paper
Context-Driven Incremental Compression for Multi-Turn Dialogue Generation
paperactiveprovisional
context-driven-incremental-compression-for-multi-turn-dialogue-generation-3ad82d7d·1 events·first seen 6d agoAliases: Context-Driven Incremental Compression for Multi-Turn Dialogue Generation
Co-occurring entities
More like this (12)
Context-Driven Incremental Compression for Multi-Turn Dialogue GenerationContext-Driven Incremental CompressionAdaptive Turn-Taking for Real-time Multi-Party Voice AgentsMeasuring Semantic Progress in Multi-turn Dialogue via Information GainArtificial Analysis Conversational DynamicsPlanning-aligned Token Compression for Long-Context Autonomous DrivingEnd-to-End Context Compression at ScaleEfficient ASR Training with Conversations that Never HappenedAdaptive Multi-Resolution Procedural Knowledge Compression for Large Language ModelsMulti-Faceted Interactivity Alignment in Full-Duplex Speech ModelsMulti-Turn Evaluation of Deep Research Agents Under Process-Level FeedbackLearning to Hear Hesitation: Continual Learning for Disfluency-Aware ASR
Recent events (1)
C-DIC: Context-Driven Incremental Compression for efficient long-horizon multi-turn dialogue
A new arXiv preprint introduces Context-Driven Incremental Compression (C-DIC), a method for managing growing dialogue history in conversational agents by treating conversations as interleaved contextual threads with revisable per-thread compression states stored in a compact dialogue memory. A retrieve-revise-write-back loop shares information across turns and updates stale memories, while truncated backpropagation-through-time (TBPTT) is adapted to learn cross-turn dependencies. Experiments on long-form dialogue benchmarks show stable inference latency and perplexity over hundreds of turns, addressing compounding errors seen in existing context compressors.