Rethinking Text-based Protein Understanding: Retrieval or LLM?

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wu, Juntong, Liu, Zijing, Cao, He, Li, Hao, Feng, Bin, Shu, Zishan, Yu, Ke, Yuan, Li, Li, Yu
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915608019337216
author Wu, Juntong
Liu, Zijing
Cao, He
Li, Hao
Feng, Bin
Shu, Zishan
Yu, Ke
Yuan, Li
Li, Yu
author_facet Wu, Juntong
Liu, Zijing
Cao, He
Li, Hao
Feng, Bin
Shu, Zishan
Yu, Ke
Yuan, Li
Li, Yu
contents In recent years, protein-text models have gained significant attention for their potential in protein generation and understanding. Current approaches focus on integrating protein-related knowledge into large language models through continued pretraining and multi-modal alignment, enabling simultaneous comprehension of textual descriptions and protein sequences. Through a thorough analysis of existing model architectures and text-based protein understanding benchmarks, we identify significant data leakage issues present in current benchmarks. Moreover, conventional metrics derived from natural language processing fail to accurately assess the model's performance in this domain. To address these limitations, we reorganize existing datasets and introduce a novel evaluation framework based on biological entities. Motivated by our observation, we propose a retrieval-enhanced method, which significantly outperforms fine-tuned LLMs for protein-to-text generation and shows accuracy and efficiency in training-free scenarios. Our code and data can be seen at https://github.com/IDEA-XL/RAPM.
format Preprint
id arxiv_https___arxiv_org_abs_2505_20354
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Rethinking Text-based Protein Understanding: Retrieval or LLM?
Wu, Juntong
Liu, Zijing
Cao, He
Li, Hao
Feng, Bin
Shu, Zishan
Yu, Ke
Yuan, Li
Li, Yu
Computation and Language
Artificial Intelligence
In recent years, protein-text models have gained significant attention for their potential in protein generation and understanding. Current approaches focus on integrating protein-related knowledge into large language models through continued pretraining and multi-modal alignment, enabling simultaneous comprehension of textual descriptions and protein sequences. Through a thorough analysis of existing model architectures and text-based protein understanding benchmarks, we identify significant data leakage issues present in current benchmarks. Moreover, conventional metrics derived from natural language processing fail to accurately assess the model's performance in this domain. To address these limitations, we reorganize existing datasets and introduce a novel evaluation framework based on biological entities. Motivated by our observation, we propose a retrieval-enhanced method, which significantly outperforms fine-tuned LLMs for protein-to-text generation and shows accuracy and efficiency in training-free scenarios. Our code and data can be seen at https://github.com/IDEA-XL/RAPM.
title Rethinking Text-based Protein Understanding: Retrieval or LLM?
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2505.20354