DART-SD: Diamond-topology Aware Retrieval and Tuning for Self-Distillation of Multi-Turn Tool-Calling Agents

  • 类型：arxiv
  • 标识：2608.18524
  • 链接：https://arxiv.org/abs/2608.18524
  • 主分类：agent
  • 形态：method
  • 被引：2
  • 被引来源：Semantic Scholar
  • S2被引：2
  • OpenAlex被引：0
  • 影响力被引：0
  • TLDR：This work proposes DART-SD (Diamond-topology Aware Retrieval and Tuning for Self-Distillation), a novel framework that shifts the paradigm from global forcing to topology-guided localized correction, and significantly outperforms traditional full-trajectory baselines.
  • OpenAlex ID：W7203915543
  • OpenAlex DOI：10.48550/arxiv.2608.18524
  • DOI：10.48550/arxiv.2608.18524
  • DOI来源：OpenAlex
  • 开放获取：green
  • 开放获取链接：https://arxiv.org/pdf/2608.18524
  • OpenAlex更新：2026-09-02
  • 副分类：rag
  • 待LLM分类：否
  • 标题中文：DART-SD：面向多轮工具调用 Agent 自蒸馏的菱形拓扑感知检索与微调
  • TLDR中文：本文提出 DART-SD (Diamond-topology Aware Retrieval and Tuning for Self-Distillation)，一种将范式从全局 forcing 转向拓扑引导的局部修正的新框架，相对传统 full-trajectory 基线取得显著提升。
  • 来源文件：
  • /inbox/tom/_candidates/2026-08-31-agent-rag-longcontext-candidates.json
  • [OpenAlex backfill]
  • [S2 enrich]