This work proposes Root-Cause Trajectory Attribution (RCTA), a training-free method that organizes original candidate records and explicit handoff instructions for attribution that reaches 24.1% exact root-step accuracy and 51.1% responsible-role accuracy.
Abstract
In long agent executions, an early error can persist through later actions and checks, while evidence needed to trace its origin is dispersed across the history. Short histories offer limited tests of recovering error origins across substantial subsequent execution. We introduce LongRCA Bench: 1,140 complete failed trajectories from five sources, all human-annotated for responsible roles and earliest decisive root-cause steps. Reference roots precede completion by a median of 48 recorded steps; 28.4% of trajectories contain over 100 subsequent steps. With DeepSeek-V4-Flash on the full benchmark, the strongest of five evaluated baselines achieves 13.2% exact root-step accuracy. We propose Root-Cause Trajectory Attribution (RCTA), a training-free method that organizes original candidate records and explicit handoff instructions for attribution. Segment summaries and a trajectory outline guide candidate retrieval; available handoff records supply upstream instruction context for the final instruction-execution comparison. With the same backbone and scoring protocol, RCTA reaches 24.1% exact root-step accuracy and 51.1% responsible-role accuracy. LongRCA-Mini provides 200 fixed trajectories for lower-cost comparative screening. Even with RCTA, fewer than one quarter of reference roots are recovered exactly.
DynSTEER bridges the granularity gap through stage-wise dynamic evaluation that segments rollouts at key execution nodes and adapts its multi-level review strategy based on stage-level results; it compiles a path-tolerant milestone graph from available task inputs to preserve diverse legal paths without reference leaka...
Zhichao Shi, Xuhui Jiang, Wen-Jie Zhang et al.· 1 citation
Recovering from failures in long-horizon, stateful language agents is challenging because the action introducing a defect, the symptom exposing it, and the persistent state requiring repair frequently diverge. Existing methods largely focus on retrospective failure attribution, leaving sequential diagnosis and state re...
CUA-Recovery is introduced, a benchmark comprising test and training suites across multiple desktop applications using ReRail, a framework for synthesizing long-horizon tasks and trajectories in a persistent desktop environment that shows that agents struggle both to recognize inherited errors and to complete the task.
Xue Hu, Hao-Ming Li, Yuxuan Wang et al.· 0 citations
Parallel search may generate a correct answer that final-answer voting fails to select. We formulate this consolidation stage as trajectory selection and introduce TRACE (Trajectory Ranking with Aggregated Cross-Rollout Evidence), a lightweight learned selector that ranks completed trajectories using the search evidenc...
MemTrace is a provenance-aware memory system that preserves execution history and aligns its reuse with the evolving task (e.g., iterative cross-file repair) and repository state and retrieves only what the next action requires.
Hong-Ming Xu, Le Zhou, Zhong-He Jin et al.· 0 citations
Long-horizon agentic workflows require models to sustain repeated state-dependent actions all while the context grows, sub-task complexity changes, and new data arrives. Each situation represents an independent axis along which an agent may fail. An agent reconciling a long ledger, for example, must repeatedly read its...
Jeffrey Willette, Krishna C. Puvvada, Boris Ginsburg· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.