跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 20 天前AI 評分42

BEND 2.0:評測設計 DNA 所需的能力

BEND 2.0: Benchmarking the Capabilities Needed to Design DNA

AI 導讀

BEND 2.0 評估 DNA 語言模型預測突變效應及指導序列設計的能力,結果顯示線性讀出可媲美或超越模型原生輸出,功能性後訓練亦能提升表現。基準涵蓋五個實驗數據集的八項任務,模型選擇及表徵取得方式都會影響表現。模型在情境特定效應及多細胞目標設計方面仍有改進空間。

正文

Published on Nov 2

Authors:

,

,

,

,

,

,

,

Abstract

BEND 2.0 evaluates DNA language models on predicting mutation effects and guiding sequence design, revealing that linear readouts and post-training improve performance but context-specific and multi-objective design remain challenging.

Genomic foundation models are increasingly used to propose DNA sequences and edits. But successful design requires capabilities that current benchmarks do not measure. Here we introduce BEND 2.0 (BENchmark for DNA language models), which tests how well models predict the effects of DNA changes and whether these predictions can guide sequence design. Across eight tasks built on five experimental datasets, we find that model choice and how representations are accessed both affect performance. Linear readouts fitted to frozen representations can match or exceed models' native outputs, and functional post-training substantially improves the models' capabilities. Linear readouts transfer well across experimental assays and genes, suggesting that models capture functional patterns beyond the data used to fit individual predictors. At the same time, strong prediction of average effects can conceal weaker prediction of context-specific effects, and accurate prediction of combined mutation effects does not imply accurate prediction of their interactions. In design tasks, model predictions can reduce the number of candidates needed to find a desired response, but still leave substantial room for better design choices, particularly for context-specific and multi-cell objectives. Our benchmark reveals that genomic models now display signs of general utility but have yet to saturate on performance in even simple tasks.

View arXiv page View PDF Project page GitHub Add to collection

Get this paper in your agent:

hf papers read 2611.03806

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2611.03806 in a model README.md to link it from this page.

Datasets citing this paper 1

yhsure/BEND2

Viewer •

Updated 28 minutes ago

•

198k

•

189

•

1

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2611.03806 in a Space README.md to link it from this page.

Collections including this paper 1

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co