Pending updates

[arvindpdmn]

New developments can go into the mindmap:

• Distillation: save to traditional LLM
• Saving ground truth datasets
• Prompt caching
• Prompt unit testing / save checkpoints
• OpenLlama, LongLlama, FoT: https://github.com/CStanKonrad/long_llama
• Pruning: Sheared LLaMA
• Small/medium LLM: p. 33, https://arxiv.org/pdf/2402.06196
• TinyAgent, LLMCompiler http://bair.berkeley.edu/blog/2024/05/29/tiny-agent/
• Hailo AI processor, plus Raspberry PI
• LoRA variants
• MobileLLM
• MIT GenSQL
• Reasoning https://arxiv.org/abs/2407.11511 https://arxiv.org/abs/2407.11229
• Mind https://arxiv.org/pdf/2407.11015
• EfficientQAT
• Explaining LLM https://arxiv.org/abs/2303.16537
• https://arxiv.org/pdf/2403.08819