Beyond One-Size-Fits-All: Inversion Learning for Highly Effective NLG Evaluation Prompts
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909780015054848 |
|---|---|
| author | Hong, Hanhua Xiao, Chenghao Wang, Yang Liu, Yiqi Rong, Wenge Lin, Chenghua |
| author_facet | Hong, Hanhua Xiao, Chenghao Wang, Yang Liu, Yiqi Rong, Wenge Lin, Chenghua |
| contents | Evaluating natural language generation systems is challenging due to the diversity of valid outputs. While human evaluation is the gold standard, it suffers from inconsistencies, lack of standardisation, and demographic biases, limiting reproducibility. LLM-based evaluators offer a scalable alternative but are highly sensitive to prompt design, where small variations can lead to significant discrepancies. In this work, we propose an inversion learning method that learns effective reverse mappings from model outputs back to their input instructions, enabling the automatic generation of highly effective, model-specific evaluation prompts. Our method requires only a single evaluation sample and eliminates the need for time-consuming manual prompt engineering, thereby improving both efficiency and robustness. Our work contributes toward a new direction for more robust and efficient LLM-based evaluation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_21117 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Beyond One-Size-Fits-All: Inversion Learning for Highly Effective NLG Evaluation Prompts Hong, Hanhua Xiao, Chenghao Wang, Yang Liu, Yiqi Rong, Wenge Lin, Chenghua Computation and Language Evaluating natural language generation systems is challenging due to the diversity of valid outputs. While human evaluation is the gold standard, it suffers from inconsistencies, lack of standardisation, and demographic biases, limiting reproducibility. LLM-based evaluators offer a scalable alternative but are highly sensitive to prompt design, where small variations can lead to significant discrepancies. In this work, we propose an inversion learning method that learns effective reverse mappings from model outputs back to their input instructions, enabling the automatic generation of highly effective, model-specific evaluation prompts. Our method requires only a single evaluation sample and eliminates the need for time-consuming manual prompt engineering, thereby improving both efficiency and robustness. Our work contributes toward a new direction for more robust and efficient LLM-based evaluation. |
| title | Beyond One-Size-Fits-All: Inversion Learning for Highly Effective NLG Evaluation Prompts |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2504.21117 |