📄 LIMI: Less Is More for Agencies New research shows that when it comes to agency AI, quality beats quantity! 💡 Key Takeaways: - 78 carefully curated demos scored 73.5% on AgencyBench – outshining models trained on 10k examples! 📈 - Results are higher than SOTA: Kimi-K2 (24.1%), DeepSeek (11.9%), Qwen3 (27.5%), GLM-4.5 (45.1%). 🤯 - Introducing the Agency Efficiency principle: AI autonomy comes from strategic data selection, not sheer volume! 🔍 Check it out: https://github.com/GAIR-NLP/LIMI
📄 LIMI: Less Is More for Agencies
New research shows that when it comes to agency AI, quality beats quantity! 💡
Key Takeaways:
- 78 carefully curated demos scored 73.5% on AgencyBench – outshining models trained on 10k examples! 📈
- Results are higher than SOTA: Kimi-K2 (24.1%), DeepSeek (11.9%), Qwen3 (27.5%), GLM-4.5 (45.1%). 🤯
- Introducing the Agency Efficiency principle: AI autonomy comes from strategic data selection, not sheer volume! 🔍
Check it out: https://github.com/GAIR-NLP/LIMI
