Get Started
Research questionHow can agent decisions be reconstructed for auditing and controlled replay when tool state and authorization context are missing?An agent’s final output does not reveal the evidence, tool state, authorization, or action path that produced a committed decision. Missing or unobserved execution state makes it difficult to distinguish a faithfully replayed decision from an unexplained divergence.
AI Agents
Alignment & Safety
Evaluation & Benchmarks
Latest papersRecent research connected to this question, newest first.DNative-Twin: Decision Graphs and Digital Twins for Reconstructable Agentic DecisionsThe source examines enterprise decision processes using a graph-native digital twin that records typed decision trajectories and re-executes decision mechanisms. Evidence comes from three public process logs and controlled replay suites; the reported experiments find that graph structure localizes represented changes but cannot determine consequences of unobserved tool state. Adding replay-contract state and verification results improved unresolved-divergence recall, while the held-out set contained no critical-class instance. Reported median end-to-end time increased from 0.794 to 8.889 seconds across the tested case volumes.research paper · Sep 3, 2026
Related questions
How can we measure whether long-horizon tool-using agents query hidden state and execute stated plans?How can LLM agents reuse execution traces without losing temporal and outcome-dependent behavior?How can AI agents adapt execution routes as runtime evidence invalidates their planned continuation?How can tool-using agents prioritize and compress action-relevant context instead of relying on semantic similarity?
Home
Topics
Search
Library