Pith Number

pith:HUNVO5KQ

pith:2026:HUNVO5KQQWMUFOYZBSABM2ACOA

not attested not anchored not stored refs resolved

DEVIS-GRPO: Unleashing GRPO on Dynamic Extreme View Synthesis

Fang Liu, Huimin Wu, Licheng Jiao, Lingling Li, Qing Li, Yi Zuo

Accumulating small camera increments during sampling lets a policy-gradient model handle extreme-view video generation without paired large-motion training data.

arxiv:2605.16937 v1 · 2026-05-16 · 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{HUNVO5KQQWMUFOYZBSABM2ACOA}

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

we propose Dynamic Extreme VIew Synthesis-GRPO (DEVIS-GRPO), a GRPO-based framework for trajectory-controlled video generation, the first online policy gradient method for extreme view video generation. Central to our approach is a novel sampling strategy: Accumulative Dynamic Extreme VIew Synthesis (ADEVIS), which achieves large-view camera motions by progressively accumulating small-view increments.

C2weakest assumption

Progressively accumulating small-view increments reliably produces high-quality large-view motions and increased sampling diversity without introducing artifacts that the multi-level reward cannot filter, eliminating the need for expensive paired large-view videos.

C3one line summary

DEVIS-GRPO applies online policy gradients with an accumulative small-to-large view sampling strategy and multi-level rewards to improve trajectory-controlled extreme view video generation, reporting gains on Kubric-4D, iPhone, and DL3DV datasets.

References

79 extracted · 79 resolved · 14 Pith anchors

[1] Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages =

[2] Cami2v: Camera-controlled image-to-video diffu- sion model

[3] Lindell and Sergey Tulyakov , booktitle = 2025

[4] CameraCtrl: Enabling Camera Control for Text-to-Video Generation · arXiv:2404.02101

[5] Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages =

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-20T00:03:31.822922Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

3d1b577550859942bb190c801668027001d05739d5da11c4d5d3c2bfacf1b265

Aliases

arxiv: 2605.16937 · arxiv_version: 2605.16937v1 · doi: 10.48550/arxiv.2605.16937 · pith_short_12: HUNVO5KQQWMU · pith_short_16: HUNVO5KQQWMUFOYZ · pith_short_8: HUNVO5KQ

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/HUNVO5KQQWMUFOYZBSABM2ACOA \
  | 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: 3d1b577550859942bb190c801668027001d05739d5da11c4d5d3c2bfacf1b265

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "ed21ecbe4c16978bc4cc0c4b52444d042d443d967df93a45ac6d8bce5c85130f",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-05-16T11:14:18Z",
    "title_canon_sha256": "33bb592aa691709011a3424ff6fa021ffd3a630b83d0994051e6aae9fffe466d"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.16937",
    "kind": "arxiv",
    "version": 1
  }
}