Slipstream: Trajectory-Grounded Compaction Validation for Long-Horizon Agents

ArXi:2605.08580v1 Announce Type: cross To cope with the large contexts that long-horizon LLM agents produce, modern frameworks increasingly rely on compaction -- invoking an LLM to rewrite the accumulated trajectory into a shorter summary that the agent resumes from. Today, compaction runs synchronously on the critical path of agent execution but this can unpredictably degrade accuracy due to a structural validation gap: the compactor must condense context but is fundamentally unaware of precisely what information the agent will need later.