small language models
AI models designed for natural language processing that have fewer parameters compared to larger models, which may be more efficient and suitable for specific applications, especially on limited resources.
- A Token is Worth over 1,000 Tokens: Efficient Knowledge Distillation through Low-Rank Clone
- BREAD: Branched Rollouts from Expert Anchors Bridge SFT & RL for Reasoning
- Distilling LLM Agent into Small Models with Retrieval and Code Tools
- Nemotron-Flash: Towards Latency-Optimal Hybrid Small Language Models
- R2R: Efficiently Navigating Divergent Reasoning Paths with Small-Large Model Token Routing
- Self-Refining Language Model Anonymizers via Adversarial Distillation
- Unlocking SLM Potential for Data Analysis Code Generation via Non-Parametric Knowledge Distillation