pith:KQGQ3HUF
SAGE: Sign-Adaptive Gradient for Memory-Efficient LLM Optimization
SAGE replaces AdamW in hybrid LLM optimizers by adding a bounded per-dimension scale that tames embedding-layer gradients, cutting memory while raising perplexity performance.
arxiv:2604.07663 v2 · 2026-04-09 · cs.LG
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{KQGQ3HUFVD5QJXTDZURY2OGC6I}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
On Llama models up to 1.3B parameters, our SAGE-based hybrid achieves new state-of-the-art perplexity, outperforming all baselines, including SinkGD hybrid, while significantly reducing optimizer state memory.
That the high-variance, sparse gradients in the embedding layer can be sufficiently controlled by a single per-dimension scale provably bounded by 1.0 without reintroducing the instability that forced prior methods back to AdamW.
SAGE replaces AdamW in memory-efficient LLM hybrids with a Lion-style sign update plus a provably bounded O(d) adaptive scale, delivering SOTA perplexity on Llama-1.3B while cutting optimizer-state memory.
References
Formal links
Receipt and verification
| First computed | 2026-06-26T01:15:51.382011Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
540d0d9e85a8fb04de63cd238d38c2f2182086c8d712ff6ee20cb40ffaa9e74d
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/KQGQ3HUFVD5QJXTDZURY2OGC6I \
| 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: 540d0d9e85a8fb04de63cd238d38c2f2182086c8d712ff6ee20cb40ffaa9e74d
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "5a8f77f162041ef977b19efbf3b0e41f621029f50f0f7f23836c3f9a532a3e23",
"cross_cats_sorted": [],
"license": "http://creativecommons.org/licenses/by/4.0/",
"primary_cat": "cs.LG",
"submitted_at": "2026-04-09T00:07:38Z",
"title_canon_sha256": "aa16abe577c9cc3b4f2638308385ac68083532f098b0c3f4f4a2f80d4b295b19"
},
"schema_version": "1.0",
"source": {
"id": "2604.07663",
"kind": "arxiv",
"version": 2
}
}