ResearchCL4SE Context Learning Benchmark for SE Tasks 24.7% ImprovementarXiv·high signalFirst standardized eval for context engineering in coding. 13K+ samples, 24.7% avg improvement. Tells which context types matter for which tasks.SourceSource pagearXiv↳ Follow the threadStack layer / Update threadSLBench Finds Up to 70% Unsafe Execution When Codex and Claude Code Follow Logical Relations in Skill FilesarXivStack layer / ContrastLong-Horizon-Terminal-Bench: Best Frontier Agent Clears Just 15% of Real Terminal Tasks, Mean 4.3%arXiv (Tencent Hunyuan) / HuggingFace Daily PapersPolicy dependency / ContrastMake context compaction a trained policy, not an inference-time heuristic — CompactionRL lifts SWE-bench up to +7 pointsarXiv 2607.05378Stack layer'Stop Guessing When to Stop Testing' Pushes Sequential Evaluation to Cut Benchmark CostarXivUpdate thread / Follow-up threadEvoAgentBench: A Benchmark That Isolates Whether Agents Can Turn Past Sessions Into Reusable ProceduresarXivPolicy dependency / Stack layerFirst Recon AI ships AI Security Runtime GA — inline policy enforcement on every agent-to-agent and agent-to-tool callBusiness WirePolicy dependency / Stack layerCodenotary ships AgentMon 3 for agent runtime policy and compliance loggingAI Agent StoreStack layerSelf-Guided Test-Time Training: Let the Model Pick Its Own Evidence Spans for +15% Long-Context AccuracyarXiv / HuggingFace Daily Papers