analiza aiqbench
This survey addresses the critical challenge of verifying autonomous LLM agents by formalizing execution provenance and evidence tracing as foundational elements for accountability. By mapping agent behaviors—including tool use, memory access, and multi-agent collaboration—into a unified typed graph framework, it establishes necessary standards for auditing complex AI systems beyond simple output accuracy. The work provides an essential taxonomy and methodological directions for building trustworthy, auditable, and recoverable next-generation agents.
From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents
arXiv:2606.04990v2 Announce Type: replace-cross Abstract: Large language model (LLM)-based agents are evolving from passive text generators into autonomous systems capable of planning, tool use, retrieval, memory access, environmental interaction, and multi-agent collaboration. These capabilities expand agent autonomy, but also make agent behavior harder to verify, debug, and audit. Final-answer accuracy alone cannot explain how an output was produced, which evidence supported each claim, whether tool calls were justified, how memory influenced later decisions, or where failures originated. This survey examines evidence tracing and execution provenance as foundations for process-level accountability in trustworthy LLM agents. We define execution provenance as the typed graph of an agent execution and evidence tracing as its projection onto evidence-support relations. This perspective connects retrieval grounding, claim support, tool-use safety, memory lineage, observability, debugging, audit, and recovery within a unified framework. We introduce a taxonomy covering trace sources, evidence and execution units, provenance relations, tracing granularity and timing, representation forms, and trust functions. We then review key methodological directions, including provenance representation, evidence attribution, tool-use provenance, runtime guardrails, provenance-bearing memory, observability, and failure diagnosis. Finally, we discuss benchmarks, datasets, metrics, and open challenges for building provenance-aware, auditable, and recoverable agent systems.
Abstract:Large language model (LLM)-based agents are evolving from passive text generators into autonomous systems capable of planning, tool use, retrieval, memory access, environmental interaction, and multi-agent collaboration. These capabilities expand agent autonomy, but also make agent behavior harder to verify, debug, and audit. Final-answer accuracy alone cannot explain how an output was produced, which evidence supported each claim, whether tool calls were justified, how memory influenced later decisions, or where failures originated. This survey examines evidence tracing and execution provenance as foundations for process-level accountability in trustworthy LLM agents. We define execution provenance as the typed graph of an agent execution and evidence tracing as its projection onto evidence-support relations. This perspective connects retrieval grounding, claim support, tool-use safety, memory lineage, observability, debugging, audit, and recovery within a unified framework. We introduce a taxonomy covering trace sources, evidence and execution units, provenance relations, tracing granularity and timing, representation forms, and trust functions. We then review key methodological directions, including provenance representation, evidence attribution, tool-use provenance, runtime guardrails, provenance-bearing memory, observability, and failure diagnosis. Finally, we discuss benchmarks, datasets, metrics, and open challenges for building provenance-aware, auditable, and recoverable agent systems.
Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
arXiv:2606.04990 [cs.CR]
(or arXiv:2606.04990v2 [cs.CR] for this version)
https://doi.org/10.48550/arXiv.2606.04990 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 286 cuvinte
Subiecte conexe
[AINews] not much happened today
Latent Space · 18 sept. 2026
From Agent Traces to Trust: Evidence Tracing and Execution Provenance in LLM Agents
arXiv · 4 iun. 2026
From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents
arXiv · 12 sept. 2026
From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents
arXiv · 17 iun. 2026