Generation Model Survey

[Aidenzich]

Model Efficiency and Resource Utilization Improvements

• [x] INSTRUCTION TUNING WITH GPT-4
• [x] Full Parameter Fine-tuning for Large Language Models with Limited Resources
• [x] ICML'23 SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models
• [x] A Theory on Adam Instability in Large-Scale Machine Learning
• [x] Patch Diffusion: Faster and More Data-Efficient Training of Diffusion Models

The abilities and limitations of LLMs

• [x] Evaluating Large Language Models Trained on Code
• [x] Can Large Language Models Infer Causation from Correlation?
• [x] Are Emergent Abilities of Large Language Models a Mirage?
• [x] Scaling Transformer to 1M tokens and beyond with RMT
• [x] Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation
• [x] Faith and Fate: Limits of Transformers on Compositionality
• [x] LongNet: Scaling Transformers to 1,000,000,000 Tokens
• [x] Less is More: Why Use Retrieval Instead of Larger Context Windows

  Multi-Modal

• [x] LLaMA-Adapter V2
• [x] Visual Instruction Tuning

Interpretability

• [x] What the DAAM: Interpreting Stable Diffusion Using Cross Attention

Articles

• [x] LLM Powered Autonomous Agents

Prompt Engineering

• [ ] System 2 Attention (is something you might need too)