pith. sign in

Seong Joon Oh

Identifiers

  • name variant Seong Joon Oh 0.60 · backfill

Papers (15)

  1. It Takes Two: Complementary Self-Distillation for Contextual Integrity in LLMs cs.LG · 2026 · author #8
  2. Sparse Autoencoders enable Robust and Interpretable Fine-tuning of CLIP models cs.CV · 2026 · author #4
  3. MEME: Multi-entity & Evolving Memory Evaluation cs.LG · 2026 · author #5
  4. When Do Diffusion Models learn to Generate Multiple Objects? cs.CV · 2026 · author #4
  5. Privacy Collapse: Benign Fine-Tuning Can Break Contextual Privacy in Language Models cs.CL · 2026 · author #4
  6. Dr.LLM: Dynamic Layer Routing in LLMs cs.CL · 2025 · author #5
  7. Sequential Attacks on Agents for Long-Term Adversarial Goals cs.LG · 2018 · author #2
  8. Natural and Effective Obfuscation by Head Inpainting cs.CV · 2017 · author #3
  9. Towards Reverse-Engineering Black-Box Neural Networks stat.ML · 2017 · author #1
  10. Person Recognition in Personal Photo Collections cs.CV · 2017 · author #1
  11. Generating Descriptions with Grounded and Co-Referenced People cs.CV · 2017 · author #4
  12. Adversarial Image Perturbation for Privacy Protection -- A Game Theory Perspective cs.CV · 2017 · author #1
  13. Exploiting saliency for object segmentation from image level labels cs.CV · 2017 · author #1
  14. Faceless Person Recognition; Privacy Implications in Social Media cs.CV · 2016 · author #1
  15. Person Recognition in Personal Photo Collections cs.CV · 2015 · author #1

Mentions

  • 2605.20258 #8 · arxiv_oai · confidence 0.70 Seong Joon Oh
  • 2510.12773 #5 · arxiv_oai · confidence 0.70 Seong Joon Oh
  • 2605.15961 #4 · arxiv_oai · confidence 0.70 Seong Joon Oh

Frequent Coauthors