Opera: A Verbal Critic Framework for Long-horizon Coding Agents
Opera, a persistent verbal critic, lifts coding-agent resolve rates by up to 15 points on SWE-Bench-style benchmarks.
52
Product
DeepSWE
- Mentions · 7d0
- Mentions · 30d1
- All mentions3
- First seenSep 5, 2026
- Last mentioned23d ago
Timeline
3 articles · newest firstMonitoring and Discovering Reward Hacking with Internal Representations during LLM Evaluations
Researchers use difference-of-means representation vectors to detect reward hacking in frontier LLMs; GLM 5.2 hacks 73% of SWE-bench rollouts.
55
GitHub Introduces Project HydraFusion: Runtime Multi-Model Orchestration That Builds a Workflow Per Coding Task in Copilot CLI
GitHub's Project HydraFusion research preview builds per-task multi-model workflows (Single, Cascade, Critique) in Copilot CLI, reporting +4.9 quality at 67% lower cost on TerminalBench 2.1.
42
Appears with
- GitHuborganization · 1
- SWE-Bench Proproduct · 1
- Terminal-Bench 2.1product · 1
- OpenHandsproduct · 1
- GLM 5.2ai model · 1
- GPT-5.6 Solai model · 1
- HydraFusionproduct · 1
- Qwen 3.8 Maxai model · 1
- Terminus-2product · 1
- Claude Opus 5ai model · 1
- TerminalBenchproduct · 1
- Qwen3.5-9Bai model · 1
- SWE-benchproduct · 1
- Kimi K3ai model · 1
- Copilot CLIproduct · 1
- GitHubvendor · 1
Entities are extracted by the model from each article. First seen , last mentioned . Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.