Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

LEASH: Uncertainty-Aware Weak-to-Strong Escalation and Oversight for Computer-Use Agents

Qingni Wang, Changdae Oh, Yuzhe Yang, Sharon Li +3 more

Published

Oct 15, 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

Oct 15, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics General Long Horizon

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

Computer-use agents increasingly tackle long-horizon tasks, where running a frontier model at every step is expensive and a small open-weight model alone often fails. Weak-to-strong escalation offers a middle ground where a weak agent acts by default and hands control to a strong one only when needed. The core challenge is deciding when to escalate, which requires an uncertainty signal cheap enough to check at every step. However, even after handoff, the strong agent can still get lost, but a black-box API that the strong agent is usually served exposes no internals to monitor. To this end, we propose LEASH, a training-free framework that unleashes the strong agent only when the weak agent's latent uncertainty signals it is needed, and keeps it on a leash afterward. LEASH has three layers: (1) initial task uncertainty for early routing, (2) step-level uncertainty for runtime escalation, and, after handoff, (3) a shadow observer for continued oversight of the strong black-box agent. Here, we use latent uncertainty as a fuel for all three layers. Specifically, we compare the weak model's hidden states with those from its past successful and failed trajectories, and a state that looks more like the failures signals higher uncertainty. During weak-agent execution, our uncertainty estimator reuses activations already produced during action generation, so it needs no extra forward pass or dedicated monitor training. Experiments on OSWorld across four contemporary model backbones show that LEASH consistently improves over standalone weak agents and outperforms random escalation by 6.55--11.94 points at matched escalation rates. With Qwen3.5-35B-A3B, LEASH achieves 73.66% task success at $0.4117 per task, matching always-on Claude Opus 5 (72.90%) while reducing inference cost by 19.2%.

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.

"Computer-use agents increasingly tackle long-horizon tasks, where running a frontier model at every step is expensive and a small open-weight model alone often fails."

Evaluation Modes

strong

Automatic Metrics

Includes extracted eval setup.

"Computer-use agents increasingly tackle long-horizon tasks, where running a frontier model at every step is expensive and a small open-weight model alone often fails."

Quality Controls

missing

Not reported

No explicit QC controls found.

"Computer-use agents increasingly tackle long-horizon tasks, where running a frontier model at every step is expensive and a small open-weight model alone often fails."

Benchmarks / Datasets

strong

OSWorld

Useful for quick benchmark comparison.

"Experiments on OSWorld across four contemporary model backbones show that LEASH consistently improves over standalone weak agents and outperforms random escalation by 6.55--11.94 points at matched escalation rates."

Reported Metrics

strong

Task success, Inference cost

Useful for evaluation criteria comparison.

"With Qwen3.5-35B-A3B, LEASH achieves 73.66% task success at $0.4117 per task, matching always-on Claude Opus 5 (72.90%) while reducing inference cost by 19.2%."

Benchmarks and datasets

OSWorld

Reported metrics

task successinference cost
Human feedback details
Uses human feedback
No
Feedback types
None
Rater population
Not reported
Expertise required
General
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

Computer-use agents increasingly tackle long-horizon tasks, where running a frontier model at every step is expensive and a small open-weight model alone often fails.

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

Key takeaways

  • Computer-use agents increasingly tackle long-horizon tasks, where running a frontier model at every step is expensive and a small open-weight model alone often fails.
  • Weak-to-strong escalation offers a middle ground where a weak agent acts by default and hands control to a strong one only when needed.
  • The core challenge is deciding when to escalate, which requires an uncertainty signal cheap enough to check at every step.

Researcher actions

  • Compare this paper against nearby papers in the same arXiv category before using it for protocol decisions.
  • Validate inferred eval signals (Tool-use evaluation, 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

Contribution summary

  • Computer-use agents increasingly tackle long-horizon tasks, where running a frontier model at every step is expensive and a small open-weight model alone often fails.
  • Weak-to-strong escalation offers a middle ground where a weak agent acts by default and hands control to a strong one only when needed.
  • To this end, we propose LEASH, a training-free framework that unleashes the strong agent only when the weak agent's latent uncertainty signals it is needed, and keeps it on a leash afterward.

Why it matters for eval

  • Computer-use agents increasingly tackle long-horizon tasks, where running a frontier model at every step is expensive and a small open-weight model alone often fails.
  • To this end, we propose LEASH, a training-free framework that unleashes the strong agent only when the weak agent's latent uncertainty signals it is needed, and keeps it on a leash afterward.

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: OSWorld

  • Metric reporting is present

    Detected: task success, inference cost