pith:63QS6NZH
Topology-Aware Layer Pruning for Large Vision-Language Models
Persistent homology on layer point clouds guides pruning to retain critical transitions in vision-language models.
arxiv:2604.16502 v2 · 2026-04-14 · cs.CV
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{63QS6NZHBNB2J6ZF5GA4QFYUVE}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
Extensive experiments on diverse multimodal benchmarks demonstrate that the proposed framework consistently outperforms existing pruning methods across a wide range of sparsity ratios.
That representing layer-wise hidden states as point clouds and quantifying their evolution with simplicial complexes and zigzag persistent homology accurately captures global and dynamic representational transitions better than local similarity metrics or static proxies.
A topology-aware pruning framework models layer representation evolution in LVLMs via simplicial complexes and zigzag persistent homology to enable adaptive removal of layers while outperforming existing methods on multimodal benchmarks.
Receipt and verification
| First computed | 2026-06-05T01:14:38.872545Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
f6e12f37270b43a4fb25e981c81714a92dff83e212367a7576c332752d6cb2ed
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/63QS6NZHBNB2J6ZF5GA4QFYUVE \
| 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: f6e12f37270b43a4fb25e981c81714a92dff83e212367a7576c332752d6cb2ed
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "a8f0eaf27c35ec2b67f410da0c14fece81a52783ea0040d90adc5720abf044a4",
"cross_cats_sorted": [],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "cs.CV",
"submitted_at": "2026-04-14T14:36:53Z",
"title_canon_sha256": "f11f8d27992ce8b86e9a3e4ccc015133a30ff5f475b9e36ec9c7f741a984560f"
},
"schema_version": "1.0",
"source": {
"id": "2604.16502",
"kind": "arxiv",
"version": 2
}
}