27 citations · 33 across the 9 of their papers we have counts for
12 papers
Stability-Aware Prompt Optimization for Clinical Data Abstraction
Arinbjörn Kolbeinsson, Daniel Timbie, Sajjan Narsinghani +1
Large language models used for clinical abstraction are sensitive to prompt wording, yet most work treats prompts as fixed and studies uncertainty in isolation. We argue these shou…
Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces
Mike A. Merrill, Alexander G. Shaw, Nicholas Carlini +82
AI agents may soon become capable of autonomously completing valuable, long-horizon tasks in diverse domains. Current benchmarks either do not measure real-world tasks, or are not…
Adversarial Negotiation Dynamics in Generative Language Models
Arinbjörn Kolbeinsson, Benedikt Kolbeinsson
Generative language models are increasingly used for contract drafting and enhancement, creating a scenario where competing parties deploy different language models against each ot…
Composable Interventions for Language Models
Arinbjorn Kolbeinsson, Kyle O'Brien, Tianjin Huang +8
Test-time interventions for language models can enhance factual accuracy, mitigate harmful outputs, and improve model efficiency without costly retraining. But despite a flood of n…
Generative models for wearables data
Arinbjörn Kolbeinsson, Luca Foschini
Data scarcity is a common obstacle in medical research due to the high costs associated with data collection and the complexity of gaining access to and utilizing data. Synthesizin…
Self-supervision of wearable sensors time-series data for influenza detection
Arinbjörn Kolbeinsson, Piyusha Gade, Raghu Kainkaryam +2
Self-supervision may boost model performance in downstream tasks. However, there is no principled way of selecting the self-supervised objectives that yield the most adaptable mode…
Patch-based Brain Age Estimation from MR Images
Kyriaki-Margarita Bintsi, Vasileios Baltatzis, Arinbjörn Kolbeinsson +2
Brain age estimation from Magnetic Resonance Images (MRI) derives the difference between a subject's biological brain age and their chronological age. This is a potential biomarker…
Biologically inspired architectures for sample-efficient deep reinforcement learning
Pierre H. Richemond, Arinbjörn Kolbeinsson, Yike Guo
Deep reinforcement learning requires a heavy price in terms of sample efficiency and overparameterization in the neural networks used for function approximation. In this work, we u…
How to Incorporate Monotonicity in Deep Networks While Preserving Flexibility?
Akhil Gupta, Naman Shukla, Lavanya Marla +2
The importance of domain knowledge in enhancing model performance and making reliable predictions in the real-world is critical. This has led to an increased focus on specific mode…
Adaptive Model Selection Framework: An Application to Airline Pricing
Naman Shukla, Arinbjörn Kolbeinsson, Lavanya Marla +1
Multiple machine learning and prediction models are often used for the same prediction or recommendation task. In our recent work, where we develop and deploy airline ancillary pri…
Dynamic Pricing for Airline Ancillaries with Customer Context
Naman Shukla, Arinbjörn Kolbeinsson, Ken Otwell +2
Ancillaries have become a major source of revenue and profitability in the travel industry. Yet, conventional pricing strategies are based on business rules that are poorly optimiz…
Tensor Dropout for Robust Learning
Arinbjörn Kolbeinsson, Jean Kossaifi, Yannis Panagakis +4
CNNs achieve remarkable performance by leveraging deep, over-parametrized architectures, trained on large datasets. However, they have limited generalization ability to data outsid…