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

Large Language Models (LLMs) have shown outstanding performance across widerange of downstream tasks. This competency is attributed to their substantialparameter size and pre-training on extensive corpus. Moreover, LLMs haveexhibited enhanced reasoning capabilities in tackling complex reasoning tasks,owing to the utilization of a method named ``Chain-of-Thought (CoT)prompting''. This method is designed to generate intermediate reasoning stepsthat guide the inference of the final answer. However, it is essential tohighlight that these advanced reasoning abilities appear to emerge in modelswith a minimum of 10 billion parameters, thereby limiting its efficacy insituations where computational resources are constrained. In this paper, weinvestigate the possibility of transferring the reasoning capabilities of LLMsto smaller models via knowledge distillation. Specifically, we propose Sci-CoT,a two-stage framework that separates the processes of generating rationales andinferring answers. This method enables a more efficient use of rationalesduring the answer inference stage, leading to improved performance onscientific question-answering tasks. Utilizing Sci-CoT, our 80-millionparameter model is able to exceed the performance of BLOOM-176B in the ARC-Easydataset under the few shot setting.

Sci-CoT: Leveraging Large Language Models for Enhanced Knowledge Distillation in Small Models for Scientific QA