GATS framework achieves 100% success rate on agent planning tasks with zero LLM calls during inference
Graph-Augmented Tree Search combines UCB1-based tree search with a layered world model to outperform LATS and ReAct on multi-step planning benchmarks.
1 source · cross-referenced
- GATS eliminates LLM calls during planning by using a three-layer world model and UCB1-based tree search.
- Achieves 100% success rate on synthetic planning tasks and a 12-scenario stress test, outperforming LATS and ReAct.
- Requires zero LLM calls per task during planning, compared to 37 for LATS, and produces deterministic plans with zero variance.
A new planning framework called GATS (Graph-Augmented Tree Search) introduces a three-layer world model to reduce reliance on LLM inference during multi-step agent planning tasks. The framework combines systematic UCB1-based tree search with exact symbolic action matching (L1), statistics learned from execution logs (L2), and LLM-based prediction for unknown actions (L3).
In synthetic planning tasks with branching paths and dead-ends, GATS achieved a 100% success rate, outperforming LATS (92%) and ReAct (64%). On a 12-scenario stress test spanning coding workflows, web navigation, and long-horizon tasks, GATS maintained a 100% success rate, while LATS dropped to 88.9% and ReAct to 23.9%.
GATS requires zero LLM calls per task during planning, compared to 37 per task for LATS, and produces deterministic plans with zero variance across runs. The authors argue this approach substantially outperforms LLM-guided exploration for agent planning by reducing computational costs and eliminating stochastic behavior.
- Nov 10, 2026 · Latent Space — swyx
Handshake AI details specialized training data pipelines for finance, bio, and alignment agents
Trust78 - Nov 9, 2026 · Latent Space — swyx
Stanford course and Monaco engineering team adapt to coding agents in software workflows
Trust79 - Nov 8, 2026 · The Verge — AI
Apple accelerates Siri roadmap to add third-party app control and agentic features
Trust74