| Mobile Web

Search results for multimodal AI

AI & Enterprise

Kakao releases two Kanana-2 vision encoder models as open source

Kakao said on Tuesday it released two versions of its in-house Kanana-2 vision encoder as open source on Hugging Face. The models are a 426 million-parameter medium version and a 98 million-parameter tiny version. Kakao said its evaluations across 42 benchmarks in nine areas showed both models outperformed global vision encoders used for comparison. It also said training efficiency rose about fourfold and processing latency fell by more than 60 percent.

Telecommunications & Media

Korea broadcast watchdog to back overseas push for 11 AI-powered TV content titles

South Korea\'s Korea Communications Commission and the Korea Broadcasting and Communications Commission and KCA selected 11 domestic broadcast content titles using AI technology for an overseas distribution support programme. The selected works will be introduced to overseas investors at the Asia TV Forum & Market in Singapore on Dec. 3-5. The lineup includes four AI-based production support titles and seven AI-assisted planning programmes. The agencies will support investment consultations and promotion at ATF.

AI & Enterprise

ETRI, KAIST take top spot in robot navigation challenge

South Korean researchers won first place in an international competition with robot AI that understands human speech and navigates to a destination on its own. The Electronics and Telecommunications Research Institute and the Korea Advanced Institute of Science and Technology formed a joint team that topped the VLNVerse challenge at the ECCV 2026 computer vision conference in Malmo, Sweden, beating 112 teams. The team recorded a 90.7 percent average success rate and won on earlier submission time.