188 citations · 199 across the 16 of their papers we have counts for
19 papers
Can AI agents conduct open-ended AI research? Early evidence from two case studies
Peter Kirgis, Sayash Kapoor, Andrew Schwartz +21
Forecasts of explosive AI progress hinge on AI agents automating AI research. But evidence on whether agents can carry out open-ended AI research is thin. Current evaluations eithe…
Life After Benchmark Saturation: A Case Study of CORE-Bench
Nitya Nadgir, Sayash Kapoor, Kangheng Liu +11
When a benchmark's accuracy saturates, it is often retired and replaced with a more challenging version. We show that this approach privileges accuracy and misses the opportunity t…
Open-World Evaluations for Measuring Frontier AI Capabilities
Sayash Kapoor, Peter Kirgis, Andrew Schwartz +15
Benchmark-based evaluation remains important for tracking frontier AI progress. But it can both overstate and understate deployed capability because it privileges tasks that can be…
Log analysis is necessary for credible evaluation of AI agents
Peter Kirgis, Sayash Kapoor, Stephan Rabanser +8
Agent benchmarks typically report only final outcomes: pass or fail. This threatens evaluation credibility in three ways. First, scores may be inflated or deflated by shortcuts and…
Towards a Science of AI Agent Reliability
Stephan Rabanser, Sayash Kapoor, Peter Kirgis +3
AI agents are increasingly deployed to execute important tasks. While rising accuracy scores on standard benchmarks suggest rapid progress, many agents still continue to fail in pr…
What Does It Take to Build a Performant Selective Classifier?
Stephan Rabanser, Nicolas Papernot
Selective classifiers improve model reliability by abstaining on inputs the model deems uncertain. However, few practical approaches achieve the gold-standard performance of a perf…
Uncertainty-Driven Reliability: Selective Prediction and Trustworthy Deployment in Modern Machine Learning
Stephan Rabanser
Machine learning (ML) systems are increasingly deployed in high-stakes domains where reliability is paramount. This thesis investigates how uncertainty estimation can enhance the s…
Cascadia: An Efficient Cascade Serving System for Large Language Models
Youhe Jiang, Fangcheng Fu, Wanru Zhao +4
Recent advances in large language models (LLMs) have intensified the need to deliver both rapid responses and high-quality outputs. More powerful models yield better results but in…
Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention
Stephan Rabanser, Ali Shahin Shamsabadi, Olive Franzese +3
Cautious predictions -- where a machine learning model abstains when uncertain -- are crucial for limiting harmful errors in safety-critical applications. In this work, we identify…
Suitability Filter: A Statistical Framework for Classifier Evaluation in Real-World Deployment Settings
Angéline Pouget, Mohammad Yaghini, Stephan Rabanser +1
Deploying machine learning models in safety-critical domains poses a key challenge: ensuring reliable model performance on downstream user data without access to ground truth label…
Gatekeeper: Improving Model Cascades Through Confidence Tuning
Stephan Rabanser, Nathalie Rauschmayr, Achin Kulshrestha +5
Large-scale machine learning models deliver strong performance across a wide range of tasks but come with significant computational and resource constraints. To mitigate these chal…
Robust and Actively Secure Serverless Collaborative Learning
Olive Franzese, Adam Dziedzic, Christopher A. Choquette-Choo +7
Collaborative machine learning (ML) is widely used to enable institutions to learn better models from distributed data. While collaborative approaches to learning intuitively prote…
Training Private Models That Know What They Don't Know
Stephan Rabanser, Anvith Thudi, Abhradeep Thakurta +2
Training reliable deep learning models which avoid making overconfident but incorrect predictions is a longstanding challenge. This challenge is further exacerbated when learning h…
-DkNN: Out-of-Distribution Detection Through Statistical Testing of Deep Representations
Adam Dziedzic, Stephan Rabanser, Mohammad Yaghini +3
The lack of well-calibrated confidence estimates makes neural networks inadequate in safety-critical domains such as autonomous driving or healthcare. In these settings, having the…
Intrinsic Anomaly Detection for Multi-Variate Time Series
Stephan Rabanser, Tim Januschowski, Kashif Rasul +6
We introduce a novel, practically relevant variation of the anomaly detection problem in multi-variate time series: intrinsic anomaly detection. It appears in diverse practical sce…
Selective Prediction via Training Dynamics
Stephan Rabanser, Anvith Thudi, Kimia Hamidieh +5
Selective Prediction is the task of rejecting inputs a model would predict incorrectly on. This involves a trade-off between input space coverage (how many data points are accepted…
The Effectiveness of Discretization in Forecasting: An Empirical Study on Neural Time Series Models
Stephan Rabanser, Tim Januschowski, Valentin Flunkert +2
Time series modeling techniques based on deep learning have seen many advancements in recent years, especially in data-abundant settings and with the central aim of learning global…
Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift
Stephan Rabanser, Stephan Günnemann, Zachary C. Lipton
We might hope that when faced with unexpected inputs, well-designed software systems would fire off warnings. Machine learning (ML) systems, however, which depend strongly on prope…
Introduction to Tensor Decompositions and their Applications in Machine Learning
Stephan Rabanser, Oleksandr Shchur, Stephan Günnemann
Tensors are multidimensional arrays of numerical values and therefore generalize matrices to multiple dimensions. While tensors first emerged in the psychometrics community in the…