caf6bb6e357e1fde407eede488534ed061bd7b76c85ade173ceefd6409d"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.01896",
    "kind": "arxiv",
    "version": 2
  }
}