Cheoneum Park

Latest

• CoinRAG: Contextualized Information Nugget KV Cache Reuse for Long-Context RAG
• Lost in Mimicry: Verified Training-Free Post-Editing for LLM Translationese
• Global Optimization and Inference-Time Region Grafting for Agentic Workflows
• ScholarBench: A Bilingual Benchmark for Abstraction, Comprehension, and Reasoning Evaluation in Academic Contexts