Hwiyeol Jo
Scholar

Hwiyeol Jo

Google Scholar ID: hFSHr0gAAAAJ
Research Scientist, at NAVER Cloud
AI with CogSciNLP using ML
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
236
 
H-index
7
 
i10-index
6
 
Publications
20
 
Co-authors
0
 
Contact
CVOpen ↗GitHubOpen ↗
Publications
7 items
ParaPairAudioBench: Paralinguistic Pairwise Audio Benchmark for LALM-as-a-Judge
2026
Cited
0
What Models Know, How Well They Know It: Knowledge-Weighted Fine-Tuning for Learning When to Say "I Don't Know"
2026
Cited
0
OmniACBench: A Benchmark for Evaluating Context-Grounded Acoustic Control in Omni-Modal Models
2026
Cited
0
Finding Answers in Thought Matters: Revisiting Evaluation on Large Language Models with Reasoning
2025
Cited
0
Enhancing Hallucination Detection via Future Context
2025
Cited
0
ZeroDL: Zero-shot Distribution Learning for Text Clustering via Large Language Models
arXiv.org · 2024
Cited
0
Taxonomy and Analysis of Sensitive User Queries in Generative AI Search
arXiv.org · 2024
Cited
0
Resume
Academic Achievements
  • - Preprint/Under Reviews:
  • - Finding Answers in Thought Matters: Revisiting Evaluation on Large Language Models with Reasoning
  • - HyperCLOVA X THINK Technical Report
  • - Accepted:
  • - ZeroDL: Zero-shot Distribution Learning for Text Clustering via Large Language Models (ACL2025 Findings)
  • - Taxonomy and Analysis of Sensitive User Queries in Generative AI Search (NAACL2025 Findings)
  • - Self-supervised Post-processing Method to Enrich Pretrained Word Vectors (EMNLP2023 Findings)
  • - A Self-Supervised Integration Method of Pretrained Language Models and Word Definitions (ACL2023 Findings)
  • - Modeling Mathematical Notation Semantics in Academic Papers (EMNLP2021 Findings)
  • - Devil’s Advocate: Novel Boosting Ensemble Method from Psychological Fi (incomplete)
Research Experience
  • - Research Scientist at NAVER Cloud
  • - Adjunct Professor at Konkuk University
  • - Core Contributor to several research projects including:
  • - Investigating the Influence of Prompt-Specific Shortcuts in AI Generated Text Detection
  • - Ruminating Word Representations with Random Masker, proposed an iterative training method for better word representation and model regularization
Education
  • - M.S. in Computer Science & Engineering, Seoul National University, Republic of Korea, Advisor: Byoung-Tak Zhang, Master Thesis: Data Augmentation Technique with Knowledge Extraction for Text Question Answering by Deep Neural Networks
  • - B.S. in Computer Science & Engineering, Konkuk University, Republic of Korea
  • - Military service at Third Republic Of Korea Army; TROKA
Background
  • - Research Interests: Artificial Intelligence with Cognitive Science, Machine Learning for Natural Language Processing to investigate human minds
  • - Research Keywords: Knowledge Transfer through Representation Learning, incorporating (lexical) semantics, information retrieval, general machine learning frameworks for self-supervised representation learning (Large Language Models are no more research topic but research tools)