Skip To Main Content
Been Kim on Muck Rack

Been Kim

As seen in: Medium, AI Magazine
Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work. Claim your profile
Back to Been Kim’s profile

Articles by Been Kim

QuestBench: Can LLMs ask the right question to acquire information in reasoning tasks?

|
By Been Kim, Zi Wang
| arXiv
Open in Who Shared Issue with article?

[2502.07586] We Can't Understand AI Using our Existing Vocabulary

|
By Robert Geirhos, Been Kim
| arXiv
Open in Who Shared Issue with article?

developing a language to shape our relationships with AI

|
By Been Kim
| Medium
By Been Kim, Google Research, Brain Team AI can be found in many applications as varied as protein structure and function prediction from sequence (e.g., AlphaFold and ProtENN), language understanding and generation (e.g., GPT-3 and PaLM), and many more. AI is also making important decisions.
Open in Who Shared Issue with article?

[1906.02715] Visualizing and Measuring the Geometry of BERT

|
By Ann Yuan, Been Kim, Adam Pearce, Emily Reif
| arXiv
Authors:Andy Coenen, Emily Reif, Ann Yuan, Been Kim, Adam Pearce, Fernanda Viégas, Martin Wattenberg (Submitted on 6 Jun 2019) Abstract: Transformer architectures show significant promise for natural language processing. Given that a single pretrained model can be fine-tuned to perform well on many different tasks, these networks appear to extract generally useful linguistic features. A natural question is how such networks represent this information internally.
Open in Who Shared Issue with article?

Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)

|
By Been Kim, Martin P. Wattenberg, Justin Gilmer, Carrie Cai
| ai.google
The interpretation of deep learning models is a challenge due to their size, complexity, and often opaque internal state. In addition, many systems, such as image classifiers, operate on low-level features rather than high-level concepts. To address these challenges, we introduce Concept Activation Vectors (CAVs), which provide an interpretation of a neural net's internal state in terms of human-friendly concepts.
Open in Who Shared Issue with article?

[1806.10758] Evaluating Feature Importance Estimates

|
By Been Kim, Dumitru Erhan, Pieter-Jan Kindermans
| arXiv
Abstract: Estimating the influence of a given feature to a model prediction is challenging. We introduce ROAR, RemOve And Retrain, a benchmark to evaluate the accuracy of interpretability methods that estimate input feature importance in deep neural networks.
Open in Who Shared Issue with article?

[1802.00682] How do Humans Understand Explanations from Machine Learning Systems? An Evaluation of the Human-Interpretability of Explanation

|
By Emily Chen, Been Kim, Jeffrey He, Sam Gershman
| arXiv
Abstract: Recent years have seen a boom in interest in machine learning systems that can provide a human-understandable rationale for their predictions or decisions. However, exactly what kinds of explanation are truly human-interpretable remains poorly understood. This work advances our understanding of what makes explanations interpretable in the specific context of verification.
Open in Who Shared Issue with article?

[1702.08608] Towards A Rigorous Science of Interpretable Machine Learning

|
By Been Kim
| arXiv
(Submitted on 28 Feb 2017 (v1), last revised 2 Mar 2017 (this version, v2)) Submission history From: Been Kim [view email] [v1] Tue, 28 Feb 2017 02:19:20 GMT (97kb,D) [v2] Thu, 2 Mar 2017 19:32:10 GMT (663kb,D)
Open in Who Shared Issue with article?
Show More