PARSER：為長上下文 LLM 代理並行閱讀、深度推理

摘要

序列記憶代理透過逐塊閱讀並維持緊湊的記憶狀態來處理長文件，同時將文件遍歷與推理深度耦合。這種耦合引入了對證據位置的敏感度，並使推理延遲與文件長度呈線性關係。我們提出PARSER，它將閱讀與推理解耦。一組輕量級子代理各自綁定到單個塊，並行讀取整個文件，而一個主代理透過迭代的分散-聚集回合進行深度推理：在每一回合中，它向所有子代理廣播查詢，聚合返回的證據，並根據目前為止發現的內容制定更深入的後續查詢。這種解耦設計將所有可學習行為集中在主代理中，並使用強化學習進行優化，而子代理則保持為凍結的現成模型。在上下文範圍從7K到896K符元的多跳問答中，具有4B主幹的PARSER平均優於最強的序列記憶基準5.7分，在896K符元時高出12.0分。擴展到9B主幹時，PARSER超越DeepSeek-V4-Pro 6.3分。對照實驗證實，PARSER對證據位置、順序和距離的擾動具有穩健性，這些條件會導致序列方法出現大的準確率波動，同時將推理延遲降低多達11倍。

English

Sequential memory agents process long documents by reading chunks one after another while maintaining a compact memory state, coupling document traversal to reasoning depth. This coupling introduces sensitivity to evidence placement and ties inference latency linearly to document length. We introduce PARSER, which decouples reading from reasoning. A bank of lightweight subagents each bound to a single chunk read the entire document in parallel, while a lead agent reasons in depth through iterative scatter--gather rounds: at each round it broadcasts a query to all subagents, aggregates the returned evidence, and formulates a deeper follow-up query conditioned on what has been found so far. This decoupled design concentrates all learnable behavior in the lead agent, which is optimized with reinforcement learning, while the subagents remain frozen off-the-shelf models. On multi-hop QA with contexts ranging from 7K to 896K tokens, PARSER with a 4B backbone outperforms the strongest sequential memory baseline by 5.7 points on average and by 12.0 points at 896K tokens. Scaling to a 9B backbone, PARSER surpasses DeepSeek-V4-Pro by 6.3 points. Controlled experiments confirm that PARSER is robust to perturbations in evidence position, order, and distance, conditions that cause large accuracy swings in sequential methods, while reducing inference latency by up to 11x.