pith:KZPVITFK
GRPO-TTA: Test-Time Visual Tuning for Vision-Language Models via GRPO-Driven Reinforcement Learning
GRPO-TTA applies GRPO to test-time visual tuning of vision-language models via group-wise policy optimization on unlabeled class candidates, outperforming prior TTA methods especially under natural distribution shifts.
arxiv:2605.03403 v2 · 2026-05-05 · cs.CV · cs.LG
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{KZPVITFKGLIULSSP4T46XSM6WS}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
GRPO-TTA consistently outperforms existing test-time adaptation methods, with notably larger performance gains under natural distribution shifts.
That constructing output groups by sampling top-K class candidates from CLIP similarity distributions enables effective probability-driven optimization without ground-truth labels, and that the designed alignment and dispersion rewards guide effective visual encoder tuning at test time.
GRPO-TTA applies GRPO to test-time visual tuning of vision-language models via group-wise policy optimization on unlabeled class candidates, outperforming prior TTA methods especially under natural distribution shifts.
Receipt and verification
| First computed | 2026-06-02T02:04:53.599649Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
565f544caa32d145ca4fe4f9ebc99eb48ecd132ea4c99eab95eded35b501572a
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/KZPVITFKGLIULSSP4T46XSM6WS \
| 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: 565f544caa32d145ca4fe4f9ebc99eb48ecd132ea4c99eab95eded35b501572a
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "d6dc7942558f67369e1d00e8f14e57f14270fc3fff6b7780143eb5e23359686b",
"cross_cats_sorted": [
"cs.LG"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "cs.CV",
"submitted_at": "2026-05-05T06:23:20Z",
"title_canon_sha256": "d74a1d201abbf5de94cf02c297d491c4464bc71ece9b99f4de06303adb1e1d4a"
},
"schema_version": "1.0",
"source": {
"id": "2605.03403",
"kind": "arxiv",
"version": 2
}
}