Papers by Huacheng Song

3 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 .
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.
PhonoThink: Improving Large Language Models’ Reasoning on Chinese Phonological Ambiguities (2025.emnlp-main)

Copied to clipboard

Challenge: Effectively resolving phonological ambiguities is crucial for robust natural language processing, as these ambiguity are pervasive in tasks ranging from speech-to-text, spelling correction, to offensive language detection.
Approach: They propose a framework to enhance LLMs’ phonological capability through a multiple-stage training approach.
Outcome: The proposed framework enables the base model to achieve comparable performance to a much larger model.

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