Set-shifting Behavioral Test for Harnessed Agents

arXiv:2607.13396v1 Announce Type: new Abstract: What happens to an LLM agent's tool choice when the reliable tool silently changes within an ongoing session? We borrow set-shifting from cognitive psychology to study how well agents adapt to hidden reliability shifts. Our benchmark mounts tool-skill libraries with redundancies, where many tools solve the same task but differ in hidden reliability. In our evaluation framework, a branched schedule shifts the reliable tool group at hidden boundaries and pairs every shift with a no-shift control. We find that agents, by default, settle on a small recurring routine within a few turns of each boundary, with call shares concentrating on a few discrete values after each reliability shift. We score the set-shifting accuracy for each agent trajectory: the joint probability of routing to the target tool group in every post-shift window. We test open-weight LLMs in an open-source agentic harness and find qualitatively distinct failure modes across the same set of routines. We also find that set framing, how the toolset presents the alternatives as competing or complementary, shifts the routing dynamics.

Related

• The Surface You Test Is Not the Surface That Breaks
• When Retrieval Metrics Mislead: Measuring Policy Signal in Long-Horizon Tool-Use Agents
• RS-Claw: Progressive Active Tool Exploration via Hierarchical Skill Trees for Remote Sensing Agents
• FALAT: Tracing Failures in LLM Agent Trajectories via Dependency-Guided Search

Source: arXiv cs.AI | 2026-07-16