Hyesu Lim
  • Home
  • News
  • Publications
  • Contact
  • CV
  • Publications
    • Beyond SAEs: Consistent Concept Discovery with Sparse Coding
    • VisualScratchpad: Grounding Visual Concepts in Large Vision Language Models
    • ConceptScope: Characterizing Dataset Bias via Disentangled Visual Concepts
    • CytoSAE: Interpretable Cell Embeddings for Hematology
    • Sparse Autoencoders Reveal Selective Remapping of Visual Concepts During Adaptation
    • Towards Calibrated Robust Fine-Tuning of Vision-Language Models
    • TTN: A Domain-Shift Aware Batch Normalization in Test-Time Adaptation
    • AVocaDo: Strategy for Adapting Vocabulary to Downstream Domain
  • Writing
    • Welcome to My New Site

VisualScratchpad: Grounding Visual Concepts in Large Vision Language Models

April 1, 2026·
Hyesu Lim
,
Jinho Choi
,
Taekyung Kim
,
Byeongho Heo
,
Jaegul Choo
,
Dongyoon Han
· 0 min read
PDF Code Project Poster
projects interpretability
Last updated on April 1, 2026
Vision-Language Models Interpretability
← Beyond SAEs: Consistent Concept Discovery with Sparse Coding October 16, 2026
ConceptScope: Characterizing Dataset Bias via Disentangled Visual Concepts December 1, 2025 →

Made with Hugo Blox. Duplicate this template →