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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

A recent trend in artificial intelligence is the use of pretrained models forlanguage and vision tasks, which have achieved extraordinary performance butalso puzzling failures. Probing these models' abilities in diverse ways istherefore critical to the field. In this paper, we explore the reliability ofmodels, where we define a reliable model as one that not only achieves strongpredictive performance but also performs well consistently over manydecision-making tasks involving uncertainty (e.g., selective prediction, openset recognition), robust generalization (e.g., accuracy and proper scoringrules such as log-likelihood on in- and out-of-distribution datasets), andadaptation (e.g., active learning, few-shot uncertainty). We devise 10 types oftasks over 40 datasets in order to evaluate different aspects of reliability onboth vision and language domains. To improve reliability, we developed ViT-Plexand T5-Plex, pretrained large model extensions for vision and languagemodalities, respectively. Plex greatly improves the state-of-the-art acrossreliability tasks, and simplifies the traditional protocol as it improves theout-of-the-box performance and does not require designing scores or tuning themodel for each task. We demonstrate scaling effects over model sizes up to 1Bparameters and pretraining dataset sizes up to 4B examples. We also demonstratePlex's capabilities on challenging tasks including zero-shot open setrecognition, active learning, and uncertainty in conversational languageunderstanding.

Plex: Towards Reliability using Pretrained Large Model Extensions