Papers by Levon Haroutunian

3 papers
Improving Cross-Domain Low-Resource Text Generation through LLM Post-Editing: A Programmer-Interpreter Approach (2024.findings-eacl)

Copied to clipboard

Challenge: Large pre-trained language models such as GPT-3.5 and GPT-4 have gained significant attention in natural language research due to limited computational resources or inaccessible parameters.
Approach: They propose a neural programmer-interpreter approach that preserves the domain generalization ability of LLMs while editing their output.
Outcome: The proposed framework significantly improves GPT-3.5’s performance in logical form-to-text conversion and low-resource machine translation, surpassing other state-of-the-art (SOTA) LLM post-editing methods in cross-domain settings.
Ethical Considerations for Low-resourced Machine Translation (2022.acl-srw)

Copied to clipboard

Challenge: a paper examines the ethical implications of machine translation for low-resourced languages . a value scenario illustrates potential harms that low-rsourced language communities may face .
Approach: They propose to use Armenian as a case study to investigate ethical implications of machine translation for low-resourced languages.
Outcome: The proposed model is based on a value-scenario model of machine translation for low-resourced languages . the model is used to identify potential harms that low-income speakers may face .
Multilingual unsupervised sequence segmentation transfers to extremely low-resource languages (2022.acl-long)

Copied to clipboard

Challenge: Unsupervised sequence segmentation is a key component of low-resource languages where there is little or no gold-standard data on which to train supervised models.
Approach: They propose to pre-train a Masked Segmental Language Model multilingually to achieve unsupervised segmentation performance in extremely low-resource languages.
Outcome: The proposed model outperforms a monolingual model and a pre-trained model on Quechua in 6/10 settings.

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