AI Briefing#korean-nlp
The latest AI and developer news about #korean-nlp, with the original source and a short summary.
The latest AI and developer news about #korean-nlp, with the original source and a short summary.
kordoc, an open-source tool that converts HWP, PDF, and Office documents to Markdown and integrates with AI agents as an MCP server, has been released.
KOLongDoc, a VLM benchmark for evaluating comprehension of long Korean public institution documents, has been released.
The Kakao Kanana organization has released 'Kanana-v-embedding', a 2-billion parameter multimodal embedding model, demonstrating superior search performance compared to Jina Embeddings v4 through training on the Korean-specific dataset KoEmbed.
Kakao has open-sourced the 'Kanana Safeguard' series, consisting of three AI guardrail models specialized for Korean user environments, under the Apache 2.0 license.
Kanana Nano-2.1B maintains 93% of Kanana Essence's Korean performance despite a 26% size reduction.
APEACH, a Korean hate speech evaluation dataset generated via crowdsourcing, was released to address the bias and copyright issues of existing web-crawled data.
Achieves an F1 score of 97.11 with a CNN model based on the Sejong Corpus, offering fast performance without GPU requirements via C++ implementation