Scholar
Hyeonbin Hwang
Google Scholar ID: RoEV6T0AAAAJ
KAIST
Citations & Impact
Citations
377
H-index
6
i10-index
6
Publications
12
Co-authors
29
list available
Publications
Resume (English only)
Academic Achievements
- ['"Let's Predict Sentence by Sentence" - COLM 2025 RAM 2 WS (Oral)', '"Differential Information Distribution: A Bayesian Perspective on Direct Preference Optimization" - Preprint (under review)', '"The Coverage Principle: A Framework for Understanding Compositional Generalization" - Preprint (under review)', '"The CoT Encyclopedia: Analyzing, Predicting, and Controlling how a Reasoning Model will Think" - Preprint (under review)', '"BiGGen Bench: A Comprehensive Benchmark for Generative Language Models" - NAACL 2025 (Best Paper)', '"Knowledge Entropy Decay during Language Model Pretraining Hinders New Knowledge Acquisition" - ICLR 2025 (Oral), AAAI 2025 KnowFM WS (Best Paper)', '"Self-Explore to Avoid the Pit: Improving the Reasoning Capabilities of Language Models" - EMNLP 2024 (Findings), ACL 2024 NLRSE WS (Oral)']
Research Experience
- Intern at KRAFTON AI Companion Team (2025 - present)
- PUBG Ally Project (May 2025 - Feb 2025)
- KAIST AI - Research on instruction following, self-feedback, and fine-grained evaluation (2024 - present)
- LK Lab, KAIST - Research on instruction following, self-feedback, and fine-grained evaluation (2022 - 2024)
- NAVER Healthcare AI - EHR summarization (March 2022 - August 2022)
- CVL Lab, KAIST - 2D image detection and classification (June 2021 - September 2021)
Background
- A Master's student with research interests in LLM reasoning, synthetic data creation, and the design of new learning methods.
Miscellany
- Open to collaborations. Feel free to reach out!
