ResearchResearch paperLarge Language Models · Reinforcement Learning · Efficiency & Inference1 source · Oct 8, 2026

Opera: A Verbal Critic Framework for Long-horizon Coding Agents

We present Opera, a verbal critic framework that treats each correction as a persistent note, followed until the diagnosed problem is resolved.

Long-horizon coding agents need timely corrections, yet feedback can be ineffective or even harmful when it misjudges ongoing work or fails to address the underlying problem.
Existing critics focus on evaluating trajectories and generating feedback, but rarely track what happens after feedback is delivered.
Opera decides when to review through periodic and event-driven triggers, diagnoses issues with typed operators, audits feedback against visible evidence before delivery, and tracks the agent's subsequent actions to distinguish mere compliance from actual resolution.
Beyond inference, Opera-guided rollouts provide approximately on-policy training data: fine-tuning Qwen3.5-9B on them improves its resolve rate on held-out SWE-Bench Pro repositories by 10.2 percentage points without a critic at inference time, matching fine-tuning on rollouts from a stronger model, while preserving its performance when switching harness, i.e., from Openhands to Terminus-2, which the latter substantially degrades.

[1]Opera: A Verbal Critic Framework for Long-horizon Coding Agents
Hugging Face Daily Papers · Oct 8, 12:00 AM
We present Opera, a verbal critic framework that treats each correction as a persistent note, followed until the diagnosed problem is resolved.
Long-horizon coding agents need timely corrections, yet feedback can be ineffective or even harmful when it misjudges ongoing work or fails to address the underlying problem.

Extractive summary: sentences quoted from the sources.

Before this