Revisable by Design: A Theory of Streaming LLM Agent Execution

TL;DR

This paper introduces a streaming paradigm for LLM agents that allows concurrent execution and user intervention, formalizes it through a reversibility taxonomy, and presents an optimal reactive algorithm validated by experiments.

Contribution

It formalizes the streaming paradigm with a reversibility taxonomy and proposes the Revision Absorber algorithm for efficient, flexible agent revisions during execution.

Findings

The reversibility of actions bounds an agent's flexibility.

Conflicting compensable actions incur unavoidable adaptation costs.

The Revision Absorber matches full-restart quality while reducing wasted steps.

Abstract

Current LLM agents operate under an implicit but universal assumption: execution is a transaction -- the user submits a request, the agent works in isolation, and only upon completion does the dialogue resume. This forces users into a binary choice: wait for a potentially incorrect output, or interrupt and lose all progress. We reject this assumption and propose the stream paradigm, in which agent execution and user intervention are concurrent, interleaved processes sharing a bidirectional channel. We formalize this paradigm through a reversibility taxonomy that classifies every agent action as Idempotent, Reversible, Compensable, or Irreversible, and arrive at a core conclusion: an agent's flexibility is bounded by its reversibility. We prove that conflicting compensable actions impose unavoidable adaptation costs and that conflicting irreversible actions make full specification satisfaction impossible -- these costs are properties of the action space, not of the algorithm. Guided by this insight, we present the Revision Absorber, a reactive algorithm based on the Earliest-Conflict Rollback rule that is structurally optimal under mild assumptions. Experiments on StreamBench with real LLM agents validate all predictions: the Absorber matches the quality of a brute-force full-restart baseline while wasting an order of magnitude fewer steps of already-completed work, turning mid-execution revisions from a dead-end into a first-class interaction.