@inproceedings{parashar-etal-2023-prompting,
title = "Prompting Scientific Names for Zero-Shot Species Recognition",
author = "Parashar, Shubham and
Lin, Zhiqiu and
Li, Yanan and
Kong, Shu",
editor = "Bouamor, Houda and
Pino, Juan and
Bali, Kalika",
booktitle = "Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2023",
address = "Singapore",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.emnlp-main.610",
doi = "10.18653/v1/2023.emnlp-main.610",
pages = "9856--9861",
abstract = "Trained on web-scale image-text pairs, Vision-Language Models (VLMs) such as CLIP can recognize images of common objects in a zero-shot fashion. However, it is underexplored how to use CLIP for zero-shot recognition of highly specialized concepts, e.g., species of birds, plants, and animals, for which their scientific names are written in Latin or Greek. Indeed, CLIP performs poorly for zero-shot species recognition with prompts that use scientific names, e.g., {``}a photo of Lepus Timidus{''} (which is a scientific name in Latin). This is because these names are usually not included in CLIP{'}s training set. To improve performance, we explore using large-language models (LLMs) to generate descriptions (e.g., of species color and shape) and additionally use them in prompts. However, this method improves only marginally. Instead, we are motivated to translate scientific names (e.g., Lepus Timidus) to common English names (e.g., mountain hare) and use such in the prompts. We find that common names are more likely to be included in CLIP{'}s training set, and prompting them achieves 2{\textasciitilde}5 times higher accuracy on benchmarking datasets of fine-grained species recognition.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="parashar-etal-2023-prompting">
<titleInfo>
<title>Prompting Scientific Names for Zero-Shot Species Recognition</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shubham</namePart>
<namePart type="family">Parashar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhiqiu</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yanan</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shu</namePart>
<namePart type="family">Kong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Houda</namePart>
<namePart type="family">Bouamor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Juan</namePart>
<namePart type="family">Pino</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kalika</namePart>
<namePart type="family">Bali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Singapore</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Trained on web-scale image-text pairs, Vision-Language Models (VLMs) such as CLIP can recognize images of common objects in a zero-shot fashion. However, it is underexplored how to use CLIP for zero-shot recognition of highly specialized concepts, e.g., species of birds, plants, and animals, for which their scientific names are written in Latin or Greek. Indeed, CLIP performs poorly for zero-shot species recognition with prompts that use scientific names, e.g., “a photo of Lepus Timidus” (which is a scientific name in Latin). This is because these names are usually not included in CLIP’s training set. To improve performance, we explore using large-language models (LLMs) to generate descriptions (e.g., of species color and shape) and additionally use them in prompts. However, this method improves only marginally. Instead, we are motivated to translate scientific names (e.g., Lepus Timidus) to common English names (e.g., mountain hare) and use such in the prompts. We find that common names are more likely to be included in CLIP’s training set, and prompting them achieves 2~5 times higher accuracy on benchmarking datasets of fine-grained species recognition.</abstract>
<identifier type="citekey">parashar-etal-2023-prompting</identifier>
<identifier type="doi">10.18653/v1/2023.emnlp-main.610</identifier>
<location>
<url>https://aclanthology.org/2023.emnlp-main.610</url>
</location>
<part>
<date>2023-12</date>
<extent unit="page">
<start>9856</start>
<end>9861</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Prompting Scientific Names for Zero-Shot Species Recognition
%A Parashar, Shubham
%A Lin, Zhiqiu
%A Li, Yanan
%A Kong, Shu
%Y Bouamor, Houda
%Y Pino, Juan
%Y Bali, Kalika
%S Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F parashar-etal-2023-prompting
%X Trained on web-scale image-text pairs, Vision-Language Models (VLMs) such as CLIP can recognize images of common objects in a zero-shot fashion. However, it is underexplored how to use CLIP for zero-shot recognition of highly specialized concepts, e.g., species of birds, plants, and animals, for which their scientific names are written in Latin or Greek. Indeed, CLIP performs poorly for zero-shot species recognition with prompts that use scientific names, e.g., “a photo of Lepus Timidus” (which is a scientific name in Latin). This is because these names are usually not included in CLIP’s training set. To improve performance, we explore using large-language models (LLMs) to generate descriptions (e.g., of species color and shape) and additionally use them in prompts. However, this method improves only marginally. Instead, we are motivated to translate scientific names (e.g., Lepus Timidus) to common English names (e.g., mountain hare) and use such in the prompts. We find that common names are more likely to be included in CLIP’s training set, and prompting them achieves 2~5 times higher accuracy on benchmarking datasets of fine-grained species recognition.
%R 10.18653/v1/2023.emnlp-main.610
%U https://aclanthology.org/2023.emnlp-main.610
%U https://doi.org/10.18653/v1/2023.emnlp-main.610
%P 9856-9861
Markdown (Informal)
[Prompting Scientific Names for Zero-Shot Species Recognition](https://aclanthology.org/2023.emnlp-main.610) (Parashar et al., EMNLP 2023)
ACL