When Agents Slow Down: Understanding LLM Agents'Test-Time Strategies via Elo-per-token Analysis
This work defines the scaling inflection point as the per-session budget where marginal Elo gains match the independent-sampling reference, and proposes Elo-per-token analysis, which tracks the best solution found at each token budget and uses a Bradley-Terry model to aggregate within-task orderings into Elo ratings ac...
Kai-Yuan Liu, Qiu-Yang Mang, Bo-Fei Peng et al.
· 2 citations