Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note3 min read

TRAJDEBUG: Tracing Error Lifecycle to Identify Critical Failures in Long-Horizon Agent Trajectories

Source

arXiv

arxiv.org/abs/2608.06346v1

Read the source

Source

  • Source name: arXiv
  • Source kind: paper
  • Authors / organization: Yunjia Qi, Zehua Yin, Xintong Shi, Hao Peng, Songyuanyi Lu, Yixian Liu, Richeng Xuan, Yuhong Liu, Zhichao Hu, Xiaozhi Wang, Lei Hou, Bin Xu, Juanzi Li
  • Venue: arXiv: cs.AI
  • Published: 2026-08-06
  • Collected: 2026-08-07
  • URL: http://arxiv.org/abs/2608.06346v1
  • PDF: https://arxiv.org/pdf/2608.06346v1
  • DOI: UNKNOWN

Plain-Language Brief

The paper proposes TrajDebug, a framework for error-lifecycle tracing in long-trajectory agentic systems. The goal is to identify critical errors responsible for final failures, especially in complex domains where cascading errors can occur. The framework addresses challenges in long trajectories and multiple local errors by using multi-granularity history compression and evidence-based error identification. It also includes TrajErrBench, a benchmark of failed trajectories from Tau2Bench and SWE-Bench Pro. Experiments show that TrajDebug outperforms existing baselines and provides actionable feedback for improving downstream agent success.

What the Document Says

  • The paper presents TrajDebug, a framework designed to trace error lifecycles in long-trajectory LLM-based agentic systems.
  • The main goal is to identify critical errors that lead to final failures in complex domains where cascading errors can occur.
  • Two main challenges are addressed: identifying individual errors in long trajectories and dealing with multiple local errors with different downstream effects.
  • TrajDebug uses multi-granularity history compression and evidence-based error identification for long-trajectory error discovery.
  • It also supports critical attribution by tracing each error's resolution status and terminal impact.
  • The paper introduces TrajErrBench, a benchmark of 486 manually annotated failed trajectories from Tau2Bench and SWE-Bench Pro.
  • The experiments show that TrajDebug outperforms existing baselines across diverse agent benchmarks.
  • Application studies demonstrate that the diagnoses provided by TrajDebug can lead to actionable feedback for improving downstream agent success.
  • The authors state they will release the codes and data for further research.

Key Concepts Mentioned

  • Trajectory
  • Error Lifecycle
  • LLM-based agentic systems
  • Cascading errors
  • Debugging
  • Multi-granularity history compression
  • Evidence-based error identification

Claims Made by the Document

  • TrajDebug addresses long-trajectory error discovery.
  • TrajDebug supports critical attribution.
  • The TrajErrBench benchmark includes 486 manually annotated failed trajectories.
  • TrajDebug outperforms existing baselines.

Evidence Used

  • The paper describes the TrajDebug framework and its methods.
  • Experiments are presented demonstrating the performance of TrajDebug.

Methods / Approach

  • Multi-granularity history compression
  • Evidence-based error identification

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

LLM-based agentic systems have shown remarkable capabilities in complex domains, while suffering from cascading errors and difficulty in debugging. Critical error detection aims to locate the earliest error step in a failed trajectory that is responsible for the final failure. However, progress faces two main challenges. First, long trajectories make it difficult to identify individual errors, since the evidence for judging a step may be scattered across distant instructions, observations, and prior context. Second, failed trajectories often contain multiple local errors with different downstream effects, only some of which remain responsible for the final failure. In this work, we propose TrajDebug, an error-lifecycle tracing framework that addresses long-trajectory error discovery with multi-granularity history compression and evidence-based error identification, and supports critical attribution by tracing each error's resolution status and terminal impact. We further construct TrajErrBench, a benchmark of 486 manually annotated failed trajectories from Tau2Bench and SWE-Bench Pro, covering realistic tool-use and coding scenarios. Experiments across diverse agent benchmarks show that TrajDebug achieves the best overall performance over existing baselines, and application studies further demonstrate that its diagnoses provide actionable feedback for improving downstream agent success. We will release the codes and data to facilitate further research.