Gabriel Mukobi

3 papers · 2024–2025 · 2 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🐣 Hot Topic Early Bird 👥 Mega-Team · Squad ❓ The Questioner (2)

Venues

ICML (2) NeurIPS (1)

Top co-authors

Steven Basart (2) Alice Gatti (2) Alexander Pan (2) Stephen Fitz (2) Mantas Mazeika (2) Dan Hendrycks (2) Long Phan (1) Adam Alfred Hunt (1) David Campbell (1) Uday Tupakula (1)

Research topics

Artificial Intelligence (3) Machine Learning (3) Core AI (3) Natural Language Processing (2) Applications (2) Optimization & Theory · Machine Learning (2) Deep Learning (1) Core Methods (1) Learning Types (1) Optimization & Theory · Deep Learning (1)

Keywords

benchmark evaluation (1) ai safety (1) spectral analysis (1) model scaling (1) safety benchmark (1) capabilities component (1) scaling behavior (1) probability mass (1) large language model (1) negative log-likelihood (1) model capabilities (1) scale correlation (1) multiple-choice benchmark (1) unlearning (1) hazardous knowledge (1) biosecurity (1) proxy measurement (1) wmdp benchmark (1) frontier ai models (1) predictable evaluations (1)

Papers

Why Has Predicting Downstream Capabilities of Frontier AI Models with Scale Remained Elusive? ICML 2025 Safetywashing: Do AI Safety Benchmarks Actually Measure Safety Progress? NeurIPS 2024 The WMDP Benchmark: Measuring and Reducing Malicious Use with Unlearning ICML 2024