Revision-Aware Independent Agent Graphs for Dynamic Reasoning

2026-10-01

AI Summary

This study addresses the challenges of version selection and reasoning failures caused by event stream revisions in dynamic task routing. To this end, it proposes the first Revision-aware Independent Agent Graph (RIAG) architecture. By decoupling temporal parsing from logical reasoning and integrating an immutable document identity cache, deterministic temporal resolution, and an audit-based repair mechanism, RIAG enables efficient dynamic question answering. Experimental results demonstrate that RIAG achieves a joint accuracy of 54.24% while requiring only 0.62 model invocations per query. This minimal computational overhead allows the proposed method to significantly outperform existing approaches, effectively balancing timeliness and accuracy.

Abstract

Conventional reasoning protocols present a fixed, preselected task, so they cannot test whether an agent propagates relevant updates, preserves unaffected work, or reconstructs a historical task binding. We therefore study \emph{dynamic task routing}, in which an event stream revises task bindings and a system must select the document version valid at each query time before solving it. To study this problem, we repurpose six widely used benchmarks: MMLU, MMLU-Pro, MedMCQA, MATH, GPQA, and HumanEval into 31{,}119 dynamic episodes comprising 373{,}428 temporally categorized queries. This setting exposes a central trade-off: recomputing after every event wastes work, whereas unguarded reuse returns stale conclusions. We introduce the Revision-Aware Independent Agent Graph (RIAG), a bounded multi-agent policy that separates deterministic temporal resolution from task reasoning. RIAG caches solutions by immutable document identity, starts each fresh task with two unexposed attempts, and conditionally invokes audit and repair, using at most four calls per document version. On this collection, homogeneous RIAG achieves 54.24\% joint routing-and-answer accuracy at 0.62 calls/query, compared with 32.22\% at 18.00 calls/query for the strongest comparison method; heterogeneous RIAG reaches 49.78\% at 0.63 calls/query.

Problem

• dynamic task routing
• temporal reasoning
• revision-aware reasoning
• multi-agent systems
• event stream

Innovation

• Dynamic Task Routing
• Revision-Aware Independent Agent Graph
• Multi-Agent Policy
• Temporal Resolution
• Bounded Reasoning