Pith Number

pith:6DVG7MV3

pith:2019:6DVG7MV3CDNLUNQDZKFYQHEVL4

not attested not anchored not stored refs resolved

Language Models as Knowledge Bases?

Alexander H. Miller, Anton Bakhtin, Fabio Petroni, Patrick Lewis, Sebastian Riedel, Tim Rockt\"aschel, Yuxiang WU

Pretrained language models already contain relational knowledge that can be accessed through fill-in-the-blank queries without any fine-tuning.

arxiv:1909.01066 v2 · 2019-09-03 · 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{6DVG7MV3CDNLUNQDZKFYQHEVL4}

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

without fine-tuning, BERT contains relational knowledge competitive with traditional NLP methods that have some access to oracle knowledge

C2weakest assumption

That cloze-statement accuracy directly measures stored factual knowledge rather than surface-level statistical patterns or memorization of training co-occurrences.

C3one line summary

BERT stores relational knowledge extractable via cloze queries without fine-tuning and matches supervised baselines on open-domain QA tasks.

References

300 extracted · 300 resolved · 41 Pith anchors

[1] Overview of the

[2] Learning and evaluating general linguistic intelligence 1901 · arXiv:1901.11373

[3] Lample, Guillaume and Ott, Myle and Conneau, Alexis and Denoyer, Ludovic and Ranzato, Marc'Aurelio , month = nov, year =. Phrase-. Proceedings of the 2018 2018

[4] BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding 2018 · arXiv:1810.04805

[5] Transactions of the Association of Computational Linguistics , author =

Formal links

2 machine-checked theorem links

Cited by

24 papers in Pith

Identifying the Achilles' Heel: An Iterative Method for Dynamically Uncovering Factual Errors in Large Language Models

CounterBench: Evaluating and Improving Counterfactual Reasoning in Large Language Models

Adapting Vision-Language Models for Neutrino Event Classification in High-Energy Physics

AtlasKV: Augmenting LLMs with Billion-Scale Knowledge Graphs in 20GB VRAM

Understanding New-Knowledge-Induced Factual Hallucinations in LLMs: Analysis and Interpretation

Receipt and verification

First computed	2026-05-17T23:38:48.128807Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

f0ea6fb2bb10daba3603ca8b881c955f1db39e171f12f9fa973aed36e815f0c0

Aliases

arxiv: 1909.01066 · arxiv_version: 1909.01066v2 · doi: 10.48550/arxiv.1909.01066 · pith_short_12: 6DVG7MV3CDNL · pith_short_16: 6DVG7MV3CDNLUNQD · pith_short_8: 6DVG7MV3

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/6DVG7MV3CDNLUNQDZKFYQHEVL4 \
  | 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: f0ea6fb2bb10daba3603ca8b881c955f1db39e171f12f9fa973aed36e815f0c0

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "b4b9d25078db0085f26c4e02c555f1804072205847706550f402b5812b96137f",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2019-09-03T11:11:08Z",
    "title_canon_sha256": "806066cd870d90a814cc6da54f47e7c2bdbe38284f76cd67d7056b51a66920b5"
  },
  "schema_version": "1.0",
  "source": {
    "id": "1909.01066",
    "kind": "arxiv",
    "version": 2
  }
}