ASSESS: A Semantic and Structural Evaluation Framework for Statement Similarity

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Liu, Xiaoyang, Zhu, Tao, Dong, Zineng, Liu, Yuntian, Guo, Qingfeng, Liu, Zhaoxuan, Chen, Yu, Luo, Tao
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912889385779200
author Liu, Xiaoyang
Zhu, Tao
Dong, Zineng
Liu, Yuntian
Guo, Qingfeng
Liu, Zhaoxuan
Chen, Yu
Luo, Tao
author_facet Liu, Xiaoyang
Zhu, Tao
Dong, Zineng
Liu, Yuntian
Guo, Qingfeng
Liu, Zhaoxuan
Chen, Yu
Luo, Tao
contents Despite significant strides in statement autoformalization, a critical gap remains in the development of automated evaluation metrics capable of assessing formal translation quality. Existing metrics often fail to balance semantic and structural information: string-based methods neglect semantics, whereas proof-based approaches offer no graded similarity when proofs fail. To address these issues, we introduce ASSESS (A Semantic and Structural Evaluation Framework for Statement Similarity), which captures syntactic structure by transforming formal statements into operator trees and computes a real-valued similarity score using our novel TransTED (Transformation Tree Edit Distance) Similarity metric by incorporating semantic transformations. For rigorous validation, we present EPLA (Evaluating Provability and Likeness for Autoformalization), a benchmark comprising 1,247 expert-annotated formal statement pairs derived from miniF2F and ProofNet, distinctively labeled for both semantic provability and structural likeness. Experiments on the EPLA benchmark demonstrate that TransTED Similarity surpasses existing methods, achieving state-of-the-art accuracy and Kappa score. The benchmark dataset, code, and detailed experimental results are available at https://github.com/XiaoyangLiu-sjtu/ASSESS.
format Preprint
id arxiv_https___arxiv_org_abs_2509_22246
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ASSESS: A Semantic and Structural Evaluation Framework for Statement Similarity
Liu, Xiaoyang
Zhu, Tao
Dong, Zineng
Liu, Yuntian
Guo, Qingfeng
Liu, Zhaoxuan
Chen, Yu
Luo, Tao
Machine Learning
Artificial Intelligence
Despite significant strides in statement autoformalization, a critical gap remains in the development of automated evaluation metrics capable of assessing formal translation quality. Existing metrics often fail to balance semantic and structural information: string-based methods neglect semantics, whereas proof-based approaches offer no graded similarity when proofs fail. To address these issues, we introduce ASSESS (A Semantic and Structural Evaluation Framework for Statement Similarity), which captures syntactic structure by transforming formal statements into operator trees and computes a real-valued similarity score using our novel TransTED (Transformation Tree Edit Distance) Similarity metric by incorporating semantic transformations. For rigorous validation, we present EPLA (Evaluating Provability and Likeness for Autoformalization), a benchmark comprising 1,247 expert-annotated formal statement pairs derived from miniF2F and ProofNet, distinctively labeled for both semantic provability and structural likeness. Experiments on the EPLA benchmark demonstrate that TransTED Similarity surpasses existing methods, achieving state-of-the-art accuracy and Kappa score. The benchmark dataset, code, and detailed experimental results are available at https://github.com/XiaoyangLiu-sjtu/ASSESS.
title ASSESS: A Semantic and Structural Evaluation Framework for Statement Similarity
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2509.22246