Firehose

14 SEP 2026 · Paper

When Agents Slow Down: Understanding LLM Agents' Test-Time Strategies via Elo-per-token Analysis

This paper investigates how large language model (LLM) agents adapt their performance during long tasks, and how their test-time strategies impact their scalability. Practitioners might care because understanding these strategies can help improve the performance of LLM agents in real-world applications.

large language models continual learning test-time strategies Elo ratings token budget open-ended tasks

Browse by tag

When Agents Slow Down: Understanding LLM Agents' Test-Time Strategies via Elo-per-token Analysis