Sign InOpen Brain
arXivPaperNeeds Review

Asymmetric Capacity Allocation in Self-Refinement Pipelines

Self-refinement pipelines need not use equally capable models: invest capacity in generation and revision, while a small critic may preserve gains at lower compute cost.

arXiv · Aug 21, 2026
Open Source Open MarkdownOpen JSON
Source Summary

Across **5 benchmarks**, the study varies **6 Qwen3 sizes** and **4 Gemma 3 sizes** across generation, critique, and revision. Larger generators and refiners generally help, while an undersized refiner can reduce performance.

Practical Implication

For agent pipelines, allocate the strongest affordable models to generation and revision. Keep critique in the loop, but test a smaller critic: results were largely insensitive to critic size, and even a small critic beat omitting critique.

Agent-Ready Context
Across **5 benchmarks**, the study varies **6 Qwen3 sizes** and **4 Gemma 3 sizes** across generation, critique, and revision. Larger generators and refiners generally help, while an undersized refiner can reduce performance.

For agent pipelines, allocate the strongest affordable models to generation and revision. Keep critique in the loop, but test a smaller critic: results were largely insensitive to critic size, and even a small critic beat omitting critique.

These are stage-level findings across two model families, not a universal routing formula. Builders still need workload-specific evaluations before turning asymmetric allocation into a fixed policy.
Connected Context · Feed7 Judgment

This narrows model routing from choosing one model per task or turn to allocating capacity by pipeline stage: generation and revision are the sensitive positions, while critique may remain useful with a cheaper model. It supports asymmetric deployment, but only as a workload-tested policy because the evidence spans two model families and does not capture production costs or switching behavior.

Context Map
agentcodingresearch#harness-engineering#model-selection#agent-reliability
Uncertainty
These are stage-level findings across two model families, not a universal routing formula. Builders still need workload-specific evaluations before turning asymmetric allocation into a fixed policy.