上下文驱动的增量压缩用于多轮对话生成
Context-Driven Incremental Compression for Multi-Turn Dialogue Generation
AI总结：
提出上下文驱动的增量压缩（C-DIC），通过可修订的线程压缩状态和轻量级检索-修订-写回循环，实现跨轮信息共享，稳定长对话性能。
摘要：
现代对话代理在每一轮都会处理不断增长的对话历史，导致冗余的注意力和编码成本随对话长度增加。简单的截断或摘要会降低保真度，而现有的上下文压缩器缺乏跨轮记忆共享或修订，导致信息丢失和长对话中的累积错误。我们重新审视了对话动态下的上下文压缩，并经验性地展示了其脆弱性。为了提高效率和鲁棒性，我们引入了上下文驱动的增量压缩（C-DIC），它将对话视为交织的上下文线程，并在单个紧凑的对话记忆中存储每个线程的可修订压缩状态。在每一轮，一个轻量级的检索、修订和写回循环在轮次之间共享信息并更新过时的记忆，从而稳定长期行为。此外，我们将截断反向传播（TBPTT）适应于我们的多轮设置，学习跨轮依赖关系而无需完整历史反向传播。在长对话基准上的大量实验证明了C-DIC的优越性能和效率；值得注意的是，C-DIC在数百轮对话中表现出稳定的推理延迟和困惑度，为高质量对话建模提供了一条可扩展的路径。
英文摘要：
Modern conversational agents condition on an ever-growing dialogue history at each turn, incurring redundant attention and encoding costs that grow with conversation length. Naive truncation or summarization degrades fidelity, while existing context compressors lack cross-turn memory sharing or revision, causing information loss and compounding errors in long dialogues. We revisit the context compression under conversational dynamics and empirically present its fragility. To improve both efficiency and robustness, we introduce Context-Driven Incremental Compression (C-DIC), which treats a conversation as interleaved contextual threads and stores revisable per-thread compression states in a single, compact dialogue memory. At each turn, a lightweight retrieve, revise, and write-back loop shares information across turns and updates stale memories, stabilizing long-horizon behavior. In addition, we adapt truncated backpropagation-through-time (TBPTT) to our multi-turn setting, learning cross-turn dependencies without full-history backpropagation. Extensive experiments on long-form dialogue benchmarks demonstrate superior performance and efficiency of C-DIC; notably, C-DIC shows stable inference latency and perplexity over hundreds of dialogue turns, supporting a scalable path to high-quality dialogue modeling.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
