@inproceedings{sandhan-etal-2022-translist,
title = "{T}rans{LIST}: A Transformer-Based Linguistically Informed {S}anskrit Tokenizer",
author = "Sandhan, Jivnesh and
Singha, Rathin and
Rao, Narein and
Samanta, Suvendu and
Behera, Laxmidhar and
Goyal, Pawan",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.513",
doi = "10.18653/v1/2022.findings-emnlp.513",
pages = "6902--6912",
abstract = "Sanskrit Word Segmentation (SWS) is essential in making digitized texts available and in deploying downstream tasks. It is, however, non-trivial because of the sandhi phenomenon that modifies the characters at the word boundaries, and needs special treatment. Existing lexicon driven approaches for SWS make use of Sanskrit Heritage Reader, a lexicon-driven shallow parser, to generate the complete candidate solution space, over which various methods are applied to produce the most valid solution. However, these approaches fail while encountering out-of-vocabulary tokens. On the other hand, purely engineering methods for SWS have made use of recent advances in deep learning, but cannot make use of the latent word information on availability. To mitigate the shortcomings of both families of approaches, we propose Transformer based Linguistically Informed Sanskrit Tokenizer (TransLIST) consisting of (1) a module that encodes the character input along with latent-word information, which takes into account the sandhi phenomenon specific to SWS and is apt to work with partial or no candidate solutions, (2) a novel soft-masked attention to prioritize potential candidate words and (3) a novel path ranking algorithm to rectify the corrupted predictions. Experiments on the benchmark datasets for SWS show that TransLIST outperforms the current state-of-the-art system by an average 7.2 points absolute gain in terms of perfect match (PM) metric.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sandhan-etal-2022-translist">
<titleInfo>
<title>TransLIST: A Transformer-Based Linguistically Informed Sanskrit Tokenizer</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jivnesh</namePart>
<namePart type="family">Sandhan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rathin</namePart>
<namePart type="family">Singha</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Narein</namePart>
<namePart type="family">Rao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Suvendu</namePart>
<namePart type="family">Samanta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Laxmidhar</namePart>
<namePart type="family">Behera</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pawan</namePart>
<namePart type="family">Goyal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2022</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Sanskrit Word Segmentation (SWS) is essential in making digitized texts available and in deploying downstream tasks. It is, however, non-trivial because of the sandhi phenomenon that modifies the characters at the word boundaries, and needs special treatment. Existing lexicon driven approaches for SWS make use of Sanskrit Heritage Reader, a lexicon-driven shallow parser, to generate the complete candidate solution space, over which various methods are applied to produce the most valid solution. However, these approaches fail while encountering out-of-vocabulary tokens. On the other hand, purely engineering methods for SWS have made use of recent advances in deep learning, but cannot make use of the latent word information on availability. To mitigate the shortcomings of both families of approaches, we propose Transformer based Linguistically Informed Sanskrit Tokenizer (TransLIST) consisting of (1) a module that encodes the character input along with latent-word information, which takes into account the sandhi phenomenon specific to SWS and is apt to work with partial or no candidate solutions, (2) a novel soft-masked attention to prioritize potential candidate words and (3) a novel path ranking algorithm to rectify the corrupted predictions. Experiments on the benchmark datasets for SWS show that TransLIST outperforms the current state-of-the-art system by an average 7.2 points absolute gain in terms of perfect match (PM) metric.</abstract>
<identifier type="citekey">sandhan-etal-2022-translist</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.513</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.513</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>6902</start>
<end>6912</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TransLIST: A Transformer-Based Linguistically Informed Sanskrit Tokenizer
%A Sandhan, Jivnesh
%A Singha, Rathin
%A Rao, Narein
%A Samanta, Suvendu
%A Behera, Laxmidhar
%A Goyal, Pawan
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Findings of the Association for Computational Linguistics: EMNLP 2022
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F sandhan-etal-2022-translist
%X Sanskrit Word Segmentation (SWS) is essential in making digitized texts available and in deploying downstream tasks. It is, however, non-trivial because of the sandhi phenomenon that modifies the characters at the word boundaries, and needs special treatment. Existing lexicon driven approaches for SWS make use of Sanskrit Heritage Reader, a lexicon-driven shallow parser, to generate the complete candidate solution space, over which various methods are applied to produce the most valid solution. However, these approaches fail while encountering out-of-vocabulary tokens. On the other hand, purely engineering methods for SWS have made use of recent advances in deep learning, but cannot make use of the latent word information on availability. To mitigate the shortcomings of both families of approaches, we propose Transformer based Linguistically Informed Sanskrit Tokenizer (TransLIST) consisting of (1) a module that encodes the character input along with latent-word information, which takes into account the sandhi phenomenon specific to SWS and is apt to work with partial or no candidate solutions, (2) a novel soft-masked attention to prioritize potential candidate words and (3) a novel path ranking algorithm to rectify the corrupted predictions. Experiments on the benchmark datasets for SWS show that TransLIST outperforms the current state-of-the-art system by an average 7.2 points absolute gain in terms of perfect match (PM) metric.
%R 10.18653/v1/2022.findings-emnlp.513
%U https://aclanthology.org/2022.findings-emnlp.513
%U https://doi.org/10.18653/v1/2022.findings-emnlp.513
%P 6902-6912
Markdown (Informal)
[TransLIST: A Transformer-Based Linguistically Informed Sanskrit Tokenizer](https://aclanthology.org/2022.findings-emnlp.513) (Sandhan et al., Findings 2022)
ACL