Neither Layer Alone: Epistemic Integrity Requires Hierarchical Joint Design for Long-Running AI Agents

• URL: http://arxiv.org/abs/2606.04017v1
• Date: Mon, 01 Jun 2026 00:56:23 GMT
• Title: Neither Layer Alone: Epistemic Integrity Requires Hierarchical Joint Design for Long-Running AI Agents
• Abstract summary: Long-running AI agents fail when independently evolving model and harness layers change the semantics of belief, capability, and goal commitments across their boundary.<n>This paper argues that Agent Epistemic Integrity (AEI) must be treated as a first-class architectural constraint.
• Score: 0.6381783966294292
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Long-running AI agents fail not only when inference fails or tools are underspecified, but when independently evolving model and harness layers change the semantics of belief, capability, and goal commitments across their boundary - a failure class this paper terms Interface Volatility. This paper argues that Agent Epistemic Integrity (AEI) must be treated as a first-class architectural constraint, achievable only through joint model-harness design organized around an explicit interface contract. The central claim is that the model-harness interface contract is the precondition for joint design; its operational form is a four-level hierarchy - goal validity, action-archetype sequencing, tool-instance selection, and invocation-level failure discrimination - that specifies what the boundary must preserve and what structured outputs the model must return for the contract to hold across levels. This reframes long-running agent design away from flat action loops and toward contract-preserving control over persistent state. Evaluation and training should therefore derive from the contract itself, testing whether belief, tool, and goal commitments hold across session boundaries and independent layer upgrades.

Related papers

• Reality Is the Final Verifier: On Two Key Gaps in Agentic Software Engineering [78.51705689800838]
  We propose an assurance-revision loop that uses deployment evidence to revise the requirements, model, or evaluator.<n>We then cast assured agentic development as a resource-allocation problem over human judgment, agent capability, and compute.
  arXiv  Detail & Related papers  (2026-09-10T17:58:48Z)
• CrACK: Adversarial Attacks on Cross-Model Consistency in Collaborative Vision Foundation Models [16.64134816858303]
  Training-free collaborative pipelines that integrate Vision Foundation Models such as CLIP, SAM, and DINO achieve strong open-vocabulary dense prediction.<n>We propose CrACK, an inference-time attack that exploits this interface without modifying any input pixel, model weight, or training data.<n>Experiments show that CrACK causes catastrophic degradation while every individual model continues to produce its unchanged standalone output.
  arXiv  Detail & Related papers  (2026-09-07T13:47:33Z)
• Advancing Omnimodal Embodied Agents from Isolated Skills to Everyday Physical Autonomy [79.09714980186278]
  Building persistent embodied agents in unstructured environments requires unified orchestration of heterogeneous tools.<n>We present OmniAct, a framework integrating a multimodal semantic planner for skill routing across unified action spaces.<n>It achieves consistent improvements in end-to-end success across all complexity levels.
  arXiv  Detail & Related papers  (2026-06-25T16:36:35Z)
• PRISM: : Planning and Reasoning with Intent in Simulated Embodied Environments [59.07829883257003]
  Built on five multi-room apartments, PRISM structures 300 human-verified tasks into three capability tiers.<n>PRISM exposes an agent-agnostic executable action API that allows arbitrary agents to be evaluated end-to-end.
  arXiv  Detail & Related papers  (2026-05-12T04:59:47Z)
• Containment Verification: AI Safety Guarantees Independent of Alignment [13.629487815782355]
  We introduce containment verification, which locates safety guarantees in the agentic framework itself.<n>Under havoc oracle semantics, the AI is modeled as an unconstrained oracle ranging over the entire typed action space.<n>For boundary-enforceable properties, expressed over modeled boundary events, action arguments, and state, we prove a universal guarantee.
  arXiv  Detail & Related papers  (2026-05-09T16:36:45Z)
• Prediction and Empowerment: A Theory of Agency through Bridge Interfaces [0.10152838128195464]
  We study agency under partial observability in deterministic physical or simulated worlds.<n>We prove a separation between prediction, compression, and empowerment.<n>For modern AI agents, the results suggest a design principle rather than a theorem of inevitability.
  arXiv  Detail & Related papers  (2026-05-07T14:30:46Z)
• Bridging Values and Behavior: A Hierarchical Framework for Proactive Embodied Agents [26.613956143957548]
  textitValuePlanner is a hierarchical cognitive architecture that decouples high-level value scheduling from low-level action execution.<n>Our work offers a structured approach to bridging intrinsic values and grounded behavior for autonomous agents.
  arXiv  Detail & Related papers  (2026-04-30T10:43:55Z)
• ADEMA: A Knowledge-State Orchestration Architecture for Long-Horizon Knowledge Synthesis with LLMAgents [16.053669481561354]
  ADEMA is a knowledge-state orchestration architecture for long-horizon knowledge synthesis.<n>This paper presents ADEMA as a knowledge-state orchestration architecture for long-horizon knowledge synthesis.
  arXiv  Detail & Related papers  (2026-04-28T16:54:48Z)
• Structural Enforcement of Goal Integrity in AI Agents via Separation-of-Powers Architecture [0.10152838128195464]
  Policy-Execution-Authorization (PEA) architecture is a "separation-of-powers" design that enforces safety at the system level.<n>PEA decouples intent generation, authorization, and execution into independent, isolated layers connected via cryptographically constrained capability tokens.
  arXiv  Detail & Related papers  (2026-04-26T10:31:13Z)
• CaMeLs Can Use Computers Too: System-level Security for Computer Use Agents [60.98294016925157]
  AI agents are vulnerable to prompt injection attacks, where malicious content hijacks agent behavior to steal credentials or cause financial loss.<n>We introduce Single-Shot Planning for CUAs, where a trusted planner generates a complete execution graph with conditional branches before any observation of potentially malicious content.<n>Although this architectural isolation successfully prevents instruction injections, we show that additional measures are needed to prevent Branch Steering attacks.
  arXiv  Detail & Related papers  (2026-01-14T23:06:35Z)
• Towards Efficient Agents: A Co-Design of Inference Architecture and System [66.59916327634639]
  This paper presents AgentInfer, a unified framework for end-to-end agent acceleration.<n>We decompose the problem into four synergistic components: AgentCollab, AgentSched, AgentSAM, and AgentCompress.<n>Experiments on the BrowseComp-zh and DeepDiver benchmarks demonstrate that through the synergistic collaboration of these methods, AgentInfer reduces ineffective token consumption by over 50%.
  arXiv  Detail & Related papers  (2025-12-20T12:06:13Z)
• Geometrically-Constrained Agent for Spatial Reasoning [53.93718394870856]
  Vision Language Models exhibit a fundamental semantic-to-geometric gap in spatial reasoning.<n>Current paradigms fail to bridge this gap.<n>We propose a training-free agentic paradigm that resolves this gap by introducing a formal task constraint.
  arXiv  Detail & Related papers  (2025-11-27T17:50:37Z)
• DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models [51.76664843721462]
  DeepThinkVLA is a new architecture for Vision-Language-Action models.<n>It generates sequential CoT with causal attention and switches to bidirectional attention for fast decoding of action vectors.<n>It achieves a 97.0% success rate on the LIBERO benchmark.
  arXiv  Detail & Related papers  (2025-10-31T05:26:16Z)
• ERIS: An Energy-Guided Feature Disentanglement Framework for Out-of-Distribution Time Series Classification [51.07970070817353]
  An ideal time series classification (TSC) should be able to capture invariant representations.<n>Current methods are largely unguided, lacking the semantic direction required to isolate truly universal features.<n>We propose an end-to-end Energy-Regularized Information for Shift-Robustness framework to enable guided and reliable feature disentanglement.
  arXiv  Detail & Related papers  (2025-08-19T12:13:41Z)
• Towards Continual Learning Desiderata via HSIC-Bottleneck Orthogonalization and Equiangular Embedding [55.107555305760954]
  We propose a conceptually simple yet effective method that attributes forgetting to layer-wise parameter overwriting and the resulting decision boundary distortion. Our method achieves competitive accuracy performance, even with absolute superiority of zero exemplar buffer and 1.02x the base model.
  arXiv  Detail & Related papers  (2024-01-17T09:01:29Z)

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.