autoformalization

Tag

Cards List
#autoformalization

AuXSpec: Learning to Autoformalize Program Specifications via Executable Postconditions

Hugging Face Daily Papers ↗ · 5d ago Cached

AuXSpec introduces a Lean framework that pairs declarative postconditions with executable counterparts, enabling execution-based agreement as an RL reward for training specification-generation models. Models trained with this reward achieve up to 84.3% SpecGen pass@4, surpassing prior open-source SOTA and frontier models on the Verina benchmark.

0 favorites 0 likes
#autoformalization

PaperFormalizer: Author-in-the-Loop Autoformalization of Research Papers in Lean

arXiv cs.AI ↗ · 2026-10-26

本文介绍 PaperFormalizer——一个基于 Lean 的作者参与式自动形式化流程,可从 LaTeX 论文出发构建语句依赖图、审计语句并将其形式化证明,同时在语义存在歧义时让作者做出决策,从而产出经 Lean 验证的论文版本。

0 favorites 0 likes
#autoformalization

Beyond Compilation: Evaluating Faithful Natural-Language-to-Lean Statement Formalization

Hugging Face Daily Papers ↗ · 2026-10-15 Cached

该论文评估了Lean编译成功是否等同于忠实的自然语言到Lean语句形式化,发现多个系统在编译通过与语义正确之间存在1.3至29.5个百分点的显著差距,且工具增强型代理的大部分差距源于真实的翻译错误。

0 favorites 0 likes
#autoformalization

What mathematicians should know about the Lean Theorem Prover: reliability & AI

Hacker News Top ↗ · 2026-10-09 Cached

Thomas Hales writes a guest post for Terence Tao's blog explaining what mathematicians should know about the Lean theorem prover, covering formal proof reliability, the growth of the mathlib library, and how AI-driven autoformalization has become a practical reality in 2026.

0 favorites 0 likes
#autoformalization

Law And Order: Tax Law Autoformalization

arXiv cs.AI ↗ · 2026-10-05 Cached

Law&Order is a neuro-symbolic framework that combines LLM synthesis with cell-level symbolic verification and iterative error repair to automatically formalize tax forms into executable programs, achieving 100% accuracy on 51 held-out TaxCalcBench returns compared to 66% for the best LLM alone.

0 favorites 0 likes
#autoformalization

@mweber_PU: Moving from individual proofs to large-scale autoformalization requires new tools. We introduce Choir, an open protocol…

X AI KOLs Timeline ↗ · 2026-10-02 Cached

The authors introduce Choir, an open, modular protocol for distributed multi-agent autoformalization that decomposes projects into GitHub-based tasks with deterministic trust-gate verification, supporting Lean 4, Isabelle, and Rocq; it includes a preprint and an open-source repo funded by DARPA's expMath program.

0 favorites 0 likes
#autoformalization

When Scientific Contradictions Are Lost in Translation

arXiv cs.AI ↗ · 2026-10-01 Cached

This arXiv paper studies how language models decide whether two scientific findings are comparable before resolving contradictions. In a controlled task with unsatisfiable XOR constraints translated into lab reports, GPT-5.6 Sol and Claude Opo 5 recover the best-supported assignment well when constraints are explicit, but models often defer to biological expectations when findings are presented in scientific prose.

0 favorites 0 likes
#autoformalization

Sage: Formalization with Semantic Correction

arXiv cs.LG ↗ · 2026-09-30 Cached

华为 Lagrange 数学计算研究中心提出 Sage,一个四阶段分解生成管线加双重信号语义校正循环的自动形式化框架,将答案泄漏率从70.9%降至2.7%,在 Omni-MATH NP 上达到73.3% pass@4,并在新提出的 IMO-Unformalized 基准上零样本达到87.4%验证保真度。

0 favorites 0 likes
#autoformalization

Beyond Solver Verdicts: Generative Reward Models for Autoformalization

arXiv cs.LG ↗ · 2026-09-11 Cached

This paper proposes Generative Verification (GenV), a method using generative reward models to detect reference-equivalence failures in autoformalization, addressing vulnerabilities in neurosymbolic systems and improving verification accuracy.

0 favorites 0 likes
#autoformalization

SHADOWBENCH: Toward Reliable Automatic Evaluation of Semantic Alignment in Autoformalization

arXiv cs.CL ↗ · 2026-09-01 Cached

This paper introduces SA-Pass, a method for evaluating semantic alignment in autoformalization, and presents ShadowBench, a Lean 4 benchmark with 178 problems, demonstrating high agreement with expert judgments.

0 favorites 0 likes
#autoformalization

FormalTCS: Benchmarking End-to-End Frontier Formal Theoretical Computer Science Research of Large Language Models

arXiv cs.CL ↗ · 2026-08-21 Cached

FormalTCS is a benchmark for evaluating large language models on end-to-end theoretical computer science research, revealing significant limitations, especially in autoformalization.

0 favorites 0 likes
#autoformalization

MathForm: Scaling Mathematical Autoformalization with Knowledge Retrieval and Verification-Guided Refinement

Hugging Face Daily Papers ↗ · 2026-08-14 Cached

MathForm introduces a framework for mathematical autoformalization using knowledge retrieval and verification-guided refinement, yielding the FormalVerse dataset and an 8B model that outperforms specialized baselines.

0 favorites 0 likes
#autoformalization

LeanFlow: A Case Study in Workflow-Driven Lean Autoformalization

arXiv cs.AI ↗ · 2026-07-24 Cached

LeanFlow is an LLM agent system for translating mathematical papers into formalized Lean projects, evaluated through case studies and benchmarks with Kimi-K2.6 and GPT-5.5, achieving high completion rates within budget constraints.

0 favorites 0 likes
#autoformalization

Theory-Level Autoformalization: From Isolated Statements to Unified Formal Knowledge Bases

arXiv cs.AI ↗ · 2026-07-16 Cached

This position paper argues for theory-level autoformalization, which formalizes entire theories including axioms, definitions, and lemmas as coherent libraries, rather than isolated statements. It discusses the significance, alternative views, open challenges, and proposes paths forward for this shift in formalization research.

0 favorites 0 likes
#autoformalization

Beyond the Library: An Agentic Framework for Autoformalizing Research Mathematics

arXiv cs.AI ↗ · 2026-07-01 Cached

Presents an agentic framework using general coding LLMs to autoformalize research-level mathematics into Lean 4 code, evaluated on Putnam problems and STOC conference papers.

0 favorites 0 likes
#autoformalization

Leanstral 1.5

Hacker News Top ↗ · 2026-06-30 Cached

Mistral AI releases Leanstral 1.5, an updated Lean 4 formal proof engineering model optimized for automated theorem proving and autoformalization, with 119B total parameters and 6.5B active parameters.

0 favorites 0 likes
#autoformalization

Verifiable Geometry Problem Solving: Solver-Driven Autoformalization and Theorem Proposing

arXiv cs.AI ↗ · 2026-06-29 Cached

This paper introduces SD-GPS, a solver-driven framework for geometry problem solving that uses autoformalization guided by solver feedback and verified theorem proposing to overcome bottlenecks in neuro-symbolic systems.

0 favorites 0 likes
#autoformalization

The Signal-Coverage Matrix: Stratifying Type and Semantic Errors in Statement Autoformalization

arXiv cs.CL ↗ · 2026-06-29 Cached

This paper introduces a signal-coverage matrix that decomposes type-correctness gains in autoformalization into four strata, revealing the mechanisms behind LLM refinements and showing that headline metrics can obscure which errors are actually resolved.

0 favorites 0 likes
#autoformalization

Autoformalization of Agent Instructions into Policy-as-Code

arXiv cs.AI ↗ · 2026-06-26 Cached

This paper presents an autoformalization pipeline that translates agent prompts, MCP tool descriptions, and natural language policy documents into formally verified policies using an LLM-based generator-critic loop, achieving better coverage than hand-coded enforcement on MedAgentBench.

0 favorites 0 likes
#autoformalization

Closing the Loop: Formally Verified Law as a Reward Signal for Self-Improving Legal AI

arXiv cs.LG ↗ · 2026-06-24 Cached

This paper presents an architecture that uses formally verified law as a reward signal for training legal AI, adaptively autoformalizing legal rules into a formal calculus and employing a verifier to ensure provable correctness, demonstrated on German and US law examples.

0 favorites 0 likes
Next →
← Back to home

Submit Feedback