Myungsub Choi
  • Bio
  • News
  • Experience
  • Publications
  • Talks
  • Projects
ESC

Searching...

No results found

↑↓ Navigate ↵ Select
Powered by Hugo Blox
  • Talks
    • AI 에이전트의 런타임 신뢰성: 검증, 방어, 예측
    • As AI Gets Smarter, How Far Can Enterprises Delegate?
    • AI의 빈틈을 메우는 스타트업: 양질의 데이터·검증·금융 규제·거버넌스로 완성되는 신뢰
    • From Focusing Lenses to Focusing Risks: An Engineer's Journey into Agentic AI Safety
    • Cracking the Enterprise Code: How GenAI Startups Can Win Enterprise Deals
    • 신뢰할 수 있는 AI는 어떻게 만들어지는가 — Mission-Critical 환경의 RAG, Judge, 그리고 Agent 통제
    • Tynapse Startup Pitch at the 2026 AWS AI Pitch Competition
  • Publications
    • Beyond the Verdict: Evidence-Aligned Evaluation of Visual Prompt-Injection Guardrails
    • Risk Is Not Review Value: Wrong-Answer Exposure Under Bounded Review Budgets
    • Cross-Agent Campaign Attribution: Linking Asynchronous Attacks Across LLM Agents
    • Stable Autofocus with Focal Consistency Loss
    • Exploring positional characteristics of dual-pixel data for camera autofocus
    • Learning to learn task-adaptive hyperparameters for few-shot learning
    • Referring object manipulation of natural images with conditional classifier-free guidance
    • Visual tracking by adaptive continual meta-learning
    • Dynavsr: Dynamic adaptive blind video super-resolution
    • Motion-aware dynamic architecture for efficient frame interpolation
    • Searching for controllable image restoration networks
    • Test-time adaptation for video frame interpolation via meta-learning
    • Channel Attention Is All You Need for Video Frame Interpolation
    • Meta-Learning with Adaptive Hyperparameters
    • Scene-adaptive video frame interpolation via meta-learning
    • AIM 2019 challenge on video temporal super-resolution: Methods and results
    • Task-aware image downscaling
  • Blog
    • 📄 A paper on evaluating visual prompt-injection guardrails is accepted in the BEAM 2 workshop at ECCV 2026 as an Oral presentation
    • 📄 A paper on review-budget-aware LLM evaluation is accepted in the SeT-LLM workshop at KDD 2026
    • 📄 A paper on cross-agent attack attribution is accepted in the AIWILD workshop at ICML 2026
    • 📄 A paper on stabilizing autofocus is accepted in WACV 2025 as an Oral presentation
    • 📄 A paper on autofocus is accepted in ICCV 2023.
  • Experience
  • Projects
  • Projects
    • Meta-VFI
    • CAIN
    • awesome-rnn

AI 에이전트의 런타임 신뢰성: 검증, 방어, 예측

Sep 16, 2026·
Myungsub Choi
Myungsub Choi
· 0 min read
Article
Date
Sep 16, 2026 7:00 PM
Event
YAI (Yonsei University AI Society)
Location

Yonsei University, Seoul, Korea

Last updated on Sep 16, 2026
Talk
Myungsub Choi
Authors
Myungsub Choi
Head of AI
As AI Gets Smarter, How Far Can Enterprises Delegate? Sep 9, 2026 →

© 2026 Myungsub Choi. This work is licensed under CC BY NC ND 4.0

Made with Hugo Blox — Open Source. Start free →