pith:KA7ICOTD
High-Entropy Tokens as Multimodal Failure Points in Vision-Language Models
A small share of high-entropy tokens during generation concentrates most adversarial influence in vision-language models.
arxiv:2512.21815 v3 · 2025-12-26 · cs.CV · cs.LG
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{KA7ICOTDHZL4TITMUSLHMEBL3D}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
a small fraction (around 20%) of high-entropy tokens, in the evaluated representative open-source VLMs with diverse architectures, concentrates a disproportionate share of adversarial influence during autoregressive generation
That high-entropy tokens can be identified reliably during generation and that concentrating perturbations on them produces comparable semantic degradation to global attacks without requiring post-hoc selection or model-specific tuning that would invalidate the transferability claim.
High-entropy tokens act as concentrated multimodal failure points in VLMs, enabling sparse Entropy-Guided Attacks that achieve 93-95% success and 30-38% harmful rates with cross-model transfer.
Formal links
Cited by
Receipt and verification
| First computed | 2026-05-26T01:03:21.256922Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
503e813a633e57c9a26ca49676102bd8f47736e4df6975fe74e80cc419cbfb1e
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/KA7ICOTDHZL4TITMUSLHMEBL3D \
| 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: 503e813a633e57c9a26ca49676102bd8f47736e4df6975fe74e80cc419cbfb1e
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "87fcc8f582da8f4dfb1a9eb1534ecb945a80f771a6a8095ecc6320410f5b1e7f",
"cross_cats_sorted": [
"cs.LG"
],
"license": "http://creativecommons.org/licenses/by-nc-nd/4.0/",
"primary_cat": "cs.CV",
"submitted_at": "2025-12-26T01:01:25Z",
"title_canon_sha256": "48f6abf74e09a01bb218402c0ee098c18be6a07cd82f4b547d10d39545aea293"
},
"schema_version": "1.0",
"source": {
"id": "2512.21815",
"kind": "arxiv",
"version": 3
}
}