Pith Number

pith:KD4RGC2W

pith:2026:KD4RGC2WYDQ4SPX5M3DAHYSNZ6

not attested not anchored not stored refs resolved

Edit-Compass & EditReward-Compass: A Unified Benchmark for Image Editing and Reward Modeling

Xiaoling Gu, Xinyu Liu, Xuanyu Zhu, Xuehai Bai, Yang Shi, Yifan Dai, Yi-Fan Zhang, Yiyan Ji, Yuanxing Zhang, Yuran Wang

Edit-Compass and EditReward-Compass supply a unified benchmark with 2,388 fine-grained instances and 2,251 realistic preference pairs to evaluate image editing models and reward models more faithfully than prior tests.

arxiv:2605.13062 v1 · 2026-05-13 · 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{KD4RGC2WYDQ4SPX5M3DAHYSNZ6}

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

Edit-Compass and EditReward-Compass address limitations in existing benchmarks by providing fine-grained multidimensional evaluation and realistic RL scenarios for image editing models and reward models.

C2weakest assumption

The assumption that the 2,388 annotated instances and 2,251 preference pairs accurately capture human judgment and practical RL optimization settings without introducing annotation biases or unrealistic preferences.

C3one line summary

Edit-Compass and EditReward-Compass are new unified benchmarks for fine-grained image editing evaluation and realistic reward modeling in reinforcement learning optimization.

References

103 extracted · 103 resolved · 17 Pith anchors

[1] Qwen3-VL Technical Report 2025 · arXiv:2511.21631

[2] Instructpix2pix: Learning to follow image editing instructions 2023

[3] HiDream-I1: A High-Efficient Image Generative Foundation Model with Sparse Diffusion Transformer 2025 · arXiv:2505.22705

[4] Opengpt-4o-image: A comprehensive dataset for advanced image generation and editing.arXiv preprint arXiv:2509.24900, 2025b 2025

[5] Emu3.5: Native Multimodal Models are World Learners 2025 · arXiv:2510.26583

Cited by

1 paper in Pith

Artifact-Bench: Evaluating MLLMs on Detecting and Assessing the Artifacts of AI-Generated Videos

Receipt and verification

First computed	2026-05-18T03:08:59.062603Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

50f9130b56c0e1c93efd66c603e24dcf9b062cee4701948bcc16950cdb0e3e52

Aliases

arxiv: 2605.13062 · arxiv_version: 2605.13062v1 · doi: 10.48550/arxiv.2605.13062 · pith_short_12: KD4RGC2WYDQ4 · pith_short_16: KD4RGC2WYDQ4SPX5 · pith_short_8: KD4RGC2W

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/KD4RGC2WYDQ4SPX5M3DAHYSNZ6 \
  | 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: 50f9130b56c0e1c93efd66c603e24dcf9b062cee4701948bcc16950cdb0e3e52

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "de80782cc439a6d830825fd9888a79fcd677a2e8a4201643a67bf7c83605ea39",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-05-13T06:33:54Z",
    "title_canon_sha256": "dbfe075de3e94432600587951695601db9ad41f696c1e3474419cf7e357101dd"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13062",
    "kind": "arxiv",
    "version": 1
  }
}