ELMo vs GPT-4

Embeddings from Language Models (ELMo) is a groundbreaking language representation model that helps machines understand complex characteristics of word usage, including both syntax and semantics, and recognizes the variations of word use across different linguistic contexts. ELMo achieves this by utilizing word vectors that are learned as functions of the internal states of a pre-trained deep bidirectional language model (biLM). BiLM uniquely models both forward and backward language model likelihoods, and when adding ELMo to a task-specific model, the weights of the biLM are frozen. The ELMo vector is then concatenated with a baseline representation of tokens and fed into a task RNN, enhancing the model's performance by providing rich, context-aware word embeddings.

GPT-4 is the latest milestone in OpenAI’s effort in scaling up deep learning.

GPT-4 is a large multimodal model (accepting image and text inputs, emitting text outputs) that, while less capable than humans in many real-world scenarios, exhibits human-level performance on various professional and academic benchmarks. For example, it passes a simulated bar exam with a score around the top 10% of test takers; in contrast, GPT-3.5’s score was around the bottom 10%. We’ve spent 6 months iteratively aligning GPT-4 using lessons from our adversarial testing program as well as ChatGPT, resulting in our best-ever results (though far from perfect) on factuality, steerability, and refusing to go outside of guardrails.

GPT-4 is more creative and collaborative than ever before. It can generate, edit, and iterate with users on creative and technical writing tasks, such as composing songs, writing screenplays, or learning a user’s writing style.