arXiv:2603.06870v2 Announce Type: replace
Abstract: Long-horizon execution in Large Language Models (LLMs) remains unstable even when high-level strategies are provided. Evaluating on controlled algorithmic puzzles, we demonstrate that while decomposition is essential for stability, extreme decomposition creates a “no-recovery bottleneck”. We show that this bottleneck becomes critical due to highly non-uniform error distribution, where consistent errors on a few “hard” steps become irreversible.
To address this, we propose Lookahead-Enhanced Atomic Decomposition (LEAD). By incorporating short-horizon future validation and aggregating overlapping rollouts, LEAD provides enough isolation to maintain stability while retaining enough local context to correct errors. This enables the o4-mini model to solve Checkers Jumping up to complexity $n=13$, whereas extreme decomposition fails beyond $n=11$.
Behavior change beyond intervention: an activity-theoretical perspective on human-centered design of personal health technology
IntroductionModern personal technologies, such as smartphone apps with artificial intelligence (AI) capabilities, have a significant potential for helping people make necessary changes in their behavior