Pith Number

pith:BVMOWBJT

pith:2022:BVMOWBJTWNGHNGHCDHNCMWDIUJ

not attested not anchored not stored refs resolved

MagicVideo: Efficient Video Generation With Latent Diffusion Models

Daquan Zhou, Hanshu Yan, Jiashi Feng, Weimin Wang, Weiwei Lv, Yizhe Zhu

MagicVideo generates 256x256 text-to-video clips on a single GPU using 64 times fewer computations than prior video diffusion models.

arxiv:2211.11018 v2 · 2022-11-20 · 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{BVMOWBJTWNGHNGHCDHNCMWDIUJ}

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

MagicVideo can synthesize video clips with 256x256 spatial resolution on a single GPU card, which takes around 64x fewer computations than the Video Diffusion Models (VDM) in terms of FLOPs.

C2weakest assumption

That a pre-trained VAE maps video clips into a low-dimensional latent space while preserving enough spatial and temporal information for the diffusion model to produce high-fidelity, temporally coherent output without major artifacts.

C3one line summary

MagicVideo generates 256x256 text-conditioned video clips via latent diffusion with a custom 3D U-Net, achieving roughly 64 times lower compute than prior video diffusion models.

References

52 extracted · 52 resolved · 14 Pith anchors

[1] Layer Normalization 2016 · arXiv:1607.06450

[2] Fitvid: Overﬁtting in pixel-level video prediction 2021

[3] Frozen in time: A joint video and im- age encoder for end-to-end retrieval 2021

[4] Multimodal datasets: misog- yny, pornography, and malignant stereotypes 2021

[5] Quo vadis, action recognition? a new model and the kinet- ics dataset 2017

Formal links

2 machine-checked theorem links

Cited by

27 papers in Pith

Wan: Open and Advanced Large-Scale Video Generative Models

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

Bridging the Intention-Expression Gap: Aligning Multi-Dimensional Preferences via Hierarchical Relevance Feedback in Text-to-Image Diffusion

Functionalization via Structure Completion and Motion Rectification

SynMotion: Semantic-Visual Adaptation for Motion Customized Video Generation

Receipt and verification

First computed	2026-05-17T23:38:50.548857Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

0d58eb0533b34c7698e219da265868a262497b0300ed8e454bb5a47f01ecd464

Aliases

arxiv: 2211.11018 · arxiv_version: 2211.11018v2 · doi: 10.48550/arxiv.2211.11018 · pith_short_12: BVMOWBJTWNGH · pith_short_16: BVMOWBJTWNGHNGHC · pith_short_8: BVMOWBJT

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/BVMOWBJTWNGHNGHCDHNCMWDIUJ \
  | 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: 0d58eb0533b34c7698e219da265868a262497b0300ed8e454bb5a47f01ecd464

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "7259079340ecbd67a84d461387463fd1df4e15a9be3062d003f7333722207e33",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2022-11-20T16:40:31Z",
    "title_canon_sha256": "de4ab79fa9f532ee899e58950dcaac9a6f1d1a44f345fafc483d8f605d57b75b"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2211.11018",
    "kind": "arxiv",
    "version": 2
  }
}