Kolors Team, Kuaishou Technology (Haoxiang Cao et al.) · Image Synthesis Benchmark
LongBanana
LongBanana is an expert-designed and expert-verified benchmark for long-context multi-reference image synthesis, covering 90 tasks with about 32 reference images, 20 constraints and 4,670 atomic evaluation items in total. Evaluation uses an entity-relation graph and checklists for content fidelity, relational binding and global constraint satisfaction, with a 15% double-labelled subset showing 89.4% exact agreement and a 0.83 quadratic weighted kappa. The authors also introduce LongBanana-Gen, an agent-harness framework, where a ReAct-Plan-Exec-Reflect workflow gives harness-dependent gains over direct generation. Code and data are stated to be available at long-banana.com.
Open the source ↗Similar Research & Safety releases
- Random Is Almost Enough? Stable and Low-Cost Benchmark Compression via BestRand
George Ling, Shanshan Zhong, Lijin Yang, Hao Yang, Nanxiang Li, Zhongzhan Huang (Carnegie Mellon University) · Benchmark Compression · Week 46, 2026
- Dream-RSI
Google, DeepMind, Maryland & Virginia · Recursive Self-Improvement · Week 39, 2026
- WeatherNext 3
Google DeepMind · WeatherNext 3 · Week 36, 2026
- mdlARC
Mithil Vakde · Efficient ARC-AGI Model · Week 36, 2026
- WikiSkill
Tang, Rashtchian, Ferng, Tomkins, Juan and Vu · Skill Evolution · Week 36, 2026
- Planetary Prediction Engine
Google Research · Planetary Prediction · Week 35, 2026
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.