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 recently demonstrated exceptionalperformance in various Natural Language Processing (NLP) tasks. They have alsoshown the ability to perform chain-of-thought (CoT) reasoning to solve complexproblems. Recent studies have explored CoT reasoning in complex multimodalscenarios, such as the science question answering task, by fine-tuningmultimodal models with high-quality human-annotated CoT rationales. However,collecting high-quality COT rationales is usually time-consuming and costly.Besides, the annotated rationales are hardly accurate due to the externalessential information missed. To address these issues, we propose a novelmethod termed T-SciQ that aims at teaching science question answering with LLMsignals. The T-SciQ approach generates high-quality CoT rationales as teachingsignals and is advanced to train much smaller models to perform CoT reasoningin complex modalities. Additionally, we introduce a novel data mixing strategyto produce more effective teaching data samples for simple and complex sciencequestion answer problems. Extensive experimental results show that our T-SciQmethod achieves a new state-of-the-art performance on the ScienceQA benchmark,with an accuracy of 96.18%. Moreover, our approach outperforms the mostpowerful fine-tuned baseline by 4.5%. The code is publicly available athttps://github.com/T-SciQ/T-SciQ.

T-SciQ: Teaching Multimodal Chain-of-Thought Reasoning via Mixed Large Language Model Signals for Science Question Answering