The reviewed record of science sign in
Pith

arxiv: 2509.03312 · v2 · pith:LWQZ4J66 · submitted 2025-09-03 · cs.CL · cs.MA

AgenTracer: Who Is Inducing Failure in the LLM Agentic Systems?

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:LWQZ4J66record.jsonopen to challenge →

classification cs.CL cs.MA
keywords agenticfailureagentracer-8bmulti-agentsystemsagentracerattributionlike
0
0 comments X
read the original abstract

Large Language Model (LLM)-based agentic systems, often comprising multiple models, complex tool invocations, and orchestration protocols, substantially outperform monolithic agents. Yet this very sophistication amplifies their fragility, making them more prone to system failure. Pinpointing the specific agent or step responsible for an error within long execution traces defines the task of agentic system failure attribution. Current state-of-the-art reasoning LLMs, however, remain strikingly inadequate for this challenge, with accuracy generally below 10%. To address this gap, we propose AgenTracer, the first automated framework for annotating failed multi-agent trajectories via counterfactual replay and programmed fault injection, producing the curated dataset TracerTraj. Leveraging this resource, we develop AgenTracer-8B, a lightweight failure tracer trained with multi-granular reinforcement learning, capable of efficiently diagnosing errors in verbose multi-agent interactions. On the Who&When benchmark, AgenTracer-8B outperforms giant proprietary LLMs like Gemini-2.5-Pro and Claude-4-Sonnet by up to 18.18%, setting a new standard in LLM agentic failure attribution. More importantly, AgenTracer-8B delivers actionable feedback to off-the-shelf multi-agent systems like MetaGPT and MaAS with 4.8-14.2% performance gains, empowering self-correcting and self-evolving agentic AI.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 30 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. SAGE-OPD: Selective Agent-Guided Intervention for Multi-Turn On-Policy Distillation

    cs.CL 2026-06 unverdicted novelty 7.0

    SAGE-OPD improves multi-turn OPD via turn-level selective intervention, teacher-confidence weighting, and loss normalization, reporting up to 13.3% relative gain in ALFWorld unseen success rate over standard OPD.

  2. Environment-Grounded Automated Prompt Optimization for LLM Game Agents

    cs.CL 2026-06 unverdicted novelty 7.0

    A multi-agent prompt optimization system using environment feedback improves LLM agent success rates on BabyAI tasks from 0% to 72.5% on challenging coordination tasks.

  3. CASPIAN: Online Detection and Attribution of Cascade Attacks in LLM Multi-Agent Systems via Cross-Channel Causal Monitoring

    cs.MA 2026-05 unverdicted novelty 7.0

    CASPIAN introduces unified cross-channel causal monitoring via late-interaction conditional transfer entropy to detect cascade onset and attribute origin, bridge, and amplifier agents in LLM multi-agent systems.

  4. Holistic Evaluation and Failure Diagnosis of AI Agents

    cs.AI 2026-05 unverdicted novelty 7.0

    A span-decomposed evaluation framework for AI agents achieves state-of-the-art results on GAIA and SWE-Bench with up to 3.5x gains in localization accuracy by breaking traces into independent per-span judgments.

  5. Do Coding Agents Understand Least-Privilege Authorization?

    cs.CR 2026-05 unverdicted novelty 7.0

    Coding agents struggle to infer least-privilege file permissions by omitting needed accesses while granting unused or sensitive ones, but Sufficiency-Tightness Decomposition improves sensitive-task success by up to 15...

  6. ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents

    cs.AI 2026-05 conditional novelty 7.0

    ClawForge supplies a generator that turns scenario templates into reproducible command-line tasks testing state conflict handling, where the strongest frontier model scores only 45.3 percent strict accuracy.

  7. ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents

    cs.AI 2026-05 unverdicted novelty 7.0

    ClawForge is a generator framework that creates reproducible executable benchmarks for command-line agents under state conflict, with ClawForge-Bench showing frontier models reach at most 45.3% strict accuracy and tha...

  8. AgentForesight: Online Auditing for Early Failure Prediction in Multi-Agent Systems

    cs.CL 2026-05 unverdicted novelty 7.0

    AgentForesight trains a 7B model to perform online auditing of multi-agent LLM trajectories, detecting early decisive errors and outperforming larger models on custom and external benchmarks.

  9. AgentForesight: Online Auditing for Early Failure Prediction in Multi-Agent Systems

    cs.CL 2026-05 unverdicted novelty 7.0

    AgentForesight introduces an online auditor model that predicts decisive errors in multi-agent trajectories at the earliest step using a coarse-to-fine reinforcement learning recipe on a new curated dataset AFTraj-2K.

  10. Uncertainty Propagation in LLM-Based Systems

    cs.SE 2026-04 unverdicted novelty 7.0

    This paper introduces a systems-level conceptual framing and a three-level taxonomy (intra-model, system-level, socio-technical) for uncertainty propagation in compound LLM applications, along with engineering insight...

  11. What Resolve Rate Hides: Trajectory Structure Diagnostics for Coding Agents

    cs.SE 2026-07 conditional novelty 6.0

    TraceProbe normalizes coding agent trajectories into canonical actions and applies rule-based detectors to localize failure patterns and behavioral divergences that resolve rate hides.

  12. Causal Agent Replay: Counterfactual Attribution for LLM-Agent Failures

    cs.LG 2026-06 unverdicted novelty 6.0

    Causal Agent Replay attributes LLM agent failures via do-interventions on step-level structural causal models, using contrastive estimators and Monte-Carlo Shapley values validated on synthetic ground truth.

  13. Strained Coherence: A Pre-Failure Signal in Coding Agent Execution Trajectories

    cs.LG 2026-06 unverdicted novelty 6.0

    Strained coherence flagged by Claude judge on 44 coding trajectories predicts failure (94% vs 46%, p=0.003), with partial replication on second model.

  14. Signal-Driven Observation for Long-Horizon Web Agents

    cs.CL 2026-06 unverdicted novelty 6.0

    Signal-Driven Observation decouples observation from action frequency in long-horizon web agents by invoking selective task-relevant DOM reads only on signals such as URL changes or action failures.

  15. StepFinder: A Temporal Semantic Framework for Failure Attribution in Multi-Agent Systems

    cs.AI 2026-06 unverdicted novelty 6.0

    StepFinder turns execution logs into temporal semantic sequences via LLMs then uses temporal modeling plus attention to attribute failures to specific steps more accurately and 79% faster than direct LLM methods on th...

  16. From Confident Closing to Silent Failure: Characterizing False Success in LLM Agents

    cs.LG 2026-06 accept novelty 6.0

    False success occurs in 3-76% of LLM agent failures; LLM judges reach at most 0.65 AUROC while TF-IDF detectors reach 0.83-0.95 and recover 4-8x more cases at equal flag rate.

  17. FALAT: Tracing Failures in LLM Agent Trajectories via Dependency-Guided Search

    cs.AI 2026-05 unverdicted novelty 6.0

    FALAT improves failure attribution in LLM agent trajectories via dependency-guided search, achieving 46.0% step-level accuracy on algorithm-generated and 29.1% on hand-crafted trajectories in the Who&When benchmark.

  18. Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems

    cs.MA 2026-05 unverdicted novelty 6.0

    Meta-Team is a collaborative self-evolution framework that turns multi-agent execution experience into reusable improvements at agent, coordination, and team levels, outperforming baselines on six benchmarks.

  19. TrajAudit: Automated Failure Diagnosis for Agentic Coding Systems

    cs.SE 2026-05 unverdicted novelty 6.0

    TrajAudit diagnoses failures in repository-level agentic coding trajectories by filtering noise and injecting test-failure priors, achieving >24.4 pp higher localization accuracy and 18% lower token use on the new Roo...

  20. Insights Generator: Systematic Corpus-Level Trace Diagnostics for LLM Agents

    cs.AI 2026-05 unverdicted novelty 6.0

    Insights Generator is a multi-agent system that produces evidence-backed natural-language reports characterizing systematic patterns across populations of LLM agent execution traces.

  21. Insights Generator: Systematic Corpus-Level Trace Diagnostics for LLM Agents

    cs.AI 2026-05 unverdicted novelty 6.0

    Insights Generator is a multi-agent system that generates evidence-backed natural-language insights characterizing systematic patterns across corpora of LLM agent execution traces.

  22. VerifyMAS: Hypothesis Verification for Failure Attribution in LLM Multi-Agent Systems

    cs.CL 2026-05 unverdicted novelty 6.0

    VerifyMAS improves failure attribution in LLM multi-agent systems via hypothesis verification on full trajectories, error taxonomy-based data construction, and fine-tuned verifier models, outperforming prior direct-pr...

  23. PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement

    cs.AI 2026-05 unverdicted novelty 6.0

    PIVOT refines LLM agent trajectories through plan-inspect-evolve-verify stages using environment feedback, yielding up to 94% relative gains in constraint satisfaction and 3-5x token efficiency over prior refinement methods.

  24. Shepherd: Enabling Programmable Meta-Agents via Reversible Agentic Execution Traces

    cs.AI 2026-05 unverdicted novelty 6.0

    Shepherd provides a reversible execution trace substrate for LLM agents that enables meta-agents to inspect and transform runs, yielding reported gains on coding and terminal benchmarks via supervision, counterfactual...

  25. Complete Cyclic Subtask Graphs for Tool-Using LLM Agents: Flexibility, Cost, and Bottlenecks in Multi-Agent Workflows

    cs.MA 2026-04 unverdicted novelty 6.0

    Complete cyclic subtask graphs offer a lens to measure when multi-agent revisitation aids recovery and exploration versus when it increases costs or is dominated by other bottlenecks in LLM agent workflows.

  26. KYA: A Framework-Agnostic Trust Layer for Autonomous Systems with Verifiable Provenance and Hierarchical Policy Composition

    cs.CR 2026-05 unverdicted novelty 5.0

    KYA provides a framework-agnostic trust layer using inbound pipelines, policy composition, unified trust scoring, interaction multipliers, and delegation attribution to ensure authorized, conforming, and verifiable ac...

  27. A Sober Look at Agentic Misalignment in Automated Workflows

    cs.AI 2026-05 unverdicted novelty 5.0

    Agentic misalignment in multi-agent systems arises from generic utilities causing posterior collapse; Agentic Evidence Attribution using self-reflection or weak-to-strong generalization provides context-specific evide...

  28. Insights Generator: Systematic Corpus-Level Trace Diagnostics for LLM Agents

    cs.AI 2026-05 unverdicted novelty 5.0

    Insights Generator is a multi-agent system that produces evidence-backed insights from corpora of LLM agent traces and yields 30.4pp performance gains when humans apply the reports.

  29. Shepherd: Enabling Programmable Meta-Agents via Reversible Agentic Execution Traces

    cs.AI 2026-05 unverdicted novelty 5.0 partial

    Shepherd is a runtime system that formalizes meta-agent operations via typed execution traces, enabling fast forking and demonstrated improvements in agent intervention, optimization, and training on benchmarks.

  30. Agent System Operations: Categorization, Challenges, and Future Directions

    cs.MA 2026-06 unverdicted novelty 3.0

    This survey categorizes anomalies in agent systems into intra-agent and inter-agent types and introduces the AgentOps framework with four operational stages.