Global Bradley-Terry rankings of LLMs are misleading due to structured heterogeneity in user preferences, and small (λ, ν)-portfolios recover coherent subpopulations that cover over 96% of votes with just five rankings.
Warm-Started QAOA with Cus- tom Mixers Provably Converges and Computationally Beats Goemans-Williamson’s Max-Cut at Low Cir- cuit Depths
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
representative citing papers
Numerical experiments on QAOA show optimal parameters often break expected patterns, performance becomes less parameter-sensitive with depth, and component-wise iterative fixing performs competitively or better at low depth.
citing papers explorer
-
Why Global LLM Leaderboards Are Misleading: Small Portfolios for Heterogeneous Supervised ML
Global Bradley-Terry rankings of LLMs are misleading due to structured heterogeneity in user preferences, and small (λ, ν)-portfolios recover coherent subpopulations that cover over 96% of votes with just five rankings.
-
Going off Pattern? QAOA Parameter Heuristics and Potentials of Parsimony
Numerical experiments on QAOA show optimal parameters often break expected patterns, performance becomes less parameter-sensitive with depth, and component-wise iterative fixing performs competitively or better at low depth.