Dr. Been Kim
Senior Staff Research Scientist
Verified Proof of Work Artifacts
2 items catalogedEach artifact below represents an authenticated research publication, production code repository, or technical architectural framework directly authored or co-created by Dr. Been Kim. Every entry undergoes editorial source verification.
Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)
Invented TCAV (Testing with Concept Activation Vectors), an interpretability framework that uses directional derivatives in activation space to quantify how much a human-understandable concept (e.g., stripes on a zebra) contributes to a model's prediction.
Requires user-provided exemplars for concept definitions; high-level abstract concepts without clear visual or lexical exemplars can be difficult to vectorize cleanly.
Relative Representations Enable Non-Degenerate Latent Space Alignment
Co-authored breakthrough research demonstrating that representations can be compared across disparate neural networks by computing pairwise angles and similarities to anchor points, proving latent geometries are invariant across architectures.
Choice and distribution of anchor points can introduce variance in representation reconstruction accuracy across domain shifts.