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

The complementary potential of Large Language Models (LLM) assumesoff-the-shelf LLMs have heterogeneous expertise in a wide range of domains andtasks so that an ensemble of LLMs can achieve consistently better performance.Existing ensemble methods for LLMs mainly focus on reward model ranking ofoutputs, leading to significant computation overhead. To combat this issue, werevisit the complementary potential of LLMs and further elaborate it by mininglatent expertise with off-the-shelf reward models. We propose Zooter, areward-guided routing method distilling rewards on training queries to train arouting function, which can precisely distribute each query to the LLM withexpertise about it. We also integrate a tag-based label enhancement to mitigatenoise from uncertainty when using rewards as silver supervision. Zooter showscomputation efficiency in inference as it introduces only a minor computationoverhead of a routing function compared with reward model ranking methods. Weevaluate Zooter on a comprehensive benchmark collection with 26 subsets ondifferent domains and tasks. Zooter outperforms the best single model onaverage and ranks first on 44% of tasks, even surpassing multiple reward modelranking methods.

Routing to the Expert: Efficient Reward-guided Ensemble of Large Language Models