Research Papers A collection of papers focused on LLM Orca 2: Teaching Small Language Models How to Reason Paper β’ 2311.11045 β’ Published Nov 18, 2023 β’ 75 ToolTalk: Evaluating Tool-Usage in a Conversational Setting Paper β’ 2311.10775 β’ Published Nov 15, 2023 β’ 10 Adapters: A Unified Library for Parameter-Efficient and Modular Transfer Learning Paper β’ 2311.11077 β’ Published Nov 18, 2023 β’ 28 MultiLoRA: Democratizing LoRA for Better Multi-Task Learning Paper β’ 2311.11501 β’ Published Nov 20, 2023 β’ 37
Orca 2: Teaching Small Language Models How to Reason Paper β’ 2311.11045 β’ Published Nov 18, 2023 β’ 75
ToolTalk: Evaluating Tool-Usage in a Conversational Setting Paper β’ 2311.10775 β’ Published Nov 15, 2023 β’ 10
Adapters: A Unified Library for Parameter-Efficient and Modular Transfer Learning Paper β’ 2311.11077 β’ Published Nov 18, 2023 β’ 28
MultiLoRA: Democratizing LoRA for Better Multi-Task Learning Paper β’ 2311.11501 β’ Published Nov 20, 2023 β’ 37
Synthetic Data Generation A curated list of papers focusing on synthetic data generation Synthetic Data (Almost) from Scratch: Generalized Instruction Tuning for Language Models Paper β’ 2402.13064 β’ Published Feb 20, 2024 β’ 48 Textbooks Are All You Need II: phi-1.5 technical report Paper β’ 2309.05463 β’ Published Sep 11, 2023 β’ 87 DataDreamer: A Tool for Synthetic Data Generation and Reproducible LLM Workflows Paper β’ 2402.10379 β’ Published Feb 16, 2024 β’ 31 Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models Paper β’ 2312.06585 β’ Published Dec 11, 2023 β’ 29
Synthetic Data (Almost) from Scratch: Generalized Instruction Tuning for Language Models Paper β’ 2402.13064 β’ Published Feb 20, 2024 β’ 48
Textbooks Are All You Need II: phi-1.5 technical report Paper β’ 2309.05463 β’ Published Sep 11, 2023 β’ 87
DataDreamer: A Tool for Synthetic Data Generation and Reproducible LLM Workflows Paper β’ 2402.10379 β’ Published Feb 16, 2024 β’ 31
Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models Paper β’ 2312.06585 β’ Published Dec 11, 2023 β’ 29