Ask the World Before Acting: Environment Probing for Calibrated Agent World Models

• URL: http://arxiv.org/abs/2606.31422v2
• Date: Sun, 05 Jul 2026 04:18:26 GMT
• Title: Ask the World Before Acting: Environment Probing for Calibrated Agent World Models
• Abstract summary: Language agents acting over long horizons must maintain beliefs about tool states, object locations, graph edges, and subgoal dependencies.<n>We formulate environment probing as a budgeted decision problem for structured agent world models.<n>We introduce EnvProbe, a simple scoring policy that combines task criticality, staleness, verbalized uncertainty, and dependency role.
• Score: 2.571318992830046
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• World Models for Embodied Intelligence: From Plausible to Controllable to Actionable [100.13060753774657]
  World models connect perception and decision-making in embodied intelligence by maintaining hidden state, anticipating consequences, comparing interventions, and adapting when execution departs from expectations.<n>This raises a central question: which predictive capabilities improve behavior?<n>We introduce three progressively stronger capability levels: Plausible models preserve task-relevant temporal, geometric, or physical structure; Controllable models additionally predict how interventions alter that structure; and Actionable models translate predictions into measurable gains in planning, action, learning, evaluation, verification, recovery, or data selection.
  arXiv  Detail & Related papers  (2026-09-15T06:22:42Z)
• Uncertainty Quantification for LLM Agents: A Taxonomy, an Evaluation Protocol, and an Empirical Study [5.818268354990671]
  Large language models act as agents that plan, call tools, retrieve evidence, maintain memory, and interact over long horizons.<n>Existing work on quantifying uncertainty for LLMs was built almost entirely for single-turn question answering.<n>This paper argues that errors and uncertainty arise from multi-turn conversations, environments, and tools rather than from a single-turn question answering setting.
  arXiv  Detail & Related papers  (2026-09-07T12:06:49Z)
• The Intervention Gap in Latent World Models [0.0]
  Planning-time intervention fidelity is a distinct, measurable property of a learned world model.<n>Across released TD-MPC2 checkpoint sizes, episode return falls as an operator-error diagnostic on task observables grows.<n>A self-supervised world model trained without task signal preserves the same operator substantially better than a task-anchored model on the shared task.
  arXiv  Detail & Related papers  (2026-08-30T19:47:51Z)
• ScrambleToolBench: Agents Search Exhaustively Even When Their Own Map Points to the Next Step [74.83535434786145]
  We introduce ScrambleToolBench, an interactive terminal benchmark designed to isolate behavioral reasoning.<n>The benchmark requires agents to uncover hidden tool behaviors entirely through trial-and-error interaction.<n>Our evaluation of state-of-the-art language models reveals that successful initial discovery does not translate into robust adaptation.
  arXiv  Detail & Related papers  (2026-08-03T15:07:46Z)
• Foresight: Failure Detection for Long-Horizon Robotic Manipulation with Action-Conditioned World Model Latents [58.49879338545782]
  Long-horizon tasks are common in real-world robotic deployments, yet failure detection for such tasks remains underexplored.<n>We present Foresight, a failure detection framework that monitors manipulation trajectories using latent representations from an action-conditioned world model.<n>Our results suggest that action-conditioned world-model embeddings provide a scalable representation for reliable failure monitoring in long-horizon manipulation.
  arXiv  Detail & Related papers  (2026-06-22T09:32:28Z)
• CIAware-Bench: Benchmarking Control Intervention Awareness Across Frontier LLMs [100.38986535324284]
  We introduce textbfCIAware-Bench, a benchmark for measuring textbfcontrol textbfintervention (CI) awareness across frontier models.<n>CIAware-Bench tests whether models can distinguish their own trajectories from those modified by a control intervention.
  arXiv  Detail & Related papers  (2026-06-09T16:24:16Z)
• PatchWorld: Gradient-Free Optimization of Executable World Models [69.05094454437402]
  We introduce PatchWorld, a gradient-free framework that turns offline trajectories into executable Python world models.<n>Across seven AgentGym environments, PatchWorld-Simple achieves the highest code-based planning score.<n>We find that a human-specified residual-memory bias improves surface observation fidelity but weakens decision utility.
  arXiv  Detail & Related papers  (2026-05-29T06:11:14Z)
• ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents [59.626170560327274]
  We present textbfClawForge, a generator-backed benchmark framework for executable command-line categories under state conflict.<n>We instantiate this framework as the ClawForge-Bench (17 scenarios, 6 ability categories)
  arXiv  Detail & Related papers  (2026-05-13T21:34:08Z)
• Agent-BRACE: Decoupling Beliefs from Actions in Long-Horizon Tasks via Verbalized State Uncertainty [70.43119366710778]
  We introduce Agent-BRACE: Agent Belief state Representation via Abstraction and Confidence Estimation.<n>It is a method that decouples an LLM agent into a belief state model and a policy model, jointly optimized via reinforcement learning.<n>It achieves an average absolute improvement of +14.5% across long-horizon, partially observable language environments.
  arXiv  Detail & Related papers  (2026-05-12T02:37:04Z)
• When Agents Overtrust Environmental Evidence: An Extensible Agentic Framework for Benchmarking Evidence-Grounding Defects in LLM Agents [7.383366380709273]
  Large language model agents increasingly operate through environment-facing scaffolds that expose files, web pages, APIs, and logs.<n>Environmental grounding is a systems-level problem involving context admission, evidence provenance, freshness checking, verification policy, action gating, and model reasoning.<n>We introduce EnvTrustBench, an agentic framework for benchmarking this failure mode.
  arXiv  Detail & Related papers  (2026-05-09T09:32:27Z)
• The Geometric Canary: Predicting Steerability and Detecting Drift via Representational Stability [0.0]
  We show that geometric stability, the consistency of a representation's pairwise distance structure, addresses two functions.<n> unsupervised stability fails entirely for steering on real-world tasks.<n>It excels at drift detection, measuring nearly $2times$ greater geometric change than CKA during post-training alignment.
  arXiv  Detail & Related papers  (2026-04-20T01:24:45Z)
• ClawArena: Benchmarking AI Agents in Evolving Information Environments [61.664633997138004]
  ClawArena is a benchmark for evaluating AI agents in evolving information environments.<n>Each scenario maintains a complete hidden ground truth while exposing the agent only to noisy, partial, and sometimes contradictory traces.<n> evaluation is organized around three coupled challenges: multi-source conflict reasoning, dynamic belief revision, and implicit personalization.
  arXiv  Detail & Related papers  (2026-04-05T17:55:23Z)
• Current Agents Fail to Leverage World Model as Tool for Foresight [61.82522354207919]
  Generative world models offer a promising remedy: agents could use them to foresee outcomes before acting.<n>This paper empirically examines whether current agents can leverage such world models as tools to enhance their cognition.
  arXiv  Detail & Related papers  (2026-01-07T13:15:23Z)
• Scene Graph-Guided Proactive Replanning for Failure-Resilient Embodied Agent [9.370683025542686]
  We present a proactive replanning framework that detects and corrects failures at subtask boundaries.<n>Experiments in the AI2-THOR simulator demonstrate that our approach detects semantic and spatial mismatches before execution failures occur.
  arXiv  Detail & Related papers  (2025-08-15T07:48:51Z)
• Hell or High Water: Evaluating Agentic Recovery from External Failures [19.402114652658206]
  We develop a specialized agentic planning benchmark to study how well language agents search for alternative ways to achieve their goals.<n>Our benchmark confronts the agent with external failures in its workflow, such as functions that suddenly become unavailable.<n>Overall, we find that language agents struggle to formulate and execute backup plans in response to environment feedback.
  arXiv  Detail & Related papers  (2025-08-14T19:21:09Z)
• AI in a vat: Fundamental limits of efficient world modelling for agent sandboxing and interpretability [84.52205243353761]
  Recent work proposes using world models to generate controlled virtual environments in which AI agents can be tested before deployment.<n>We investigate ways of simplifying world models that remain agnostic to the AI agent under evaluation.
  arXiv  Detail & Related papers  (2025-04-06T20:35:44Z)
• AgentRefine: Enhancing Agent Generalization through Refinement Tuning [28.24897427451803]
  Large Language Model (LLM) based agents have proved their ability to perform complex tasks like humans.<n>There is still a large gap between open-sourced LLMs and commercial models like the GPT series.<n>In this paper, we focus on improving the agent generalization capabilities of LLMs via instruction tuning.
  arXiv  Detail & Related papers  (2025-01-03T08:55:19Z)
• Criticality and Safety Margins for Reinforcement Learning [53.10194953873209]
  We seek to define a criticality framework with both a quantifiable ground truth and a clear significance to users.<n>We introduce true criticality as the expected drop in reward when an agent deviates from its policy for n consecutive random actions.<n>We also introduce the concept of proxy criticality, a low-overhead metric that has a statistically monotonic relationship to true criticality.
  arXiv  Detail & Related papers  (2024-09-26T21:00:45Z)
• Model Checking for Closed-Loop Robot Reactive Planning [0.0]
  We show how model checking can be used to create multistep plans for a differential drive wheeled robot so that it can avoid immediate danger. Using a small, purpose built model checking algorithm in situ we generate plans in real-time in a way that reflects the egocentric reactive response of simple biological agents.
  arXiv  Detail & Related papers  (2023-11-16T11:02:29Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.