Pith Number

pith:HFQBZVID

pith:2026:HFQBZVIDLX4CUCQZJBDV3CIAEL

not attested not anchored not stored refs pending

Mind the Generative Details: Direct Localized Detail Preference Optimization for Video Diffusion Models

Chao Gao, Kaidong Zhang, Rui Ding, Wangmeng Zuo, Ying Chen, Yukang Ding, Zitong Huang

LocalDPO aligns text-to-video diffusion models by optimizing preferences only on locally corrupted regions of real videos.

arxiv:2601.04068 v4 · 2026-01-07 · cs.CV · cs.AI

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{HFQBZVIDLX4CUCQZJBDV3CIAEL}

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

Experiments on Wan2.1 and CogVideoX demonstrate that LocalDPO consistently improves video fidelity, temporal coherence and human preference scores over other post-training approaches, establishing a more efficient and fine-grained paradigm for video generator alignment.

C2weakest assumption

That videos created by locally masking real footage and inpainting only the masked regions with the frozen base model produce negatives whose flaws correspond to the kinds of errors humans actually dislike at the region level.

C3one line summary

LocalDPO creates localized preference pairs from real videos by applying random spatio-temporal masks and restoring masked regions with the frozen base model, then applies region-restricted DPO loss to improve fidelity and coherence in video diffusion models.

Formal links

1 machine-checked theorem link

Cited by

1 paper in Pith

CreFlow: Corrective Reflow for Sparse-Reward Embodied Video Diffusion RL

Receipt and verification

First computed	2026-05-21T01:04:20.654858Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

39601cd5035df82a0a1948475d890022d1a5caa7e4fe1cfe2ac2c83965b3c8b4

Aliases

arxiv: 2601.04068 · arxiv_version: 2601.04068v4 · doi: 10.48550/arxiv.2601.04068 · pith_short_12: HFQBZVIDLX4C · pith_short_16: HFQBZVIDLX4CUCQZ · pith_short_8: HFQBZVID

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/HFQBZVIDLX4CUCQZJBDV3CIAEL \
  | 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: 39601cd5035df82a0a1948475d890022d1a5caa7e4fe1cfe2ac2c83965b3c8b4

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "43878163f4f400605643cda096dfa433a4547fb1f1f7f6d08d8bd90c7630978b",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by-nc-sa/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-01-07T16:32:17Z",
    "title_canon_sha256": "88155bb746b5acc291ca04674b6ac289e3d0a10639fd4dfc314d8efafefd3ca0"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2601.04068",
    "kind": "arxiv",
    "version": 4
  }
}