Pith Number

pith:URMW7OAT

pith:2024:URMW7OATDXGQEQCYXI2WZDOTW6

not attested not anchored not stored refs pending

Sabotage Evaluations for Frontier Models

Buck Shlegeris, Cem Anil, David Duvenaud, Deep Ganguli, Eric Christiansen, Esin Durmus, Ethan Perez, Evan Hubinger, Holden Karnofsky, Jai Srivastav, Jared Kaplan, Joe Benton, Misha Wagner, Roger Grosse, Samuel R. Bowman, Shauna Kravec

arxiv:2410.21514 v1 · 2024-10-28 · cs.LG · cs.AI · cs.CY

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{URMW7OATDXGQEQCYXI2WZDOTW6}

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

17 papers in Pith

Comparing Linear Probes with Mahalanobis Cosine Similarity

Decoding Hidden Deception in Reasoning LLMs: Activation Explainers for Deception Auditing

Generalization Hacking: Models Can Game Reinforcement Learning by Preventing Behavioral Generalization

The Distributed Detectability Band Against Marginal-Preserving Attacks

Sycophancy Towards Researchers Drives Performative Misalignment

Receipt and verification

First computed	2026-07-05T09:27:29.324073Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

a4596fb8131dcd024058ba356c8dd3b79cc54da04b2a5bb13ae997e35a596036

Aliases

arxiv: 2410.21514 · arxiv_version: 2410.21514v1 · doi: 10.48550/arxiv.2410.21514 · pith_short_12: URMW7OATDXGQ · pith_short_16: URMW7OATDXGQEQCY · pith_short_8: URMW7OAT

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/URMW7OATDXGQEQCYXI2WZDOTW6 \
  | 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: a4596fb8131dcd024058ba356c8dd3b79cc54da04b2a5bb13ae997e35a596036

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "c851fa598fcb8e3bbd35fe7733e140c0e623d678f64c49aaee225c853e44f211",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.CY"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2024-10-28T20:34:51Z",
    "title_canon_sha256": "d6e7f142cc97eb74cb2f6111972764534052ee6be81abdb9678a396876304fa3"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2410.21514",
    "kind": "arxiv",
    "version": 1
  }
}