Instruction Fine-Tuning and In-Context Learning of Large Language Models with Symbols
Delve into advanced concepts of LLM instruction fine-tuning and in-context learning, exploring symbol tuning, FLAN-PaLM models, and cutting-edge research developments.
MotherNet: A Foundational Hypernetwork for Tabular Classification
Discover how MotherNet hypernetwork architecture revolutionizes tabular classification by generating optimized neural networks through in-context learning, offering faster training and competitive performance against XGBoost.
CMU Advanced NLP: How to Use Pre-Trained Models
Explore advanced techniques for leveraging pre-trained models in NLP, including fine-tuning, linear probing, and in-context learning. Gain insights into empirical observations and mental models for effective implementation.
Pretraining Task Diversity and the Emergence of Non-Bayesian In-Context Learning for Regression
Explores how task diversity in pretraining affects transformers' in-context learning abilities for regression, revealing a threshold beyond which models can solve novel tasks and outperform Bayesian estimators.
In-Context Learning: A Case Study of Simple Function Classes
Exploring in-context learning capabilities of AI models, focusing on Transformers and LSTMs for learning basic function classes, with insights on evaluation and architectural differences.
AI Mastery: Ultimate Crash Course in Prompt Engineering for Large Language Models
Master prompt engineering for AI models: craft effective prompts, explore techniques like n-shot and chain-of-thought, and learn practical tips to enhance communication with language models.
In-Context Learning in Deep Learning - How It Works in Large and Small Language Models
Explore in-context learning in LLMs: definition, significance, Bayesian framework, and implementation. Gain insights into this surprising capability of language models and its implications for AI.
Do Pretrained Transformers Learn In-Context by Gradient Descent?
Explore the relationship between In-Context Learning and Gradient Descent in large language models, examining theoretical connections and empirical evidence from pre-trained transformers.
Supercharging LLM Performance Without AI Training
Discover techniques to enhance large language model performance without additional training, focusing on innovative approaches and collaborative research findings.
Understanding Task Vectors in Vision-Language Models - Cross-Modal Representations
Explore how Vision-Language Models create task vectors - internal representations enabling cross-modal performance through text and images, revolutionizing AI's ability to understand and execute diverse tasks.
SETFIT Few-Shot Learning for SBERT Text Classification - Part 43
Discover how SETFIT's few-shot learning approach revolutionizes text classification, outperforming GPT-3 while using SBERT Sentence Transformers for efficient processing with limited training data.
Informative Example Selection for In-Context Learning
Explore efficient approaches for selecting informative examples in In-Context Learning, enhancing large language model performance without additional training.
Exploring Long Context Language Models - MLOps Reading Group August 2024
Explore the potential of Long-Context Language Models to replace traditional methods like retrieval, RAG, and SQL. Dive into recent research and engage in critical discussions about LCLMs' capabilities and limitations.
AutoFeedback - Scaling Human Feedback with Custom Evaluation Models
Discover how AutoFeedback combines human and model-based evaluation to improve LLM output assessment, reducing costs and enhancing efficiency while maintaining accuracy and transparency.
Learning Linear Models In-Context with Transformers
Explore in-context learning of linear models using transformers with UC Davis professor Spencer Frei in this insightful seminar.
Efficient Fine-tuning and Prompt Engineering - Lecture 20
Explore efficient fine-tuning techniques and prompt engineering strategies for optimizing large language models in machine learning applications.
Large Scale Universal Speech Generative Models
Explore cutting-edge speech generative models, including Voicebox, and their applications in multilingual TTS, style conversion, and content editing. Learn about scalable approaches for audio AI.
A Brief History of AI and AI Agents - From Language Models to Multi-Agent Systems
Explore the evolution of AI from GPT-3 to modern multi-agent systems, covering key developments in language models, function calling, AI agents, and future implications for global artificial intelligence.
Why Larger Language Models Struggle with Subjective Reasoning
Explore how larger language models handle subjective tasks, examining their limitations in emotion and morality recognition despite advanced prompting techniques and increased model sizes.
How Transformers Learn Causal Structure with Gradient Descent
Delve into the mathematical foundations of how transformer models learn and encode causal relationships through self-attention mechanisms and gradient-based training algorithms.