4 citations · 5 across the 6 of their papers we have counts for
6 papers
LLMs and the Madness of Crowds
William F. Bradley
We investigate the patterns of incorrect answers produced by large language models (LLMs) during evaluation. These errors exhibit highly non-intuitive behaviors unique to each mode…
Enhancing LLM Evaluations: The Garbling Trick
William F. Bradley
As large language models (LLMs) become increasingly powerful, traditional evaluation metrics tend to saturate, making it challenging to distinguish between models. We propose a gen…
Running in Circles: Packet Routing on Ring Networks
William F. Bradley
I analyze packet routing on unidirectional ring networks, with an eye towards establishing bounds on the expected length of the queues. Suppose we route packets by a greedy "hot po…
Superconcentration on a Pair of Butterflies
William F. Bradley
Suppose we concatenate two directed graphs, each isomorphic to a dimensional butterfly (but not necessarily identical to each other). Select any set of input and ou…
Sparse Eigenvectors of the Discrete Fourier Transform
William F. Bradley
We construct a basis of sparse eigenvectors for the N-dimensional discrete Fourier transform. The sparsity differs from the optimal by at most a factor of four. When N is a perfect…
A Recursive Definition of the Holographic Standard Signature
William F. Bradley
We provide a recursive description of the signatures realizable on the standard basis by a holographic algorithm. The description allows us to prove tight bounds on the size of pla…