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

Recent advancements in large language models (LLMs) have significantlyexpanded their functionality and skills as tool agents. In this paper, we arguethat a waveform pattern in the model's attention allocation has an impact onthe tool use performance, which degrades when the position of essentialinformation hits the trough zone. To address this issue, we propose a novelinference method named Attention Buckets. This approach enables LLMs to handlecontext by conducting parallel processes, each featuring a unique RoPE anglebase that shapes the attention waveform. Attention Buckets ensures that anattention trough of a particular process can be compensated with an attentionpeak of another run, reducing the risk of the LLM missing essential informationresiding within the attention trough. Our extensive experiments on the widelyrecognized tool use benchmark demonstrate the efficacy of our approach, where a7B-parameter open-source model enhanced by Attention Buckets achieves SOTAperformance on par with GPT-4.

Fortify the Shortest Stave in Attention: Enhancing Context Awareness of Large Language Models for Effective Tool Use