LatinPipe

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F24%3A10493029" target="_blank" >RIV/00216208:11320/24:10493029 - isvavai.cz</a>

Výsledek na webu

<a href="https://github.com/ufal/evalatin2024-latinpipe" target="_blank" >https://github.com/ufal/evalatin2024-latinpipe</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

LatinPipe

Popis výsledku v původním jazyce

LatinPipe is the winning submission to the EvaLatin 2024 Dependency Parsing shared task. LatinPipe performs state-of-the-art language analysis for Latin. Our system consists of a fine-tuned concatenation of base and large pre-trained LMs, with a dot-product attention head for parsing and softmax classification heads for morphology to jointly learn both dependency parsing and morphological analysis. It is trained by sampling from seven publicly available Latin corpora, utilizing additional harmonization of annotations to achieve a more unified annotation style. Before fine-tuning, we train the system for a few initial epochs with frozen weights. We also add additional local relative contextualization by stacking the BiLSTM layers on top of the Transformer(s). The source code is available at https://github.com/ufal/evalatin2024-latinpipe. LatinPipe demo and REST API webservice are available at https://lindat.mff.cuni.cz/services/udpipe/. The fine-tuned model for Latin is available at http://hdl.handle.n

Název v anglickém jazyce

LatinPipe

Popis výsledku anglicky

LatinPipe is the winning submission to the EvaLatin 2024 Dependency Parsing shared task. LatinPipe performs state-of-the-art language analysis for Latin. Our system consists of a fine-tuned concatenation of base and large pre-trained LMs, with a dot-product attention head for parsing and softmax classification heads for morphology to jointly learn both dependency parsing and morphological analysis. It is trained by sampling from seven publicly available Latin corpora, utilizing additional harmonization of annotations to achieve a more unified annotation style. Before fine-tuning, we train the system for a few initial epochs with frozen weights. We also add additional local relative contextualization by stacking the BiLSTM layers on top of the Transformer(s). The source code is available at https://github.com/ufal/evalatin2024-latinpipe. LatinPipe demo and REST API webservice are available at https://lindat.mff.cuni.cz/services/udpipe/. The fine-tuned model for Latin is available at http://hdl.handle.n

Druh

R - Software

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/GX20-16819X" target="_blank" >GX20-16819X: Porozumění jazyku: od syntaxe k diskurzu</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Interní identifikační kód produktu

[https://github.com/ufal/evalati

Technické parametry

Výsledek volně dostupný na adrese https://github.com/ufal/evalatin2024-latinpipe.

Ekonomické parametry

105390

IČO vlastníka výsledku

00216208

Název vlastníka

Univerzita Karlova