Pith. sign in

REVIEW 13 cited by

Formal Mathematics Statement Curriculum Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2202.01344 v1 pith:5GBRTQKY submitted 2022-02-03 cs.LG cs.AI

Formal Mathematics Statement Curriculum Learning

classification cs.LG cs.AI
keywords expertiterationformalappliedcurriculumlearningmathematicsproblems
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

We explore the use of expert iteration in the context of language modeling applied to formal mathematics. We show that at same compute budget, expert iteration, by which we mean proof search interleaved with learning, dramatically outperforms proof search only. We also observe that when applied to a collection of formal statements of sufficiently varied difficulty, expert iteration is capable of finding and solving a curriculum of increasingly difficult problems, without the need for associated ground-truth proofs. Finally, by applying this expert iteration to a manually curated set of problem statements, we achieve state-of-the-art on the miniF2F benchmark, automatically solving multiple challenging problems drawn from high school olympiads.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 13 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Diffusion-Proof: Recipe for Formal Theorem Proving Beyond Auto-Regressive Generation

    cs.LG 2026-06 unverdicted novelty 8.0

    Diffusion-Proof trains diffusion LLMs (dLLM-Prover-7B for whole proofs and dLLM-Corrector-7B for local correction) that outperform AR baselines by 1.61% on ProofNet-Test and 6.14% on MiniF2F-Test while solving one IMO...

  2. LAMP: Lean-based Agentic framework with MCP and Proof Repair

    cs.LO 2026-06 conditional novelty 7.0

    LAMP achieves 96.7% success generating verified Lean proofs for 90 Combinatorics on Words theorems by coordinating Planner, Builder, and Verifier agents with a CoW ontology accessed through Model Context Protocol.

  3. Geometric Measurements of the Axiom of Choice in Neural Proof Embeddings

    cs.LG 2026-06 unverdicted novelty 7.0

    Proofs depending on the axiom of choice show a geometric signature in neural embeddings of tactic sequences that weakens with dependency-graph distance and correlates with prover failure rates.

  4. Formalizing Mathematics at Scale

    cs.AI 2026-05 accept novelty 7.0

    A multi-agent framework called AutoformBot autoformalized 26 textbooks spanning analysis, algebra, topology, combinatorics and probability into a verified Lean 4 library of 45k declarations, demonstrating scalable for...

  5. Automating Formal Verification with Agent-Guided Tree Search

    cs.LO 2026-05 unverdicted novelty 6.0

    Agent-directed tree search improves LLM performance on Lean formal verification tasks, with context-based orchestration solving more intermediate specs at lower token cost than baseline agents.

  6. OProver: A Unified Framework for Agentic Formal Theorem Proving

    cs.CL 2026-05 unverdicted novelty 6.0

    OProver-32B achieves top Pass@32 scores on MiniF2F, ProverBench, and PutnamBench by combining continued pretraining with iterative agentic proving, retrieval, SFT on repairs, and RL on unresolved cases using a 6.86M-p...

  7. Correcting Influence: Unboxing LLM Outputs with Orthogonal Latent Spaces

    cs.LG 2026-05 unverdicted novelty 6.0

    A latent mediation framework with sparse autoencoders enables non-additive token-level influence attribution in LLMs by learning orthogonal features and back-propagating attributions.

  8. Rethinking Supervision Granularity: Segment-Level Learning for LLM-Based Theorem Proving

    cs.AI 2026-05 unverdicted novelty 6.0

    Segment-level supervision extracts coherent proof segments to train policy models that achieve 61-66% success on miniF2F, outperforming step-level and whole-proof methods while also improving existing provers.

  9. ProofSketcher: Hybrid LLM + Lightweight Proof Checker for Reliable Math/Logic Reasoning

    cs.AI 2026-04 unverdicted novelty 6.0

    A hybrid pipeline lets an LLM write high-level proof sketches in a compact DSL that a lightweight kernel then expands into explicit, checkable obligations for reliable math and logic reasoning.

  10. Measuring Representation Robustness in Large Language Models for Geometry

    cs.CL 2026-04 unverdicted novelty 6.0

    LLMs display accuracy gaps of up to 14 percentage points on the same geometry problems solely due to representation choice, with vector forms consistently weakest and a convert-then-solve prompt helping only high-capa...

  11. Llemma: An Open Language Model For Mathematics

    cs.CL 2023-10 unverdicted novelty 6.0

    Continued pretraining of Code Llama on Proof-Pile-2 yields Llemma, an open math-specialized LLM that beats known open base models on MATH and supports tool use plus formal proving out of the box.

  12. PaLM-E: An Embodied Multimodal Language Model

    cs.LG 2023-03 conditional novelty 6.0

    PaLM-E is a single 562B-parameter multimodal model that performs embodied reasoning tasks like robotic manipulation planning and visual question answering by interleaving vision, state, and text inputs with positive t...

  13. Automating Formal Verification with Reinforcement Learning and Recursive Inference

    cs.LG 2026-05 unverdicted novelty 5.0

    RLVR training raises verified Dafny pass rates from 9.7% to 31.1% on a filtered benchmark while a Lean proof scaffold lifts success from 46.2% to 69.2% on a pilot set and solves 7 of 42 prior unsolved tasks.