Image de couverture de livre

Not All Proofs Are Equal: Evaluating LLM Proof Quality Beyond Correctness

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Petrov, Ivo, Dekoninck, Jasper, Dimitrov, Dimitar I., Vechev, Martin
Format:	Preprint
Publié:	2026
Sujets:	Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

BrokenMath: A Benchmark for Sycophancy in Theorem Proving with LLMs
par: Petrov, Ivo, et autres
Publié: (2025)

Proof or Bluff? Evaluating LLMs on 2025 USA Math Olympiad
par: Petrov, Ivo, et autres
Publié: (2025)

Learning from Saturated Data: Signals Beyond Correctness for LLM Training
par: Hiss, Hanno, et autres
Publié: (2026)

MathArena: Evaluating LLMs on Uncontaminated Math Competitions
par: Balunović, Mislav, et autres
Publié: (2025)

MathConstruct: Challenging LLM Reasoning with Constructive Proofs
par: Balunović, Mislav, et autres
Publié: (2025)

Beyond Benchmarks: MathArena as an Evaluation Platform for Mathematics with LLMs
par: Dekoninck, Jasper, et autres
Publié: (2026)

Polyrating: A Cost-Effective and Bias-Aware Rating System for LLM Evaluation
par: Dekoninck, Jasper, et autres
Publié: (2024)

The Open Proof Corpus: A Large-Scale Study of LLM-Generated Mathematical Proofs
par: Dekoninck, Jasper, et autres
Publié: (2025)

A Unified Approach to Routing and Cascading for LLMs
par: Dekoninck, Jasper, et autres
Publié: (2024)

ConStat: Performance-Based Contamination Detection in Large Language Models
par: Dekoninck, Jasper, et autres
Publié: (2024)

Controlled Text Generation via Language Model Arithmetic
par: Dekoninck, Jasper, et autres
Publié: (2023)

Constrained Decoding of Diffusion LLMs with Context-Free Grammars
par: Mündler, Niels, et autres
Publié: (2025)

GRAIN: Exact Graph Reconstruction from Gradients
par: Drencheva, Maria, et autres
Publié: (2025)

DAGER: Exact Gradient Inversion for Large Language Models
par: Petrov, Ivo, et autres
Publié: (2024)

Modular Synthesis of Efficient Quantum Uncomputation
par: Venev, Hristo, et autres
Publié: (2024)

BgGPT 1.0: Extending English-centric LLMs to other languages
par: Alexandrov, Anton, et autres
Publié: (2024)

Evading Data Contamination Detection for Language Models is (too) Easy
par: Dekoninck, Jasper, et autres
Publié: (2024)

Adaptive Generation of Bias-Eliciting Questions for LLMs
par: Staab, Robin, et autres
Publié: (2025)

All Claims Are Equal, but Some Claims Are More Equal Than Others: Importance-Sensitive Factuality Evaluation of LLM Generations
par: Wanner, Miriam, et autres
Publié: (2025)

SPEAR++: Scaling Gradient Inversion via Sparsely-Used Dictionary Learning
par: Bakarsky, Alexander, et autres
Publié: (2025)

FMI_SU_Yotkova_Kastreva at SemEval-2026 Task 13: Lightweight Detection of LLM-Generated Code via Stylometric Signals
par: Yotkova, Elitsa, et autres
Publié: (2026)

Not All Correct Answers Are Equal: Why Your Distillation Source Matters
par: Tian, Xiaoyu, et autres
Publié: (2025)

SPEAR:Exact Gradient Inversion of Batches in Federated Learning
par: Dimitrov, Dimitar I., et autres
Publié: (2024)

Natural Language Translation of Formal Proofs through Informalization of Proof Steps and Recursive Summarization along Proof Structure
par: Hattori, Seiji, et autres
Publié: (2025)

Post-OCR Text Correction for Bulgarian Historical Documents
par: Beshirov, Angel, et autres
Publié: (2024)

IMProofBench: Benchmarking AI on Research-Level Mathematical Proof Generation
par: Schmitt, Johannes, et autres
Publié: (2025)

Proof-RM: A Scalable and Generalizable Reward Model for Math Proof
par: Yang, Haotong, et autres
Publié: (2026)

AlphaIntegrator: Transformer Action Search for Symbolic Integration Proofs
par: Ünsal, Mert, et autres
Publié: (2024)

BELLS: A Framework Towards Future Proof Benchmarks for the Evaluation of LLM Safeguards
par: Dorn, Diego, et autres
Publié: (2024)

Not All LLM-Generated Data Are Equal: Rethinking Data Weighting in Text Classification
par: Kuo, Hsun-Yu, et autres
Publié: (2024)

Not All Errors Are Created Equal: ASCoT Addresses Late-Stage Fragility in Efficient LLM Reasoning
par: Zhang, Dongxu, et autres
Publié: (2025)

APE-Bench: Evaluating Automated Proof Engineering for Formal Math Libraries
par: Xin, Huajian, et autres
Publié: (2025)

The Proof is in the Almond Cookies
par: van Trijp, Remi, et autres
Publié: (2025)

Proof2Hybrid: Automatic Mathematical Benchmark Synthesis for Proof-Centric Problems
par: Peng, Yebo, et autres
Publié: (2025)

Hiding in Plain Sight: Disguising Data Stealing Attacks in Federated Learning
par: Garov, Kostadin, et autres
Publié: (2023)

Tool-Call Dependency Structure is Linearly Decodable in LLM Agent Residual Streams
par: Sun, Tianda, et autres
Publié: (2026)

Proof of Time: A Benchmark for Evaluating Scientific Idea Judgments
par: Ye, Bingyang, et autres
Publié: (2026)

Reliable Fine-Grained Evaluation of Natural Language Math Proofs
par: Ma, Wenjie, et autres
Publié: (2025)

FormalProofBench: Can Models Write Graduate Level Math Proofs That Are Formally Verified?
par: Ravi, Nikil, et autres
Publié: (2026)

Dissecting Paraphrases: The Impact of Prompt Syntax and supplementary Information on Knowledge Retrieval from Pretrained Language Models
par: Linzbach, Stephan, et autres
Publié: (2024)