Human-Inspired AI Research

Advanced AI Technology Showcase

Pioneering the future of artificial intelligence through cutting-edge research and innovation

Our Technologies

Check out our state-of-the-art AI models and systems designed to push the boundaries of what's possible

2026

Current research highlights

Document AI

A human-centered document intelligence technology that transforms unstructured multimodal documents into structured knowledge for reliable search, retrieval, and reasoning.

PPFT

A privacy-preserving LLM technology that performs inference and domain adaptation using obfuscated embeddings, without sending raw sensitive text to the server.

SYNAPSE 2.0

An agentic knowledge system that structures information into graphs and hierarchical files, then searches and reasons over them to deliver grounded answers.

2025

Selected showcase lineup

Cross-Embedding

A specialized cross-lingual embedding model for cross-lingual search, bridging languages with unprecedented accuracy and efficiency.

Reasoning KULLM

This next-generation large language model effectively solves complex problems and provides rapid responses with sophisticated Korean expressions and enhanced reasoning capabilities.

RAG Specialized KULLM

Retrieval-Augmented Generation model optimized for knowledge-intensive tasks with superior context understanding.

MAGIC

A one-stop AI-powered solution for generating and translating multilingual YouTube subtitles.

About HIAI Research

Leading AI innovation through human-inspired intelligence

Best Team

  • Over 70% PhD-level workforce composition
  • More than 10 professor-level experts
  • Top specialists from global companies like Samsung, Upstage, NC
  • All members with 10+ years of NLP & AI expertise

Best Performance

  • Recognition as one of the 100 Outstanding National R&D Achievements of 2025 for technologies that ensure the truthfulness and consistency of generative AI outputs
  • World's first large language model KULLM development in academia
  • KURE - highest performing Korean embedding model
  • World 1st place in WMT Machine Translation Competition (Quality Estimation Shared Task 2022)
  • Domestic 1st place in Microsoft AI for Accessibility Hackathon
  • 6 consecutive years of excellence awards at Korean Language Processing Conference

Research Leadership

  • ACL 2026 Main Conference acceptance for 'NRLB' an AI summarization technique anyone can understand, with 12 additional papers from our lab accepted in the same cycle
  • Technical cooperation with various domestic and international companies and academia (Naver, KT, NC, Upstage, MIT, etc.)
  • Over 500 published papers domestically and internationally (ACL, EMNLP, NAACL, EACL, COLING, etc.)
  • More than 150 patent applications and registrations
  • Technology transfer exceeding 3 billion KRW
  • Research funding exceeding 40 billion KRW
  • Operating with 30 partner companies through family business agreements
  • Participating as Open Ko-LLM Leaderboard Data Sponsor

Contact Us

Get in touch with our research team

ADDRESS

Room 108, CJ Food Safety Hall, Korea University Anam Campus, 145 Anam-ro, Seongbuk-gu, Seoul, Republic of Korea