From Natural Language Policies to Executable Obligations: A Verification Harness for Dependable In-Car LLM Agents

• URL: http://arxiv.org/abs/2608.23282v1
• Date: Mon, 24 Aug 2026 14:06:21 GMT
• Title: From Natural Language Policies to Executable Obligations: A Verification Harness for Dependable In-Car LLM Agents
• Abstract summary: We present AgentGuardUtil, our entry to CAR-bench Track1, which treats the AI planer (LLM) as a fallible proposer inside a grounded verify-and-revise loop.<n>The natural-language policy shipped with each conversation is compiled, once per policy, into typed machine-checkable rules.<n>A deterministic obligation engine interprets these rules against live tool results and the simulated post-write state of the draft itself.
• Score: 0.5352699766206808
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) agents deployed in vehicles must satisfy a written operating policy on every turn: a single hallucinated identifier, omitted mandatory side-effect, or premature completion claim fails the task. We present AgentGuardUtil, our entry to CAR-bench Track~1, which treats the AI planer (LLM) as a fallible proposer inside a grounded verify-and-revise loop. Its core novelty is a runtime policy compiler: the natural-language policy shipped with each conversation is compiled, once per policy, into typed machine-checkable rules, a subset of which receive an executable form. A deterministic obligation engine interprets these rules against live tool results and the simulated post-write state of the draft itself, emitting the exact remedial calls with computed arguments rather than natural-language reminders. Around this engine, 25 deterministic gates (identifier provenance, schema and enum validity, gather-before-act, confirmation and future-time protocols) and an LLM critic produce tiered findings that drive a bounded revision loop tuned for the pass k metric.

Related papers

• CEDAR: Automata as Verifiable Interfaces for Language-Guided Embodied Action [10.197044775138883]
  We present CEDAR, a framework that grounds instructions as regular languages over environment event traces.<n>In Minecraft, with the same simulator/API observations available to a program-generating baseline, CEDAR maintains temporal and spatial constraints.<n>These results suggest that regular languages offer a practical verification layer between natural-language instructions and embodied-agent policies.
  arXiv  Detail & Related papers  (2026-08-28T00:25:48Z)
• In-Context VLA: Endowing Vision-Language-Action Models with Language via In-Context Post-Training and Agentic Tool Use [18.507118969955666]
  Vision-Language-Action models imitate expert action chunks conditioned on a static image and a fixed instruction.<n>We introduce textbfourmethod, a framework that endows a VLA with language competence through in-context post-training.<n>Our method consistently achieves SOTA results in both performance and efficiency when compared with CoT-based approaches.
  arXiv  Detail & Related papers  (2026-08-06T08:21:02Z)
• ETAS: An Effect-Typed Language for Agent Systems [3.1440390119576676]
  ETAS is a programming language for agent systems.<n>It treats model-backed agents, tool calls, prompts, typed memory, human approvals, policies, and execution traces as semantic program elements.
  arXiv  Detail & Related papers  (2026-07-20T10:11:53Z)
• PhyAgentOS: A Self-Evolving Operating System for Embodied Agents with Decoupled Cognitive Planning and Physical Execution [49.776611937968]
  We present PhyAgentOS, a foundation delivering scheduling, verification, memory, benchmarking, and safety as system-level services.<n>Its Session-Centered treats a session, not an action, as the minimum unit of scheduling, compatibility, supervised execution, evidence collection, and acceptance.<n>SessionVerifier distinguishes execution termination from semantic task completion via evidence-grounded verdicts of success, failure, or replan.<n> Benchmarking reuses the deployment session and verification path, so results trace to real execution.
  arXiv  Detail & Related papers  (2026-07-18T04:46:53Z)
• Weave of Formal Thought [51.56484100374058]
  We introduce Weave of Formal Thought (WoFT), a paradigm uniting rigorous syntactic validation with learned structural representations.<n>We present a latent-variable fine-tuning method training the language model to interleave non-terminal grammar symbols directly into generation.<n>For Python, fine-tuning StarCoder2-3B with our RWS objective reduces per-token cross-entropy by 14.3% relative to a text-only SFT baseline.
  arXiv  Detail & Related papers  (2026-06-24T15:58:11Z)
• VASO: Formally Verifiable Self-Evolving Skills for Physical AI Agents [57.240036084348354]
  We introduce VASO, a framework for verification-guided self-evolution of robot skill contracts.<n>VASO verifies logically inconsistent skill contracts, then verifies plans induced by the skill against global and local temporal specifications.<n>On Clearpath Jackal and PX4 quadcopter tasks, VASO reaches 97.2% formal-specification compliance using fewer than 100 optimization samples.
  arXiv  Detail & Related papers  (2026-06-03T20:02:35Z)
• RunAgent: Interpreting Natural-Language Plans with Constraint-Guided Execution [36.644786364066796]
  RunAgent is a multi-agent plan execution platform that interprets natural-language plans and enforces stepwise execution through constraints and rubrics.<n>We show that RunAgent outperforms baseline LLMs and state-of-the-art PlanGEN methods.
  arXiv  Detail & Related papers  (2026-05-01T17:29:45Z)
• A Physical Agentic Loop for Language-Guided Grasping with Execution-State Monitoring [5.437966695589128]
  We reformulate language-guided grasping as a bounded embodied agent operating over grounded execution states.<n>We introduce a physical agentic loop that wraps an unmodified learned manipulation primitive.<n>We validate the resulting loop on a mobile manipulator with an eye-in-hand D405 camera.
  arXiv  Detail & Related papers  (2026-04-08T08:01:35Z)
• Policy-as-Prompt: Turning AI Governance Rules into Guardrails for AI Agents [0.19336815376402716]
  We introduce a regulatory machine learning framework that converts unstructured design artifacts (like PRDs, TDDs, and code) into verifiable runtime guardrails.<n>Our Policy as Prompt method reads these documents and risk controls to build a source-linked policy tree.<n>System is built to enforce least privilege and data minimization.
  arXiv  Detail & Related papers  (2025-09-28T17:36:52Z)
• DECIDER: A Dual-System Rule-Controllable Decoding Framework for Language Generation [57.07295906718989]
  Constrained decoding approaches aim to control the meaning or style of text generated by pre-trained large language (Ms also PLMs) for various tasks at inference time.<n>These methods often guide plausible continuations by greedily and explicitly selecting targets.<n>Inspired by cognitive dual-process theory, we propose a novel decoding framework DECIDER.
  arXiv  Detail & Related papers  (2024-03-04T11:49:08Z)
• Generative Context-aware Fine-tuning of Self-supervised Speech Models [54.389711404209415]
  We study the use of generative large language models (LLM) generated context information. We propose an approach to distill the generated information during fine-tuning of self-supervised speech models. We evaluate the proposed approach using the SLUE and Libri-light benchmarks for several downstream tasks: automatic speech recognition, named entity recognition, and sentiment analysis.
  arXiv  Detail & Related papers  (2023-12-15T15:46:02Z)
• Large Language Models as General Pattern Machines [64.75501424160748]
  We show that pre-trained large language models (LLMs) are capable of autoregressively completing complex token sequences. Surprisingly, pattern completion proficiency can be partially retained even when the sequences are expressed using tokens randomly sampled from the vocabulary. In this work, we investigate how these zero-shot capabilities may be applied to problems in robotics.
  arXiv  Detail & Related papers  (2023-07-10T17:32:13Z)
• CAPE: Corrective Actions from Precondition Errors using Large Language Models [8.547766794082184]
  We propose a novel approach that attempts to propose corrective actions to resolve precondition errors during planning. CAPE improves the quality of generated plans by leveraging few-shot reasoning from action preconditions. Our improvements transfer to a Boston Dynamics Spot robot with a set of skills (specified in language) and associated preconditions, where CAPE improves the correctness metric of the executed task plans by 76.49% compared to SayCan.
  arXiv  Detail & Related papers  (2022-11-17T23:14:51Z)

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.