pith:W6SUU52J
Contextualized Visual Personalization in Vision-Language Models
CoViP trains vision-language models to link new images to a user's past visual experiences for personalized responses.
arxiv:2602.03454 v3 · 2026-02-03 · cs.CV
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{W6SUU52JXO5WYVMSTJ5FXKIXLD}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
CoViP not only improves personalized image captioning but also yields holistic gains across downstream personalization tasks.
That the reinforcement-learning post-training and caption-augmented generation cause the model to genuinely use visual context rather than learned textual patterns, and that the introduced diagnostic evaluations fully rule out shortcut solutions.
CoViP is a unified framework that improves vision-language models' personalized image captioning and downstream tasks through RL-based post-training while introducing diagnostics to confirm visual context usage.
Formal links
Cited by
Receipt and verification
| First computed | 2026-05-20T01:05:07.777211Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
b7a54a7749bbbb6c55929a7a5ba91758d49647ed06f7f643c34811a8dace9833
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/W6SUU52JXO5WYVMSTJ5FXKIXLD \
| 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: b7a54a7749bbbb6c55929a7a5ba91758d49647ed06f7f643c34811a8dace9833
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "19463fcc26754d19928e3026e2c322609af7ebc0165db5ba1af0a1b1300f97eb",
"cross_cats_sorted": [],
"license": "http://creativecommons.org/licenses/by/4.0/",
"primary_cat": "cs.CV",
"submitted_at": "2026-02-03T12:21:26Z",
"title_canon_sha256": "d3378c8dcc208a8f25d0727da31c2752b7b1c4dfdb48d5e27fdb2f42bde6e5fd"
},
"schema_version": "1.0",
"source": {
"id": "2602.03454",
"kind": "arxiv",
"version": 3
}
}