Pith Number

pith:4GFH6LTS

pith:2026:4GFH6LTS2T7JAFO2QNHEAN5OIN

not attested not anchored not stored refs resolved

ICP: Exploiting Instruction Correlation for Prefetching Irregular Memory Accesses

Buqing Xu, Ceyu Xu, Chenlu Miao, Mengming Li, Qijun Zhang, Shang Liu, Wenkai Li, Xiangfeng Sun, Yuan Xie, Zhiyao Xie

ICP prefetching exploits stable instruction correlations to handle irregular memory accesses more efficiently than address-based methods.

arxiv:2605.15645 v1 · 2026-05-15 · cs.AR

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{4GFH6LTS2T7JAFO2QNHEAN5OIN}

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

ICP outperforms the state-of-the-art temporal prefetcher Triangel by 14.0% and the indirect prefetcher DMP by 6.0%, while requiring only 2.1 KB of hardware storage, over three orders of magnitude smaller than temporal prefetchers.

C2weakest assumption

The assumption that instructions generating irregular addresses recur with stable data-dependency relationships even when the addresses themselves show weak or no temporal recurrence, enabling speculative computation of future accesses from correlated predecessors.

C3one line summary

ICP is a prefetcher that learns stable instruction correlations to speculatively compute future irregular memory accesses, outperforming Triangel by 14% and DMP by 6% with only 2.1 KB storage.

References

60 extracted · 60 resolved · 1 Pith anchors

[1] “Gate equivalent,” https://en.wikipedia.org/wiki/Gate equivalent

[2] “SPEC CPU 2006,” https://www.spec.org/cpu2006/ 2006

[3] Ghostminion: A strictness-ordered cache system for spectre mitigation, 2021

[4] Graph prefetching using data structure knowledge, 2016

[5] Software prefetching for indirect memory accesses, 2017

Receipt and verification

First computed	2026-05-20T00:01:09.841405Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

e18a7f2e72d4fe9015da834e4037ae4369bd4ccd15bde34b1b4836d92ae58261

Aliases

arxiv: 2605.15645 · arxiv_version: 2605.15645v1 · doi: 10.48550/arxiv.2605.15645 · pith_short_12: 4GFH6LTS2T7J · pith_short_16: 4GFH6LTS2T7JAFO2 · pith_short_8: 4GFH6LTS

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/4GFH6LTS2T7JAFO2QNHEAN5OIN \
  | 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: e18a7f2e72d4fe9015da834e4037ae4369bd4ccd15bde34b1b4836d92ae58261

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "f12e77c42a67208fbab7d42ea27c4f78216a9c14a17e8fcb1786a63cd83ff3db",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by-nc-sa/4.0/",
    "primary_cat": "cs.AR",
    "submitted_at": "2026-05-15T06:01:25Z",
    "title_canon_sha256": "1c2cfe19983f44b393057645d887de7cbbe4c3f1170fe464b72151f38c4c779c"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.15645",
    "kind": "arxiv",
    "version": 1
  }
}