Forschung
Wir trainieren kleine, spezialisierte Modelle, die Allzweck-LLMs bei Netzoperationen übertreffen
Neueste
Decoupling reasoning and facts in language models
On decoupling reasoning and facts in language models
Mahmoud
10 min
The Limits of prompt optimization in high-alignment model pairs
The Limits of prompt optimization in high-alignment model pairs
Eya
5 min
Machine learning
Utilization rate forecasting for EV charging infrastructure
Utilization rate forecasting for EV charging infrastructure
Abdulrahman
8 min
Machine learning
Real-Time Human Activity Recognition Using the H2O AutoML Framework and Imbalance Handling
Real-Time Human Activity Recognition Using the H2O AutoML Framework and Imbalance Handling
Eya
5 min
AI
The few-shot distraction effect
The Few-shot distraction effect —> when more examples degrade LLM constraint adherence
4 min
AI
Small vs Big - NanoChat vs GPT-OSS 20B
nanochat implements a dense transformer decoder with several architectural refinements that have become mainstream since the original GPT-2 paper.
, MNN
10 min