Pith Number

pith:2CJ6UHSW

pith:2023:2CJ6UHSWSOT6LJLF6BUWCSAXSE

not attested not anchored not stored refs resolved

Let's Verify Step by Step

Bowen Baker, Harri Edwards, Hunter Lightman, Ilya Sutskever, Jan Leike, John Schulman, Karl Cobbe, Teddy Lee, Vineet Kosaraju, Yura Burda

Process supervision outperforms outcome supervision for training models to solve MATH problems.

arxiv:2305.20050 v1 · 2023-05-31 · cs.LG · cs.AI · 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{2CJ6UHSWSOT6LJLF6BUWCSAXSE}

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

we conduct our own investigation, finding that process supervision significantly outperforms outcome supervision for training models to solve problems from the challenging MATH dataset. Our process-supervised model solves 78% of problems from a representative subset of the MATH test set.

C2weakest assumption

Human-provided step-level labels are consistent, unbiased, and sufficient to train a generalizable process reward model; the chosen subset is representative of the full MATH test distribution.

C3one line summary

Process supervision significantly outperforms outcome supervision for training models on the MATH dataset, achieving 78% accuracy on a representative test subset with active learning and a released 800k step-label dataset.

References

22 extracted · 22 resolved · 15 Pith anchors

[1] A General Language Assistant as a Laboratory for Alignment · arXiv:2112.00861

[2] Sparks of Artificial General Intelligence: Early experiments with GPT-4 · arXiv:2303.12712

[3] Training Verifiers to Solve Math Word Problems · arXiv:2110.14168

[4] 16 Solving math word problems with process- and outcome-based feedback A

[5] Reinforcement Learning with a Corrupted Reward Channel · arXiv:1705.08417

Formal links

2 machine-checked theorem links

Cited by

307 papers in Pith

LLM-as-a-Verifier: A General-Purpose Verification Framework

Weak-to-Strong Generalization via Direct On-Policy Distillation

R2IF: Aligning Reasoning with Decisions via Composite Rewards for Interpretable LLM Function Calling

InvestPhilBench: A Multi-Layer Dynamic Benchmark for Evaluating Large Language Model Procedural Reasoning in Expert Investment Philosophy

Calibration Drift Under Reasoning: How Chain-of-Thought Budgets Induce Overconfidence in Large Language Models

Receipt and verification

First computed	2026-07-05T06:16:06.287282Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

d093ea1e5693a7e5a565f069614817913ebf6c86ce0b449d728cfe2e94a3e3eb

Aliases

arxiv: 2305.20050 · arxiv_version: 2305.20050v1 · doi: 10.48550/arxiv.2305.20050 · pith_short_12: 2CJ6UHSWSOT6 · pith_short_16: 2CJ6UHSWSOT6LJLF · pith_short_8: 2CJ6UHSW

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/2CJ6UHSWSOT6LJLF6BUWCSAXSE \
  | 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: d093ea1e5693a7e5a565f069614817913ebf6c86ce0b449d728cfe2e94a3e3eb

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "71df4e49475483a0f241e9e6d9a7ad2b7912df9b0691b08f25ded42a75b81ea3",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.CL"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2023-05-31T17:24:00Z",
    "title_canon_sha256": "71855739b2da08ea015cd79f0aebab131d8019e1ccbbf9db8ac4a714de4efa73"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2305.20050",
    "kind": "arxiv",
    "version": 1
  }
}