Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

Holistic Evaluation and Failure Diagnosis of AI Agents

Netta Madvil, Gilad Dym, Alon Mecilati, Edo Dekel +11 more

Published

May 14, 2026

Citations

0

Trust level

Moderate

Usefulness score

25/100 (Low)

Extraction confidence

55% (Moderate)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

May 14, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics Long Horizon Medicine

Should you rely on this paper?

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Use this for comparison and orientation, not as your only source.

Best use

Background context only

Use if you need

A benchmark-and-metrics comparison anchor.

What to verify

Validate the evaluation procedure and quality controls in the full paper before operational use.

Main weakness

No major weakness surfaced.

Human feedback signal
Not explicit
Not explicit in abstract metadata
Evaluation signal
Detected
Eval setup described
Usefulness for eval research
25/100
Adjacent candidate

Treat as adjacent context, not a core eval-method reference.

If you are doing eval pipeline work, start here

Human Eval Papers LLM-as-Judge Papers Pairwise Preference Papers Tool-Use Evaluation Search

Abstract

AI agents execute complex multi-step processes, but current evaluation falls short: outcome metrics report success or failure without explaining why, and process-level approaches struggle to connect failure types to their precise locations within long, structured traces. We present a holistic agent evaluation framework that pairs top-down agent-level diagnosis with bottom-up span-level evaluation, decomposing analysis into independent per-span assessments. This decomposition scales to traces of arbitrary length and produces span-level rationales for each verdict. On the TRAIL benchmark, our framework achieves state-of-the-art results across all metrics on both GAIA and SWE-Bench, with relative gains over the strongest prior baselines of up to 38% on category F1, up to 3.5x on localization accuracy, and up to 12.5x on joint localization-categorization accuracy. Per-category analysis shows our framework leading in more error categories than any other evaluator. Notably, the same frontier model achieves several times higher localization accuracy when used inside our framework than as a monolithic judge over the full trace, showing that evaluation methodology, not model capability, is the bottleneck.

What we could verify

These are the protocol signals we could actually recover from the available paper metadata. Use them to decide whether this paper is worth deeper reading.

Human Feedback Types

missing

None explicit

No explicit feedback protocol extracted.

"AI agents execute complex multi-step processes, but current evaluation falls short: outcome metrics report success or failure without explaining why, and process-level approaches struggle to connect failure types to their precise locations within long, structured traces."

Evaluation Modes

strong

Automatic Metrics

Includes extracted eval setup.

"AI agents execute complex multi-step processes, but current evaluation falls short: outcome metrics report success or failure without explaining why, and process-level approaches struggle to connect failure types to their precise locations within long, structured traces."

Quality Controls

missing

Not reported

No explicit QC controls found.

"AI agents execute complex multi-step processes, but current evaluation falls short: outcome metrics report success or failure without explaining why, and process-level approaches struggle to connect failure types to their precise locations within long, structured traces."

Benchmarks / Datasets

strong

GAIA, SWE Bench

Useful for quick benchmark comparison.

"On the TRAIL benchmark, our framework achieves state-of-the-art results across all metrics on both GAIA and SWE-Bench, with relative gains over the strongest prior baselines of up to 38% on category F1, up to 3.5x on localization accuracy, and up to 12.5x on joint localization-categorization accuracy."

Reported Metrics

strong

Accuracy, F1

Useful for evaluation criteria comparison.

"On the TRAIL benchmark, our framework achieves state-of-the-art results across all metrics on both GAIA and SWE-Bench, with relative gains over the strongest prior baselines of up to 38% on category F1, up to 3.5x on localization accuracy, and up to 12.5x on joint localization-categorization accuracy."

Benchmarks and datasets

GAIASWE-bench

Reported metrics

accuracyf1
Human feedback details
Uses human feedback
No
Feedback types
None
Rater population
Not reported
Expertise required
Medicine
Evaluation details
Evaluation modes
Automatic Metrics
Agentic eval
Long Horizon
Quality controls
Not reported
Evidence quality
Moderate
Use this page as
Background context only

Research brief

Metadata summary

AI agents execute complex multi-step processes, but current evaluation falls short: outcome metrics report success or failure without explaining why, and process-level approaches struggle to connect failure types to their precise locations within long, structured traces.

Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.

Key takeaways

  • AI agents execute complex multi-step processes, but current evaluation falls short: outcome metrics report success or failure without explaining why, and process-level approaches struggle to connect failure types to their precise locations within long, structured traces.
  • We present a holistic agent evaluation framework that pairs top-down agent-level diagnosis with bottom-up span-level evaluation, decomposing analysis into independent per-span assessments.
  • This decomposition scales to traces of arbitrary length and produces span-level rationales for each verdict.

Researcher actions

  • Compare this paper against others mentioning SWE-bench.
  • Validate inferred eval signals (Automatic metrics, Long-horizon tasks) against the full paper.
  • Use related-paper links to find stronger protocol-specific references.

Caveats

  • Generated from abstract + metadata only; no PDF parsing.
  • Signals below are heuristic and may miss details reported outside the abstract.

Recommended queries

SWE-bench Accuracy

Contribution summary

  • We present a holistic agent evaluation framework that pairs top-down agent-level diagnosis with bottom-up span-level evaluation, decomposing analysis into independent per-span assessments.
  • On the TRAIL benchmark, our framework achieves state-of-the-art results across all metrics on both GAIA and SWE-Bench, with relative gains over the strongest prior baselines of up to 38% on category F1, up to 3.5x on localization accuracy,…
  • Notably, the same frontier model achieves several times higher localization accuracy when used inside our framework than as a monolithic judge over the full trace, showing that evaluation methodology, not model capability, is the…

Why it matters for eval

  • We present a holistic agent evaluation framework that pairs top-down agent-level diagnosis with bottom-up span-level evaluation, decomposing analysis into independent per-span assessments.
  • On the TRAIL benchmark, our framework achieves state-of-the-art results across all metrics on both GAIA and SWE-Bench, with relative gains over the strongest prior baselines of up to 38% on category F1, up to 3.5x on localization accuracy,…

Researcher checklist

  • Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • Evaluation mode is explicit

    Detected: Automatic Metrics

  • Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Benchmark or dataset anchors are present

    Detected: GAIA, SWE-bench

  • Metric reporting is present

    Detected: accuracy, f1