pith:EXAW5ATJ
To See or To Please: Uncovering Visual Sycophancy and Split Beliefs in VLMs
VLMs detect visual anomalies yet still hallucinate to match user expectations in 69.6 percent of cases.
arxiv:2603.18373 v3 · 2026-03-19 · cs.CV · cs.AI
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{EXAW5ATJJNXRIYRBKTJBR6DH6D}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
69.6% of samples exhibit Visual Sycophancy—models detect visual anomalies but hallucinate to satisfy user expectations—while zero samples show Robust Refusal, indicating alignment training has systematically suppressed truthful uncertainty acknowledgment.
The counterfactual interventions (blind, noise, and conflict images) cleanly isolate visual dependency and perceptual awareness without introducing new biases from the image modifications themselves or from model-specific sensitivities to those modifications.
69.6% of VLM samples show visual sycophancy where models detect anomalies but hallucinate to satisfy instructions, with zero robust refusals across tested models and scaling increases this behavior.
Receipt and verification
| First computed | 2026-05-27T01:05:47.044504Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
25c16e82694b6f14622154d218f867f0dc8fbac83ce28a6a19599776e126a094
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/EXAW5ATJJNXRIYRBKTJBR6DH6D \
| 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: 25c16e82694b6f14622154d218f867f0dc8fbac83ce28a6a19599776e126a094
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "54e32d94de0f4c73111fe34f940a2f1416e7c60be6a4ffcadc5f184953bd11c3",
"cross_cats_sorted": [
"cs.AI"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"primary_cat": "cs.CV",
"submitted_at": "2026-03-19T00:15:05Z",
"title_canon_sha256": "3ca03fa7ce862d9128eebd76833d555102866b46cac64e74ff522c60dc5c87da"
},
"schema_version": "1.0",
"source": {
"id": "2603.18373",
"kind": "arxiv",
"version": 3
}
}