pith:OGHK4LF5
Leveraging Verifier-Based Reinforcement Learning in Image Editing
A chain-of-thought verifier that decomposes editing instructions into principles delivers better rewards than general vision-language models.
arxiv:2604.27505 v2 · 2026-04-30 · cs.CV
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{OGHK4LF5S5YDNR2LIK24ONZEQF}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
our Edit-RRM surpasses powerful VLMs such as Seed-1.5-VL and Seed-1.6-VL as an editing-specific reward model, and we observe a clear scaling trend, with performance consistently improving from 3B to 7B parameters. Moreover, Edit-R1 delivers gains to editing models like FLUX.1-kontext
That breaking instructions into principles and aggregating CoT checks produces unbiased, generalizable rewards across all editing tasks without introducing new failure modes or requiring task-specific tuning that was not captured in the human preference data.
Edit-R1 trains a CoT-based reasoning reward model with GCPO and uses it to boost image editing performance over VLMs and models like FLUX.1-kontext via GRPO.
Receipt and verification
| First computed | 2026-05-21T01:04:26.577653Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
718eae2cbd977036c74b42b5c73724817838315c6900e81060bc23a0325448c2
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/OGHK4LF5S5YDNR2LIK24ONZEQF \
| 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: 718eae2cbd977036c74b42b5c73724817838315c6900e81060bc23a0325448c2
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "c61a98136ad8842d44ef0fa81fadc0b2fba34e65235a32294bf13d181b9c5a43",
"cross_cats_sorted": [],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "cs.CV",
"submitted_at": "2026-04-30T06:54:39Z",
"title_canon_sha256": "0b22305dd553af6e1129f96b32d58c87810b78b6f73175455820a852d4036d28"
},
"schema_version": "1.0",
"source": {
"id": "2604.27505",
"kind": "arxiv",
"version": 2
}
}