Pith Number

pith:FJ7SS4UW

pith:2025:FJ7SS4UWAI7H4H3KT5VTQMNL4J

not attested not anchored not stored refs resolved

LIMO: Less is More for Reasoning

Ethan Chern, Pengfei Liu, Shijie Xia, Yang Xiao, Yixin Ye, Zhen Huang

Sophisticated mathematical reasoning emerges in large language models from only a few strategically designed examples.

arxiv:2502.03387 v3 · 2025-02-05 · cs.CL · cs.AI

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{FJ7SS4UWAI7H4H3KT5VTQMNL4J}

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

sophisticated mathematical reasoning can emerge with only a few examples... through minimal but strategically designed demonstrations of cognitive processes.

C2weakest assumption

The foundation model has already comprehensively encoded the relevant domain knowledge during pre-training, and the small set of post-training examples functions as effective cognitive templates.

C3one line summary

LIMO achieves 63.3% on AIME24 and 95.6% on MATH500 via supervised fine-tuning on roughly 1% of the data used by prior models, supporting the claim that minimal strategic examples suffice when pre-training has already encoded domain knowledge.

References

299 extracted · 299 resolved · 38 Pith anchors

[1] Attention is All you Need , url =

[2] OpenAI o1 System Card , author=. 2024 , eprint= 2024

[3] SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training , author=. 2025 , eprint= 2025

[4] Benchmarking Benchmark Leakage in Large Language Models , author=. 2024 , eprint= 2024

[5] MathPile: A Billion-Token-Scale Pretraining Corpus for Math , author=. 2024 , eprint= 2024

Formal links

2 machine-checked theorem links

Cited by

34 papers in Pith

TPMM-DPO: Trajectory-aware Preference-guided Model Merging for Iterative Direct Preference Optimization

Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems

A Survey of Scaling in Large Language Model Reasoning

Learning to Foresee: Unveiling the Unlocking Efficiency of On-Policy Distillation

A Model Can Help Itself: Reward-Free Self-Training for LLM Reasoning

Receipt and verification

First computed	2026-05-17T23:38:15.428774Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

2a7f297296023e7e1f6a9f6b3831abe26735212df6938c89631204fb7806d50d

Aliases

arxiv: 2502.03387 · arxiv_version: 2502.03387v3 · doi: 10.48550/arxiv.2502.03387 · pith_short_12: FJ7SS4UWAI7H · pith_short_16: FJ7SS4UWAI7H4H3K · pith_short_8: FJ7SS4UW

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/FJ7SS4UWAI7H4H3KT5VTQMNL4J \
  | 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: 2a7f297296023e7e1f6a9f6b3831abe26735212df6938c89631204fb7806d50d

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "348bddcd4634c3d1407c8e87fe25abf1925e655f5771c0cad444c8e714a16eaf",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2025-02-05T17:23:45Z",
    "title_canon_sha256": "b2893585327c5dea96c02e8be12b73ff54b659b0d8de6de73f0599ec12965715"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2502.03387",
    "kind": "arxiv",
    "version": 3
  }
}