Pith Number

pith:KOOKJRXX

pith:2026:KOOKJRXX2VNXKGV2JG46W5KAUF

not attested not anchored not stored refs resolved

LeanSearch v2: Global Premise Retrieval for Lean 4 Theorem Proving

Bin Dong, Bryan Dai, Guoxiong Gao, Jiedong Jiang, Jingda Xu, Peihao Wu, Yutong Wang, Zeming Sun

LeanSearch v2 recovers full premise sets for Lean 4 theorems at 46.1 percent accuracy on research benchmarks.

arxiv:2605.13137 v2 · 2026-05-13 · cs.IR · 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{KOOKJRXX2VNXKGV2JG46W5KAUF}

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

On a 69-query benchmark of research-level Mathlib theorems, reasoning mode recovers 46.1% of ground-truth premise groups within 10 retrieved candidates, outperforming strong reasoning retrieval systems (38.0%) and premise-selection baselines (9.3%). In a controlled downstream evaluation with a fixed prover loop, replacing alternative retrievers with LeanSearch v2 yields the highest proof success (20% vs. 16% for the next-best system and 4% without retrieval).

C2weakest assumption

The 69-query benchmark and the fixed prover loop are assumed to be representative of real-world Lean 4 usage; the paper does not report how performance changes when the prover loop or theorem distribution is altered.

C3one line summary

LeanSearch v2 recovers 46.1% of ground-truth premise groups for research-level Lean 4 theorems within 10 candidates and raises fixed-loop proof success to 20%.

References

32 extracted · 32 resolved · 8 Pith anchors

[1] Mm-bright: A multi-task multimodal benchmark for reasoning-intensive retrieval

[2] Aristotle: IMO-level Automated Theorem Proving · arXiv:2510.01346

[3] Leanexplore: A search engine for lean 4 declarations.CoRR, abs/2506.11085

[4] Seed-prover 1.5: Mastering undergraduate-level theorem proving via learning from experience

[5] Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities · arXiv:2507.06261

Formal links

1 machine-checked theorem link

Receipt and verification

First computed	2026-05-18T03:08:57.558427Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

539ca4c6f7d55b751aba49b9eb7540a168dfa0c9d066d71ea4bb4c5290aed5a5

Aliases

arxiv: 2605.13137 · arxiv_version: 2605.13137v2 · doi: 10.48550/arxiv.2605.13137 · pith_short_12: KOOKJRXX2VNX · pith_short_16: KOOKJRXX2VNXKGV2 · pith_short_8: KOOKJRXX

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/KOOKJRXX2VNXKGV2JG46W5KAUF \
  | 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: 539ca4c6f7d55b751aba49b9eb7540a168dfa0c9d066d71ea4bb4c5290aed5a5

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "73f07a0d6d4db5983508cb527e54b89273b948ee0856f96a0cfb535578c3f400",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.IR",
    "submitted_at": "2026-05-13T08:04:57Z",
    "title_canon_sha256": "919e0266f74bd9c8661c2bf5ce1eb11fc15fcc60a20a9e3ca896d71101db3685"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13137",
    "kind": "arxiv",
    "version": 2
  }
}