AgentDebugX: An Open-Source Toolkit for Failure Observability, Attribution, and Recovery in LLM Agents
- 类型:arxiv
- 标识:2607.18754
- 链接:https://arxiv.org/abs/2607.18754
- 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:DeepDebug achieves the best strict attribution accuracy among the evaluated methods on both tested open-weight backbones, reaching 28.8 percent exact agent-and-step accuracy on qwen3.5-9b versus 21.7 percent for the strongest single-pass baseline.
- OpenAlex ID:W7170058005
- OpenAlex DOI:10.48550/arxiv.2607.18754
- DOI:10.48550/arxiv.2607.18754
- DOI来源:OpenAlex
- 开放获取:green
- 开放获取链接:https://doi.org/10.48550/arxiv.2607.18754
- OpenAlex更新:2026-08-24
- 待LLM分类:否
- 标题中文:AgentDebugX:面向 LLM Agent 失败可观测性、归因与恢复的开源工具包
- TLDR中文:DeepDebug 在两个测试的开源权重 backbone 上均取得了所评估方法中最高的严格归因准确率,在 qwen3.5-9b 上达到 28.8% 的精确 agent 与步骤准确率,而最强的单遍 baseline 为 21.7%。
- 来源文件:
- /inbox/tom/_candidates/2026-07-22-agent-rag-longcontext-candidates.json
- [S2 enrich]
- [OpenAlex backfill]