Pith. sign in
Pith Number

pith:LZNQEFNK

pith:2026:LZNQEFNKJVTKXR5OYXKCWZ7TLX
not attested not anchored not stored refs pending

LLMs Meet Isolation Kernel: Lightweight, Learning-free Binary Embeddings for Fast Retrieval

Cam-Tu Nguyen, Kai Ming Ting, Yang Xu, Zhibo Zhang

Isolation Kernel converts LLM embeddings into compact binary codes that deliver up to 16 times lower memory use and 16.7 times faster retrieval with comparable accuracy.

arxiv:2601.09159 v4 · 2026-01-14 · cs.IR

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{LZNQEFNKJVTKXR5OYXKCWZ7TLX}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp
2 Internet Archive
3 Author claim open · sign in to claim
4 Citations open
5 Replications open
Portable graph bundle live · download bundle · merged state
The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

Experiments on multiple text retrieval datasets demonstrate that IKE offers up to 16.7x faster retrieval and 16x lower memory usage than the original LLM embeddings, while maintaining comparable accuracy. Theoretically, we show that IKE works because it satisfies four essential criteria for effective binary hashing that other methods do not possess.

C2weakest assumption

That the Isolation Kernel applied to LLM embeddings preserves enough semantic similarity information to avoid meaningful accuracy loss on downstream retrieval tasks, without any task-specific tuning or validation of the kernel parameters.

C3one line summary

IKE converts LLM embeddings into binary codes via Isolation Kernel for up to 16.7x faster retrieval and 16x lower memory with comparable accuracy.

Receipt and verification
First computed 2026-07-08T01:19:09.564661Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

5e5b0215aa4d66abc7aec5d42b67f35dead430bc018d4484dbda69359fc3c9d8

Aliases

arxiv: 2601.09159 · arxiv_version: 2601.09159v4 · doi: 10.48550/arxiv.2601.09159 · pith_short_12: LZNQEFNKJVTK · pith_short_16: LZNQEFNKJVTKXR5O · pith_short_8: LZNQEFNK
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/LZNQEFNKJVTKXR5OYXKCWZ7TLX \
  | 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: 5e5b0215aa4d66abc7aec5d42b67f35dead430bc018d4484dbda69359fc3c9d8
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "f0857584fdc40e90fa2727875040cdfff66670933e7ce17180a984ca41200c29",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.IR",
    "submitted_at": "2026-01-14T04:54:09Z",
    "title_canon_sha256": "dfecff787b8312f1b2c6cf22ea3ea12d8d7565cf0eb401a4fa16bd44c298a390"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2601.09159",
    "kind": "arxiv",
    "version": 4
  }
}