Papers with direct generation method

2 papers
Can LLM Generate Culturally Relevant Commonsense QA Data? Case Study in Indonesian and Sundanese (2024.emnlp-main)

Copied to clipboard

Challenge: Large Language Models (LLMs) are increasingly being used to generate synthetic data for training and evaluating models.
Approach: They investigate the effectiveness of using Large Language Models to generate culturally relevant commonsense QA datasets for Indonesian and Sundanese languages using both LLMs and human annotators.
Outcome: The proposed model generates 4.5K questions per language, compared with 4.5k for Indonesian and 4.5km for Sundanese.
Threading the Needle: Reweaving Chain-of-Thought Reasoning to Explain Human Label Variation (2025.emnlp-main)

Copied to clipboard

Challenge: Recent advances in large language models have shown the power of chain-of-thought reasoning in improving complex decision-making tasks.
Approach: They propose a pipeline that generates chain-of-thought (CoT) explanations from CoTs with improved accuracy.
Outcome: The proposed pipeline outperforms a direct generation method and baselines on three datasets.

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