114 citations · 326 across the 19 of their papers we have counts for
22 papers
Transect: Retaining Observability for Long-Horizon LLM Agent Evaluations
Toby D. Pilditch, Konstantinos Voudouris, Alexandra Abbas +1
Frontier AI evaluations increasingly use open-ended, agentic, long-horizon tasks whose transcripts can span hundreds of pages of outputs and actions from complex multi-agent networ…
You Are What You Read: Misalignment via In-Context Persona Induction
Kyuhee Kim, Benjamin Berczi, Cozmin Ududec
Broad misalignment has been produced by finetuning on narrow data, harmful or benign, and in context only by demonstrations of the undesirable behaviour itself. We show that benign…
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…
How Inference Compute Shapes Frontier LLM Evaluation
Jessica McFadyen, Ole Jorgensen, Harry Coppock +2
AI evaluations are shifting toward harder tasks that benefit from longer trajectories involving tool use and iterative problem solving. As a result, performance is increasingly sen…
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…
Seven simple steps for log analysis in AI systems
Magda Dubois, Ekin Zorer, Maia Hamin +17
AI systems produce large volumes of logs as they interact with tools and users. Analysing these logs can help understand model capabilities, propensities, and behaviours, or assess…
Ask don't tell: Reducing sycophancy in large language models
Magda Dubois, Cozmin Ududec, Christopher Summerfield +1
Sycophancy, the tendency of large language models to favour user-affirming responses over critical engagement, has been identified as an alignment failure, particularly in high-sta…
Measuring what Matters: Construct Validity in Large Language Model Benchmarks
Andrew M. Bean, Ryan Othniel Kearns, Angelika Romanou +39
Evaluating large language models (LLMs) is crucial for both assessing their capabilities and identifying safety or robustness issues prior to deployment. Reliably measuring abstrac…
Establishing Best Practices for Building Rigorous Agentic Benchmarks
Yuxuan Zhu, Tengjun Jin, Yada Pruksachatkun +22
Benchmarks are essential for quantitatively tracking progress in AI. As AI agents become increasingly capable, researchers and practitioners have introduced agentic benchmarks to e…
Lessons from a Chimp: AI "Scheming" and the Quest for Ape Language
Christopher Summerfield, Lennart Luettgau, Magda Dubois +9
We examine recent research that asks whether current AI systems may be developing a capacity for "scheming" (covertly and strategically pursuing misaligned goals). We compare curre…
Skewed Score: A statistical framework to assess autograders
Magda Dubois, Harry Coppock, Mario Giulianelli +3
The evaluation of large language model (LLM) outputs is increasingly performed by other LLMs, a setup commonly known as "LLM-as-a-judge", or autograders. While autograders offer a…
HiBayES: A Hierarchical Bayesian Modeling Framework for AI Evaluation Statistics
Lennart Luettgau, Harry Coppock, Magda Dubois +2
As Large Language Models (LLMs) and other AI systems evolve, robustly estimating their capabilities from inherently stochastic outputs while systematically quantifying uncertainty…
Self-duality and Jordan structure of quantum theory follow from homogeneity and pure transitivity
Howard Barnum, Cozmin Ududec, John van de Wetering
Quantum theory satisfies the mathematically powerful property of self-duality, meaning that its state and effect spaces are made isomorphic by an inner product. This is why both qu…
Learning an Optimally Reduced Formulation of OPF through Meta-optimization
Alex Robson, Mahdi Jamei, Cozmin Ududec +1
With increasing share of renewables in power generation mix, system operators would need to run Optimal Power Flow (OPF) problems closer to real-time to better manage uncertainty.…
Correlations and Clustering in Wholesale Electricity Markets
Tianyu Cui, Francesco Caravelli, Cozmin Ududec
We study the structure of locational marginal prices in day-ahead and real-time wholesale electricity markets. In particular, we consider the case of two North American markets and…
Optimal growth trajectories with finite carrying capacity
Francesco Caravelli, Lorenzo Sindoni, Fabio Caccioli +1
We consider the problem of finding optimal strategies that maximize the average growth-rate of multiplicative stochastic processes. For a geometric Brownian motion the problem is s…
Multi-scaling of wholesale electricity prices
Francesco Caravelli, James Requeima, Cozmin Ududec +3
We empirically analyze the most volatile component of the electricity price time series from two North-American wholesale electricity markets. We show that these time series exhibi…
Higher-order interference and single-system postulates characterizing quantum theory
Howard Barnum, Markus P. Mueller, Cozmin Ududec
We present a new characterization of quantum theory in terms of simple physical principles that is different from previous ones in two important respects: first, it only refers to…
Information-theoretic equilibration: the appearance of irreversibility under complex quantum dynamics
Cozmin Ududec, Nathan Wiebe, Joseph Emerson
The question of how irreversibility can emerge as a generic phenomena when the underlying mechanical theory is reversible has been a long-standing fundamental problem for both clas…
The structure of reversible computation determines the self-duality of quantum theory
Markus P. Mueller, Cozmin Ududec
Predictions for measurement outcomes in physical theories are usually computed by combining two distinct notions: a state, describing the physical system, and an observable, descri…
Three Slit Experiments and the Structure of Quantum Theory
Cozmin Ududec, Howard Barnum, Joseph Emerson
In spite of the interference manifested in the double-slit experiment, quantum theory predicts that a measure of interference defined by Sorkin and involving various outcome probab…