Pith Number

pith:WXZVI4V2

pith:2025:WXZVI4V2IUP6ELAT6F3F2RTVH6

not attested not anchored not stored refs pending

How much do language models memorize?

Alexander M. Rush, Chawin Sitawarin, Chuan Guo, G. Edward Suh, John X. Morris, Kamalika Chaudhuri, Narine Kokhlikyan, Saeed Mahloujifar

arxiv:2505.24832 v3 · 2025-05-30 · cs.CL

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{WXZVI4V2IUP6ELAT6F3F2RTVH6}

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.

Cited by

19 papers in Pith

Towards Mechanistically Understanding Why Memorized Knowledge Fails to Generalize in Large Language Model Finetuning

PromptMN: Pseudo Prompting Language

Detecting Functional Memorization in Code Language Models

Fast Unlearning at Scale via Margin Self-Correction

Slower Generalization, Faster Memorization: A Sweet Spot in Algorithmic Learning

Receipt and verification

First computed	2026-07-05T11:23:29.000310Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

b5f35472ba451fe22c13f1765d46753fbf1d99dca422760385fccdf2b7e27784

Aliases

arxiv: 2505.24832 · arxiv_version: 2505.24832v3 · doi: 10.48550/arxiv.2505.24832 · pith_short_12: WXZVI4V2IUP6 · pith_short_16: WXZVI4V2IUP6ELAT · pith_short_8: WXZVI4V2

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/WXZVI4V2IUP6ELAT6F3F2RTVH6 \
  | 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: b5f35472ba451fe22c13f1765d46753fbf1d99dca422760385fccdf2b7e27784

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "bc4a9903a996f65d446d40088c32aa8de68b925250becd95fcc80ed701185193",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2025-05-30T17:34:03Z",
    "title_canon_sha256": "8969aba3fbbe132a80cbafc42ebf0d4e99391987a48849647fef45ba4d99804c"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2505.24832",
    "kind": "arxiv",
    "version": 3
  }
}