Ask the World Before Acting: Environment Probing for Calibrated Agent World Models
Language agents acting over long horizons must maintain beliefs about tool states, object locations, graph edges, and subgoal dependencies. When these beliefs drift, failures can be fixed neither by longer reasoning traces nor by ordinary self-reflection, since the missing evidence lies in the environment. We formulate environment probing as a budgeted decision problem for structured agent world models: before acting, the agent may query the current value of one belief field, update its table, and pay one interaction step. We introduce EnvProbe, a simple scoring policy that combines task criticality, staleness, verbalized uncertainty, and dependency role. A type-stratified analysis separates the benefit of belief repair from the cost of displaced task actions and predicts different behavior for procedural and spatial beliefs. In three controlled environments with gold belief states, EnvProbe improves terminal world-state accuracy over periodic probing by 11.76 percentage points on procedural tool-dependency tasks, 3.79 points on spatial tasks, and 6.45 points overall. Ablations show that task-structural terms are the main source of the gains, while self-reported uncertainty is unreliable under confident wrong beliefs. The results suggest that agent calibration should be treated as an action-selection problem over environment evidence, not only as a model-internal reasoning problem.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Black-Box Membership Inference Attack for LVLMs via Prior Knowledge-Calibrated Memory Probing
Large vision-language models (LVLMs) derive their capabilities from extensive training on vast corpora of visual and textual data. Empowered by large-scale parameters, these models often exhibit strong memorization of th…
Machine learning applied to quantum synchronization-assisted probing
A probing scheme is considered with an accessible and controllable qubit, used to probe an out-of equilibrium system consisting of a second qubit interacting with an environment. Quantum spontaneous synchronization betwe…
BIG-bench Machine LearningBeyond Next-Observation Prediction: Agent-Authored World Modeling for Sequential Decision Making
Recent studies on world modeling for Large Language Model (LLM) agents typically formulate the learning objective as next-observation prediction. However, this objective ties supervision to what a transition happens to r…
Decision MakingTACT: Mitigating Overthinking and Overacting in Coding Agents via Activation Steering
When language model agents tackle complex software engineering tasks, they often degrade over long trajectories, which we define as *agent drift*. We focus on two recurring failure modes *overthinking* and *overacting*, …
Reasoning Models Know When They're Right: Probing Hidden States for Self-Verification
Reasoning models have achieved remarkable performance on tasks like math and logical reasoning thanks to their ability to search during reasoning. However, they still suffer from overthinking, often performing unnecessar…
Logical ReasoningMath