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

Auxiliary-Task Learning (ATL) aims to improve the performance of the targettask by leveraging the knowledge obtained from related tasks. Occasionally,learning multiple tasks simultaneously results in lower accuracy than learningonly the target task, which is known as negative transfer. This problem isoften attributed to the gradient conflicts among tasks, and is frequentlytackled by coordinating the task gradients in previous works. However, theseoptimization-based methods largely overlook the auxiliary-target generalizationcapability. To better understand the root cause of negative transfer, weexperimentally investigate it from both optimization and generalizationperspectives. Based on our findings, we introduce ForkMerge, a novel approachthat periodically forks the model into multiple branches, automaticallysearches the varying task weights by minimizing target validation errors, anddynamically merges all branches to filter out detrimental task-parameterupdates. On a series of auxiliary-task learning benchmarks, ForkMergeoutperforms existing methods and effectively mitigates negative transfer.

ForkMerge: Mitigating Negative Transfer in Auxiliary-Task Learning