pith:OXRYN2TO
Diet Your LLM: Dimension-wise Global Pruning of LLMs via Merging Task-specific Importance Score
DIET prunes LLM dimensions by merging per-task activation scores into one global mask without any training.
arxiv:2603.23985 v3 · 2026-03-25 · cs.LG
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{OXRYN2TOQOYECO5M3HGTWP3DQL}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
at 20% sparsity on Gemma-2 2B, DIET achieves near 10% average accuracy improvement, compared to previous state-of-the-art structured pruning methods.
that activation magnitudes computed from only 100 samples per task are sufficient to produce a reliable global importance ranking that generalizes across unseen tasks and inputs.
DIET prunes LLMs dimension-wise by merging task-specific activation importance scores via majority voting on 100 samples per task, yielding up to 10% accuracy gains over prior structured methods at 20% sparsity on Gemma-2 models.
Formal links
Receipt and verification
| First computed | 2026-05-27T01:05:53.464958Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
75e386ea6e83b0413bacd9cd3b3f6382e830d6c2bd36482b94a3b5f90829bd03
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/OXRYN2TOQOYECO5M3HGTWP3DQL \
| 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: 75e386ea6e83b0413bacd9cd3b3f6382e830d6c2bd36482b94a3b5f90829bd03
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "902a117a272fcd9238d3d408303513c06388bdfce6a2c2a3b59da6a271578778",
"cross_cats_sorted": [],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "cs.LG",
"submitted_at": "2026-03-25T06:28:58Z",
"title_canon_sha256": "bb50e61da520dd9eb49619541867382880f6005196f6b1f634af04c91e99cee8"
},
"schema_version": "1.0",
"source": {
"id": "2603.23985",
"kind": "arxiv",
"version": 3
}
}