CONFLICTGUI: Enabling Conflict-Aware Termination for Multimodal GUI Agents

Forum topic · 小凯 · 2026-09-06

Summary

GUI agents execute natural-language instructions on user interfaces, but real users may issue infeasible instructions due to benign mistakes. This arXiv paper (2509.00009) introduces CONFLICTGUI, a benchmark covering instruction-internal conflicts and instruction-GUI context conflicts to study conflict-aware termination. Evaluation reveals severe execution-biased overcompliance: agents performing well on feasible tasks often continue blindly executing under conflicting instructions. The authors propose CONFLICTGUARD, an inference-time framework with two coupled components: a feasibility verification protocol that prompts agents to assess instruction logic and GUI-side evidence before acting, and a conditional action conditioning mechanism that shifts agents from overcompliant execution toward termination-oriented behavior. Experiments across five widely used agents show CONFLICTGUARD significantly improves average conflict-task success rates while preserving normal GUI task performance, demonstrating that lightweight inference-time intervention can substantially enhance GUI agents' ability to recognize inappropriate execution scenarios and avoid unnecessary actions.

Overview

Field: AI/ML Publication date: 2026-09-06 arXiv: 2509.00009

Graphical user interface (GUI) agents are increasingly used to execute natural-language instructions on user interfaces, yet real users may issue infeasible instructions due to benign mistakes. A reliable agent should not only know how to act, but also when not to act.

Key Contributions

CONFLICTGUI benchmark: Covers instruction-internal conflicts and instruction-GUI context conflicts to study conflict-aware termination in GUI agents.
Finding — execution-biased overcompliance: Agents that perform well on feasible tasks often continue to execute blindly under conflicting instructions.
CONFLICTGUARD framework: An inference-time method aligning an agent's feasibility awareness with its action generation, consisting of two coupled components:
A feasibility verification protocol that guides agents to assess instruction logic and GUI-side evidence before acting.
A conditional action conditioning mechanism that shifts agents from overcompliant execution toward termination-oriented behavior.

Results

Experiments on five widely used agents show that CONFLICTGUARD significantly improves average conflict-task success rates while maintaining normal GUI task performance. These results validate that lightweight inference-time intervention can substantially enhance GUI agents' ability to recognize inappropriate execution scenarios and avoid unnecessary actions.

---

*Auto-collected on 2026-09-06.*