Pith Number

pith:LVPWXSGY

pith:2026:LVPWXSGYEVADWFCGDBQYX7ZL5L

not attested not anchored not stored refs resolved

SpatialReward: Bridging the Perception Gap in Online RL for Image Editing via Explicit Spatial Reasoning

Bin Wen, Changyi Liu, Fan Yang, Han Li, Haonan Fan, Hongyang Wei, Jiankang Chen, Kaiyu Jiang, Kaiyu Tang, Shuo Yang, Tianke Zhang, Tingting Gao, Wei Chen, Yancheng Long, Yankai Yang

Anchoring rewards to predicted edit regions closes the perception gap in image editing RL

arxiv:2602.07458 v4 · 2026-02-07 · 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{LVPWXSGYEVADWFCGDBQYX7ZL5L}

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

SpatialReward serves as a robust signal in online RL, boosting OmniGen2 by +0.90 on GEdit-Bench--surpassing the leading discriminative model and doubling the gain of GPT-4.1 (+0.45).

C2weakest assumption

That predicting edit regions and anchoring reasoning to them reliably grounds semantic judgments in pixel-level evidence without the prediction step introducing new errors that offset the gains.

C3one line summary

SpatialReward is a new reward model that grounds image edit evaluations in pixel-level spatial reasoning on predicted regions, achieving SOTA on benchmarks and doubling RL gains for OmniGen2.

References

29 extracted · 29 resolved · 0 Pith anchors

[1] • Good: All edit operations in the instruction are perfectly executed

[2] •Good: High fidelity, no visible artifacts

[3] Overall AestheticsA holistic assessment of the image’s visual appeal and harmony. annotators are instructed to judge solely based on the visual outcome: •Good: Visually pleasing, professional-looking 2017

[4] Reward Model Interpretation(Section C.1): We analyze the internal attention mechanisms of SpatialReward to verify its reasoning logic and explain the metrics used for quantitative diagnosis

[5] Policy Generation Results(Section C.2): We showcase additional qualitative comparisons of the downstream policy model (OmniGen2) trained via Online RL, demonstrating the effectiveness of our reward si

Formal links

2 machine-checked theorem links

Cited by

1 paper in Pith

OpenSpatial: A Principled Data Engine for Empowering Spatial Intelligence

Receipt and verification

First computed	2026-05-18T02:44:31.312988Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

5d5f6bc8d825403b144618618bff2bead30ec8155b6065932ba15832e9e355a8

Aliases

arxiv: 2602.07458 · arxiv_version: 2602.07458v4 · doi: 10.48550/arxiv.2602.07458 · pith_short_12: LVPWXSGYEVAD · pith_short_16: LVPWXSGYEVADWFCG · pith_short_8: LVPWXSGY

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/LVPWXSGYEVADWFCGDBQYX7ZL5L \
  | 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: 5d5f6bc8d825403b144618618bff2bead30ec8155b6065932ba15832e9e355a8

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "858e2f5a747c75d8bd75320f8f48e4fc3f0a7e5cdc6e2a6ef6e7b75004144ea8",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-02-07T09:23:34Z",
    "title_canon_sha256": "1b5d4237dc6f688e08f86957fa5d701d53ea0fa79ea55d2fde65d83d45a38ce0"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2602.07458",
    "kind": "arxiv",
    "version": 4
  }
}