4 papers
SWORD: Wikidata-based Distortions Reveal Hidden Cross-Lingual Inconsistencies in LLM Factual Error Rejection
Sanghyeok Park, Minji Kang, Hosung Kwak +1
Modern LLMs demonstrate impressive multilingual performance, yet standard benchmarks primarily reward selecting correct answers rather than evaluating genuine factual understanding…
Rollout-Level Advantage-Prioritized Experience Replay for GRPO
Gyeongtae Yoo, Sanghyeok Park, Soohyuk Jang +2
Reinforcement learning from verifiable rewards with GRPO is a standard approach for post-training reasoning LLMs. It remains sample inefficient. Each rollout is used for a single g…
Automating Code Generation for Semiconductor Equipment Control from Developer Utterances with LLMs
Youngkyoung Kim, Sanghyeok Park, Misoo Kim +3
Semiconductors form the backbone of modern electronics, with their manufacturing and testing relying on highly specialized equipment and domain-specific programming languages. Equi…
ODPG: Outfitting Diffusion with Pose Guided Condition
Seohyun Lee, Jintae Park, Sanghyeok Park
Virtual Try-On (VTON) technology allows users to visualize how clothes would look on them without physically trying them on, gaining traction with the rise of digitalization and on…