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

As a novel and effective fine-tuning paradigm based on large-scalepre-trained language models (PLMs), prompt-tuning aims to reduce the gapbetween downstream tasks and pre-training objectives. While prompt-tuning hasyielded continuous advancements in various tasks, such an approach stillremains a persistent defect: prompt-tuning methods fail to generalize tospecific few-shot patterns. From the perspective of distribution analyses, wedisclose that the intrinsic issues behind the phenomenon are theover-multitudinous conceptual knowledge contained in PLMs and the abridgedknowledge for target downstream domains, which jointly result in that PLMsmis-locate the knowledge distributions corresponding to the target domains inthe universal knowledge embedding space. To this end, we intuitively explore toapproximate the unabridged target domains of downstream tasks in a debiasedmanner, and then abstract such domains to generate discriminative prompts,thereby providing the de-ambiguous guidance for PLMs. Guided by such anintuition, we propose a simple yet effective approach, namely BayesPrompt, tolearn prompts that contain the domain discriminative information against theinterference from domain-irrelevant knowledge. BayesPrompt primitivelyleverages known distributions to approximate the debiased factual distributionsof target domains and further uniformly samples certain representative featuresfrom the approximated distributions to generate the ultimate prompts for PLMs.We provide theoretical insights with the connection to domain adaptation.Empirically, our method achieves state-of-the-art performance on benchmarks.

BayesPrompt: Prompting Large-Scale Pre-Trained Language Models on Few-shot Inference via Debiased Domain Abstraction