Leaderboard

Verified proof rate by configuration

Share of theorems for which the configuration produced a proof the certifier accepted, with a 95% Wilson interval. The same model and the same theorems throughout; configurations differ only in which parts of the loop are on.

ConfigurationVerifiedRate, 95% CI≤1 call≤2≤4Median callsCost / verified
Template (no LLM)
A fixed list of automation tactics (simp, omega, linarith, nlinarith, positivity, aesop, grind, exact?, …) tried in order. No model.
46/174
26.4%
26%26%26%0$0.00000
Full (plan + retrieval + repair)
A planning call (informal sketch plus a Lean `have` skeleton that Lean type-checks), then retrieval plus repair.
13/174
7.5%
0%4%7%2$0.026
Full − skeleton
The full agent with an informal plan only, no Lean skeleton.
13/174
7.5%
0%5%7%2$0.019
BM25 retrieval + repair
Retrieval + repair, with BM25 retrieval only.
13/174
7.5%
3%5%7%2$0.015
Repair (paraphrased prompt)
Repair with a paraphrased prompt (prompt-sensitivity check).
12/174
6.9%
3%6%7%2$0.011
Full − memory
The full agent, but each repair round shows only the latest attempt.
11/174
6.3%
0%4%6%2$0.028
Repair
One draft, then up to three rounds in which the model sees Lean's exact error output.
11/174
6.3%
1%5%6%2$0.015
Full − compiler feedback
The full agent, but a failed round only says the proof failed; Lean's output is withheld.
10/174
5.7%
0%4%6%2$0.031
Full − retrieval
The full agent without retrieved lemmas.
9/174
5.2%
0%5%5%2$0.030
Dense retrieval + repair
Retrieval + repair, with dense retrieval only.
9/174
5.2%
2%3%5%2$0.021
Hybrid retrieval + repair
Retrieval plus up to three compiler-feedback repair rounds.
8/174
4.6%
2%3%5%2$0.023
Direct (paraphrased prompt)
Direct with a paraphrased prompt (prompt-sensitivity check).
5/174
2.9%
3%3%3%1$0.00403
Hybrid retrieval
One draft with eight lemmas retrieved from Mathlib (hybrid retriever) in the prompt.
3/174
1.7%
2%2%2%1$0.012
Direct ×4
Four drafts: the Direct draft plus three resampled with fixed seeds (unseeded repeats were near-identical at this gateway). The equal-budget control for repair: same number of model calls, no feedback.
2/174
1.1%
1%1%1%1.5$0.059
Direct
One model draft, checked once.
1/174
0.6%
1%1%1%1$0.030

Pre-registered contrasts

Each row compares two configurations on the theorems both attempted. +a/−b counts theorems only the treatment / only the control solved.

ContrastnDifference [95% CI]+a / −bpHolm p
RQ2 compiler-feedback repair vs one draft
Repair vs Direct
174+5.7 pts [+2.3 pts, +9.2 pts]+10 / −00.0020.025
RQ2 repair vs independent resampling at equal LLM calls
Repair vs Direct ×4
174+5.2 pts [+2.3 pts, +8.6 pts]+9 / −00.0040.047
RQ1 retrieval, single draft
Hybrid retrieval vs Direct
174+1.1 pts [−1.1 pts, +3.4 pts]+3 / −10.6251.000
RQ1 retrieval, with repair
Hybrid retrieval + repair vs Repair
174−1.7 pts [−6.3 pts, +2.9 pts]+6 / −90.6071.000
RQ3 planning on top of retrieval + repair
Full (plan + retrieval + repair) vs Hybrid retrieval + repair
174+2.9 pts [−0.6 pts, +6.3 pts]+7 / −20.1801.000
Primary: full agent vs direct generation
Full (plan + retrieval + repair) vs Direct
174+6.9 pts [+3.4 pts, +10.9 pts]+12 / −0<0.0010.007
Ablation: retrieval
Full (plan + retrieval + repair) vs Full − retrieval
174+2.3 pts [−1.7 pts, +6.3 pts]+9 / −50.4241.000
Ablation: compiler feedback
Full (plan + retrieval + repair) vs Full − compiler feedback
174+1.7 pts [+0.0 pts, +4.0 pts]+3 / −00.2501.000
Ablation: memory of earlier attempts
Full (plan + retrieval + repair) vs Full − memory
174+1.1 pts [+0.0 pts, +2.9 pts]+2 / −00.5001.000
Ablation: proof skeleton
Full (plan + retrieval + repair) vs Full − skeleton
174+0.0 pts [−3.4 pts, +3.4 pts]+4 / −41.0001.000
Retriever: BM25 vs hybrid
BM25 retrieval + repair vs Hybrid retrieval + repair
174+2.9 pts [+0.6 pts, +5.7 pts]+5 / −00.0630.688
Retriever: dense vs hybrid
Dense retrieval + repair vs Hybrid retrieval + repair
174+0.6 pts [−2.3 pts, +3.4 pts]+4 / −31.0001.000
LLM vs no-LLM automation
Direct vs Template (no LLM)
174−25.9 pts [−32.2 pts, −19.5 pts]+0 / −45<0.001<0.001
Prompt sensitivity: paraphrased prompt, direct
Direct (paraphrased prompt) vs Direct
174+2.3 pts [+0.6 pts, +4.6 pts]+4 / −00.1251.000
Prompt sensitivity: paraphrased prompt, repair
Repair (paraphrased prompt) vs Repair
174+0.6 pts [+0.0 pts, +1.7 pts]+1 / −01.0001.000

By theorem source

ConfigurationMathlib held-outAuthored
Template (no LLM)10% (12/121)64% (34/53)
Full (plan + retrieval + repair)4% (5/121)15% (8/53)
Full − skeleton4% (5/121)15% (8/53)
BM25 retrieval + repair6% (7/121)11% (6/53)
Repair (paraphrased prompt)7% (8/121)8% (4/53)
Full − memory3% (4/121)13% (7/53)
Repair7% (8/121)6% (3/53)
Full − compiler feedback3% (4/121)11% (6/53)
Full − retrieval6% (7/121)4% (2/53)
Dense retrieval + repair2% (2/121)13% (7/53)
Hybrid retrieval + repair3% (4/121)8% (4/53)
Direct (paraphrased prompt)1% (1/121)8% (4/53)
Hybrid retrieval1% (1/121)4% (2/53)
Direct ×40% (0/121)4% (2/53)
Direct0% (0/121)2% (1/53)

By family

ConfigurationAlgebraCategory theoryFunctionsInequalitiesNumber theoryProbabilitySets
Template (no LLM)33% (10/30)21% (5/24)54% (7/13)23% (7/30)17% (5/30)17% (4/24)35% (8/23)
Full (plan + retrieval + repair)7% (2/30)8% (2/24)0% (0/13)10% (3/30)10% (3/30)4% (1/24)9% (2/23)
Full − skeleton0% (0/30)8% (2/24)8% (1/13)10% (3/30)10% (3/30)4% (1/24)13% (3/23)
BM25 retrieval + repair10% (3/30)4% (1/24)8% (1/13)10% (3/30)7% (2/30)4% (1/24)9% (2/23)
Repair (paraphrased prompt)13% (4/30)8% (2/24)0% (0/13)13% (4/30)7% (2/30)0% (0/24)0% (0/23)
Full − memory7% (2/30)4% (1/24)0% (0/13)7% (2/30)10% (3/30)4% (1/24)9% (2/23)
Repair10% (3/30)8% (2/24)0% (0/13)13% (4/30)7% (2/30)0% (0/24)0% (0/23)
Full − compiler feedback3% (1/30)8% (2/24)0% (0/13)7% (2/30)7% (2/30)4% (1/24)9% (2/23)
Full − retrieval10% (3/30)4% (1/24)8% (1/13)10% (3/30)3% (1/30)0% (0/24)0% (0/23)
Dense retrieval + repair3% (1/30)8% (2/24)8% (1/13)7% (2/30)7% (2/30)4% (1/24)0% (0/23)
Hybrid retrieval + repair3% (1/30)4% (1/24)8% (1/13)3% (1/30)3% (1/30)4% (1/24)9% (2/23)
Direct (paraphrased prompt)10% (3/30)4% (1/24)0% (0/13)0% (0/30)3% (1/30)0% (0/24)0% (0/23)
Hybrid retrieval0% (0/30)0% (0/24)0% (0/13)3% (1/30)0% (0/30)4% (1/24)4% (1/23)
Direct ×40% (0/30)4% (1/24)0% (0/13)0% (0/30)3% (1/30)0% (0/24)0% (0/23)
Direct0% (0/30)4% (1/24)0% (0/13)0% (0/30)0% (0/30)0% (0/24)0% (0/23)