English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This paper studies task adaptive pre-trained model selection, anunderexplored problem of assessing pre-trained models for the target task andselect best ones from the model zoo \emph{without fine-tuning}. A few pilotworks addressed the problem in transferring supervised pre-trained models toclassification tasks, but they cannot handle emerging unsupervised pre-trainedmodels or regression tasks. In pursuit of a practical assessment method, wepropose to estimate the maximum value of label evidence given featuresextracted by pre-trained models. Unlike the maximum likelihood, the maximumevidence is \emph{immune to over-fitting}, while its expensive computation canbe dramatically reduced by our carefully designed algorithm. The Logarithm ofMaximum Evidence (LogME) can be used to assess pre-trained models for transferlearning: a pre-trained model with a high LogME value is likely to have goodtransfer performance. LogME is \emph{fast, accurate, and general},characterizing itself as the first practical method for assessing pre-trainedmodels. Compared with brute-force fine-tuning, LogME brings at most$3000\times$ speedup in wall-clock time and requires only $1\%$ memoryfootprint. It outperforms prior methods by a large margin in their setting andis applicable to new settings. It is general enough for diverse pre-trainedmodels (supervised pre-trained and unsupervised pre-trained), downstream tasks(classification and regression), and modalities (vision and language). Code isavailable at this repository:\href{https://github.com/thuml/LogME}{https://github.com/thuml/LogME}.

LogME: Practical Assessment of Pre-trained Models for Transfer Learning