@inproceedings{zhao-etal-2024-large-language,
title = "Large Language Models are In-context Teachers for Knowledge Reasoning",
author = "Zhao, Jiachen and
Yao, Zonghai and
Yang, Zhichao and
Yu, Hong",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2024",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-emnlp.961/",
doi = "10.18653/v1/2024.findings-emnlp.961",
pages = "16470--16486",
abstract = "In this work, we study in-context teaching(ICT), where a teacher provides in-context example rationales to teach a student to reasonover unseen cases. Human teachers are usually required to craft in-context demonstrations, which are costly and have high variance. We ask whether a large language model (LLM) can serve as a more effective in-context teacher for itself or otherLLMs, compared to humans. Inspired by the Encoding Specificity Hypothesis from human episodic memory, we hypothesize thatin-context exemplars crafted by the teacher should match the training data of the student. This hypothesis motivates us to propose Self-Explain where an LLM`s self-elicited explanations are used as in-context demonstrations for prompting it as they are generalized fromthe model`s training examples. Self-Explain is shown to significantly outperform using human-crafted exemplars and other baselines.Furthermore, we reveal that for ICT, rationales from different teacher LLMs or human experts that more resemble the student LLM`s self-explanations are better in-context demonstrations. This supports our encoding specificity hypothesis. We then propose Teach-Back that aligns a teacher LLM with the student to enhance the ICT performance. For example, Teach-Back enables a 7B model to teach the much larger GPT-3.5 in context, surpassing human teachers by around 5{\%} in test accuracy on medical question answering."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhao-etal-2024-large-language">
<titleInfo>
<title>Large Language Models are In-context Teachers for Knowledge Reasoning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jiachen</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zonghai</namePart>
<namePart type="family">Yao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhichao</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hong</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yaser</namePart>
<namePart type="family">Al-Onaizan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohit</namePart>
<namePart type="family">Bansal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yun-Nung</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Miami, Florida, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In this work, we study in-context teaching(ICT), where a teacher provides in-context example rationales to teach a student to reasonover unseen cases. Human teachers are usually required to craft in-context demonstrations, which are costly and have high variance. We ask whether a large language model (LLM) can serve as a more effective in-context teacher for itself or otherLLMs, compared to humans. Inspired by the Encoding Specificity Hypothesis from human episodic memory, we hypothesize thatin-context exemplars crafted by the teacher should match the training data of the student. This hypothesis motivates us to propose Self-Explain where an LLM‘s self-elicited explanations are used as in-context demonstrations for prompting it as they are generalized fromthe model‘s training examples. Self-Explain is shown to significantly outperform using human-crafted exemplars and other baselines.Furthermore, we reveal that for ICT, rationales from different teacher LLMs or human experts that more resemble the student LLM‘s self-explanations are better in-context demonstrations. This supports our encoding specificity hypothesis. We then propose Teach-Back that aligns a teacher LLM with the student to enhance the ICT performance. For example, Teach-Back enables a 7B model to teach the much larger GPT-3.5 in context, surpassing human teachers by around 5% in test accuracy on medical question answering.</abstract>
<identifier type="citekey">zhao-etal-2024-large-language</identifier>
<identifier type="doi">10.18653/v1/2024.findings-emnlp.961</identifier>
<location>
<url>https://aclanthology.org/2024.findings-emnlp.961/</url>
</location>
<part>
<date>2024-11</date>
<extent unit="page">
<start>16470</start>
<end>16486</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Large Language Models are In-context Teachers for Knowledge Reasoning
%A Zhao, Jiachen
%A Yao, Zonghai
%A Yang, Zhichao
%A Yu, Hong
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Findings of the Association for Computational Linguistics: EMNLP 2024
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F zhao-etal-2024-large-language
%X In this work, we study in-context teaching(ICT), where a teacher provides in-context example rationales to teach a student to reasonover unseen cases. Human teachers are usually required to craft in-context demonstrations, which are costly and have high variance. We ask whether a large language model (LLM) can serve as a more effective in-context teacher for itself or otherLLMs, compared to humans. Inspired by the Encoding Specificity Hypothesis from human episodic memory, we hypothesize thatin-context exemplars crafted by the teacher should match the training data of the student. This hypothesis motivates us to propose Self-Explain where an LLM‘s self-elicited explanations are used as in-context demonstrations for prompting it as they are generalized fromthe model‘s training examples. Self-Explain is shown to significantly outperform using human-crafted exemplars and other baselines.Furthermore, we reveal that for ICT, rationales from different teacher LLMs or human experts that more resemble the student LLM‘s self-explanations are better in-context demonstrations. This supports our encoding specificity hypothesis. We then propose Teach-Back that aligns a teacher LLM with the student to enhance the ICT performance. For example, Teach-Back enables a 7B model to teach the much larger GPT-3.5 in context, surpassing human teachers by around 5% in test accuracy on medical question answering.
%R 10.18653/v1/2024.findings-emnlp.961
%U https://aclanthology.org/2024.findings-emnlp.961/
%U https://doi.org/10.18653/v1/2024.findings-emnlp.961
%P 16470-16486
Markdown (Informal)
[Large Language Models are In-context Teachers for Knowledge Reasoning](https://aclanthology.org/2024.findings-emnlp.961/) (Zhao et al., Findings 2024)
ACL