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

Despite the remarkable success of foundation models, their task-specificfine-tuning paradigm makes them inconsistent with the goal of generalperception modeling. The key to eliminating this inconsistency is to usegeneralist models for general task modeling. However, existing attempts atgeneralist models are inadequate in both versatility and performance. In thispaper, we propose Uni-Perceiver v2, which is the first generalist model capableof handling major large-scale vision and vision-language tasks with competitiveperformance. Specifically, images are encoded as general region proposals,while texts are encoded via a Transformer-based language model. The encodedrepresentations are transformed by a task-agnostic decoder. Different tasks areformulated as a unified maximum likelihood estimation problem. We furtherpropose an improved optimizer to ensure stable multi-task learning with anunmixed sampling strategy, which is helpful for tasks requiring largebatch-size training. After being jointly trained on various tasks,Uni-Perceiver v2 is capable of directly handling downstream tasks without anytask-specific adaptation. Results show that Uni-Perceiver v2 outperforms allexisting generalist models in both versatility and performance. Meanwhile,compared with the commonly-recognized strong baselines that requiretasks-specific fine-tuning, Uni-Perceiver v2 achieves competitive performanceon a broad range of vision and vision-language tasks.

Uni-Perceiver v2: A Generalist Model for Large-Scale Vision and Vision-Language Tasks