| title | Thinking Fast, Anywhere: One System-1 for the Decision Points of Any Agent Workflow |
|---|---|
| authors | Rui Li, Liyang He, Zheng Zhang, Zhenya Huang, Qi Liu |
| date | 2026-11-18 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2611.23751 |
| affiliation | University of Science and Technology of China |
| upvotes | 14 (Hugging Face Daily Papers, 2026-11-21) |
| links | abs pdf hf code |
| cite | |
| related | ExeWorldBench: Benchmarking Code-Based Executable 3D World Modeling ACC: Compiling Agent Trajectories for Long-Context Training GraphForge: Training Working Agents with Graph-Anchored Workspace Synthesis Speculative Sampling in Interactive Video Generation PhysVista: Benchmarking Physical Intelligence in VLMs via a Perception-Reasoning-Assessment Loop |
| abstract | Autonomous LLM agents incur substantial cost and latency by invoking a large model (System-2) at every step, even when many intermediate actions choose among a small set of known candidates. Existing dual-system agents offload such decisions to a lightweight System-1 model, but rely on workflow-specific hooks authored per environment. We present AnyS1, a framework that turns human-authored procedural skills, whose instructions naturally specify where choices arise and which options apply, into a shared, domain-general source of decision points. Offline, AnyS1 mines grounded enumerable decisions across over 1,000 public skills, gathers execution-grounded trajectory states inside isolated containers to construct Any-DP, and trains a single calibrated 4B pointer System-1 with label smoothing. At run time, compiled workflows resolve routine decisions in one forward pass and fall back to System-2 when confidence falls below a calibrated threshold. Across three unseen benchmarks and distinct System-2 model families, AnyS1 preserves task reliability while cutting episode cost by 75-91% and model-call time by 63-92% relative to an uncompiled System-2 agent, and maintains performance under zero-shot threshold transfer. |
bibtex
@misc{li2026thinking,
title={Thinking Fast, Anywhere: One System-1 for the Decision Points of Any Agent Workflow},
author={Rui Li and Liyang He and Zheng Zhang and Zhenya Huang and Qi Liu},
year={2026},
eprint={2611.23751},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2611.23751}
}