# Step Gui Calibrated Rewards

> Train GUI agents through calibrated step-level reasoning anchored to trajectory-level validation. Use trajectory verification rather than step annotation achieving 90% accuracy with 10-100× cost reduction. Implement self-evolving training framework with generation-refinement cycles and verifiable reward signals.

- Skill: `adu2021/step-gui-calibrated-rewards` (Agent Skill)
- Install (CLI): `npx skillmds@latest add adu2021/step-gui-calibrated-rewards`
- Raw SKILL.md: https://api.skillmd.com/api/skills/adu2021/step-gui-calibrated-rewards/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- License: MIT
- Author: adu2021 (https://skillmd.com/u/adu2021)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/adu2021/step-gui-calibrated-rewards

---


## Skill Summary

Step-GUI presents three interconnected innovations for GUI automation. The core contribution is a Calibrated Step Reward System (CSRS) that anchors LLM-generated dense step-level reasoning to trajectory-level evaluation signals through validation rather than step-level annotation, achieving 90% accuracy with 10-100× cost reduction. A self-evolving training framework enables continuous improvement through generation-refinement cycles. Supporting infrastructure includes GUI-MCP (hierarchical protocol for LLM-device interaction) and AndroidDaily benchmark for real-world evaluation.

## When To Use

- Training autonomous GUI agents for mobile or web automation
- Projects where annotation cost is prohibitive for step-level supervision
- Scenarios requiring self-improving agent training loops
- Research on efficient reward calibration for RL

## When NOT To Use

- Scenarios with abundant step-level annotations making calibration unnecessary
- Real-time applications where training overhead isn't justified
- Domains without reliable trajectory-level evaluation signals (success/failure)
- Simple scripted automation already meeting requirements

## Core Technique

Three key innovations enable efficient agent training:

**1. Calibrated Step Reward System (CSRS)**
Convert expensive annotation problem into scalable data-refinement process. Rather than annotating each step, use trajectory-level validation (automated verifiers or human judges confirming task success/failure), then apply thinking models to generate rich chain-of-thought reasoning for successful trajectories. Achieves ">90% annotation accuracy with 10-100× cost reduction."

**2. Self-Evolving Training Framework**
Implement three-stage progressive training paradigm:
- Mid-Training: Generate rollouts and refine via rejection sampling
- Cold-Start Fine-Tuning: Initialize from successful trajectories
- Reinforcement Learning: Optimize with verifiable reward signals

Orchestrate two parallel data flows—generation (from model rollouts) and refinement (via rejection sampling)—enabling continuous performance enhancement.

**3. Supporting Infrastructure**
- GUI-MCP: Hierarchical protocol standardizing LLM-device interaction with privacy protection through local specialist model execution
- AndroidDaily: Evaluation benchmark grounded in real-world usage patterns with 3,146 static actions and 235 end-to-end tasks

## Implementation Notes

Establish trajectory-level evaluation: automated verifier or human judgment confirming success/failure. Apply thinking model to generate step-level reasoning chains for successful trajectories. Implement three-stage training: mid-training with rejection sampling, cold-start fine-tuning, and RL with verifiable rewards. Use hierarchical protocol for LLM-device interaction.

## References

- Original paper: Step-GUI (Dec 2025)
- Reward calibration for reinforcement learning
- GUI agent training methodologies