# 65labs

- **Event:** [AIE Code Agents Hackathon](https://cerebralvalley.ai/e/aie-code-agents-nyc)
- **When:** Nov 22 at 9:00 AM – Nov 23 at 5:00 PM (EST)
- **Where:** The Malin Flatiron, New York, NY
- **Team:** [Agrim Singh](https://cerebralvalley.ai/u/agrim), [Sherry Jiang](https://cerebralvalley.ai/u/sherryyjiang), [Ivan Leo](https://cerebralvalley.ai/u/ivanleomk)
- **GitHub:** https://github.com/agrimsingh/rl-nano
- **Demo video:** https://drive.google.com/file/d/15MmWU2-pE0z62BWbRdJ7-f8hkpsphfWn/view?usp=sharing
- **Gallery:** https://cerebralvalley.ai/e/aie-code-agents-nyc/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/aie-code-agents-nyc/hackathon/gallery/17

🍌 Nono Banana: A Controllable Benchmark for Non-English LLM Text Recognition
LLMs still struggle to reliably extract non-english text from images. This has become a huge blocker for existing models, as over a billion people read and write Mandarin alone, yet most evaluation data for this task is tiny, messy, and english-centric. The core issue is simple: there’s no controlled, scalable way to test how LLMs behave on complex scripts. The existing datasets are scraped, mislabeled, inconsistent, and impossible to tune. you can’t say ‘make this 20% harder’ or systematically test radicals, stroke density, angles, blur, or font variation.

Nono Banana fixes that. The name is a small RL wink — you keep saying ‘no no’ until the model improves — but the tech is the serious part.

---

Markdown version of https://cerebralvalley.ai/e/aie-code-agents-nyc/hackathon/gallery/17. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt