Language Model

On October 31, OpenAI announced that it is open-sourcing a new benchmark called SimpleQA, which measures the ability of language models to answer short fact-seeking questions, in order to measure the accuracy of language models. One of the open challenges in AI is how to train models to generate factually correct answers. Current language models sometimes produce incorrect output or unsubstantiated answers, a problem known as "hallucinations". Language models that can generate more accurate and less illusory answers are more reliable and can be used...

Recently, Arcee AI officially launched its latest open source language model, Arcee-Nova. This new model, based on Qwen2-72B-Instruct, quickly emerged in the open source field and became one of the best performing models. After evaluation, Arcee-Nova's performance almost reached the level of GPT-4 in May 2023, which not only marks an important milestone for Arcee AI, but also brings new hope to the entire AI community. Arcee-Nova is a complex model that...

AI2's newly released Open Language Model (OLMo) framework aims to promote research and experimentation with large-scale language models. By providing training code, models, and evaluation code on Hugging Face and GitHub, AI2 is committed to enabling academia and researchers to jointly study the science of language models, explore the impact of new pre-training data subsets on downstream performance, and study new pre-training methods and stability. The first batch of models in the project include four 7B scale final variants corresponding to different architectures, optimizers, and training hardware, and a 1B scale model, all of which are...