Papers by Kyubyung Chae
Mitigating Hallucination in Abstractive Summarization with Domain-Conditional Mutual Information (2024.findings-naacl)
Copied to clipboard
| Challenge: | Prior studies have attempted to enhance faithfulness of abstractive summarization, yet hallucination remains a persistent challenge. |
| Approach: | They propose a decoding strategy that adjusts the generation probability of each token by comparing it with the token’s marginal probability within the domain of the source text. |
| Outcome: | The proposed method significantly improves faithfulness and source relevance on the XSUM dataset. |
Assessing Socio-Cultural Alignment and Technical Safety of Sovereign LLMs (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Recent trends in LLMs development show growing interest in the use and application of sovereign LLM models. |
| Approach: | They propose a framework for extracting and evaluating socio-cultural elements of sovereign LLMs and assess their technical robustness. |
| Outcome: | The proposed framework assesses the socio-cultural elements of sovereign LLMs and their technical robustness. |
Evaluating Structure-Aware Retrieval and Safety in Statute-Centric Legal QA (2026.acl-long)
Copied to clipboard
Kyubyung Chae, Jewon Yeom, Jeongjae Park, Seunghyun Bae, Ijun Jang, Hyunbin Jin, Jinkwan Jang, Taesup Kim
| Challenge: | Legal QA benchmarks focus on case law, overlooking statute-centric regulatory reasoning . relevant evidence is distributed across hierarchically linked documents, creating statutory retrieval gap . |
| Approach: | They propose a structure- and safety-aware benchmark for statute-centric legal QA . the benchmark assesses whether models can retrieve hierarchically fragmented evidence . |
| Outcome: | The proposed benchmark evaluates whether models can retrieve hierarchically fragmented evidence and safely abstain when statutory context is insufficient. |
Model-based Preference Optimization in Abstractive Summarization without Human Feedback (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) can generate fluent summaries but often introduce inaccuracies by hallucinating content not found in the source document. |
| Approach: | They propose a method to fine-tune Large Language Models for improved summarization abilities without any human feedback. |
| Outcome: | The proposed method significantly improves the quality of generated summaries without any human feedback. |