Papers by Hongzhi Xu

8 papers
Benchmarking the Performance of Machine Translation Evaluation Metrics with Chinese Multiword Expressions (2024.lrec-main)

Copied to clipboard

Challenge: Multiword Expressions (MWEs) are hard nuts for many natural language processing tasks.
Approach: They annotate 28 types of Chinese MWEs and then examine 31 MTE metrics on groups of sentences containing different MWE.
Outcome: The results show that MT systems and MTE metrics still suffer from MWEs .
Annotating Chinese Light Verb Constructions according to PARSEME guidelines (L18-1)

Copied to clipboard

Challenge: Using existing resources, we can annotate Chinese multiword expressions using PARSEME guidelines.
Approach: They propose to use an existing resource containing Chinese light verbs to make an annotation of a Chinese UD treebank in two steps.
Outcome: The proposed annotations are based on an existing treebank containing Chinese light verbs and are consistent with the proposed guidelines.
A Deep Analysis of the Impact of Multiword Expressions and Named Entities on Chinese-English Machine Translations (2024.findings-emnlp)

Copied to clipboard

Challenge: a study on the impact of multiword expressions and multiword named entities (NEs) on the performance of Chinese-English machine translation systems is presented.
Approach: They propose to use Chinese multiword expressions and multiword named entities (NEs) to evaluate machine translation performance.
Outcome: The proposed methods show that Chinese-English machine translation systems perform significantly worse on Chinese sentences with most kinds of MWEs and NEs.
Modeling Morphological Typology for Unsupervised Learning of Language Morphology (2020.acl-main)

Copied to clipboard

Challenge: Existing approaches to morphological analysis relied on hand-built rules to identify word-internal structures.
Approach: They propose a language-independent model for fully unsupervised morphological analysis that exploits a universal framework leveraging morphology.
Outcome: The proposed model outperforms existing systems on nine typologically and genetically diverse languages and shows superior performance over leading systems.
Large-Scale Relation Learning for Question Answering over Knowledge Bases with Pre-trained Language Models (2021.emnlp-main)

Copied to clipboard

Challenge: Existing KBQA methods focus on the natural language but ignore textual information carried by the nodes and edges.
Approach: They propose to perform relation extraction, relation matching, and relation reasoning tasks to align the natural language expressions to the relations in the KB and reason over the missing connections.
Outcome: Experiments on WebQSP show that the proposed model outperforms baselines even when the KB is incomplete.
Morphological Segmentation for Low Resource Languages (2020.lrec-1)

Copied to clipboard

Challenge: a new corpus of annotated morphological data is described for the DARPA LORELEI Program . the data is annotating 9 low resource languages and root information for 7 of the languages .
Approach: This paper describes a new morphology resource created by Linguistic Data Consortium and the University of Pennsylvania for the DARPA LORELEI Program.
Outcome: The annotated corpus provides a gold standard for unsupervised morphological segmenters and analyzers . the language-specific annotation guidelines were language-independent, but included morphology paradigms and other specifications.
Annotating Chinese Word Senses with English WordNet: A Practice on OntoNotes Chinese Sense Inventories (2024.lrec-main)

Copied to clipboard

Challenge: a recent study has shown that large language models can be useful for cross-lingual applications.
Approach: They propose to annotate Chinese word senses using English WordNet synsets . they examine the relationship between two annotators and find patterns among synset .
Outcome: The proposed method shows that the annotators agree on 38% of the synsets compared with the original synset . the results highlight similarities between the synnotated synset and the WordNet structure .
Unsupervised Morphology Learning with Statistical Paradigms (C18-1)

Copied to clipboard

Challenge: Existing models treat words as concatenation of morphemes, but some use transformations like rewrite rules to recognize dependencies between morphs.
Approach: They propose an unsupervised model that exploits the notion of paradigms for morphological segmentation that can be applied to a homogeneous set of words.
Outcome: The proposed model significantly improves on the Morpho-Challenge dataset in English, Turkish, and Finnish.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations