Foundation models Papers

A 2018 paper by Matthew E. Peters and five co-authors that represents each word with the internal states of a pre-trained deep bidirectional language model, so vectors change with context; adding them to existing models improved results on six NLP tasks.

OpenAI's 2018 paper introducing GPT, a Transformer decoder pre-trained to predict the next token and then fine-tuned for specific tasks. It improved the state of the art on 9 of 12 tasks studied and set the recipe the GPT series followed.

Google's 2018 BERT paper introduced an encoder-only Transformer that reads text in both directions and is pre-trained by hiding words. It set new results on eleven language tasks and made download-and-fine-tune the default workflow.

An NVIDIA paper from 2019 presenting an intra-layer model parallelism method, implemented with a few communication operations in native PyTorch, that trained transformer models of up to 8.3 billion parameters and improved several benchmarks.

A 2020 paper finding that language model loss falls predictably, as a power law, with model size, data and compute. It gave labs a quantitative reason to build much larger models and shaped the scale-up that followed.

The GPT-3 paper showed that a language model with 175 billion parameters could perform many tasks from a few examples in its prompt, with no fine-tuning, and that this ability grew with model size.

A report from Google's Gemini Team introducing the Gemini 1.5 family, multimodal models built around very long context, with Gemini 1.5 Pro recalling information across up to 10 million tokens.[^1]

Meta's 2024 paper describing Llama 3 as a family of foundation models — 8B, 70B and 405B parameters — trained on about 15 trillion tokens and reported to match leading models such as GPT-4 across a wide range of tasks.

A technical report from DeepSeek-AI on DeepSeek-V3, a mixture-of-experts language model with 671 billion total parameters, describing its architecture, training pipeline, and evaluations, including a full training run of 2.788 million H800 GPU hours.

The Gemini 2.5 technical report, submitted to arXiv on 7 July 2025, documents Google's Gemini 2.5 Pro and 2.5 Flash models: their reasoning, long-context video handling, and cost-versus-capability positioning.

A preview report on the DeepSeek-V4 series, describing a 1.6-trillion-parameter Pro model and a 284-billion-parameter Flash model, both with one-million-token context, plus the attention, connection and optimization changes that make that length affordable.

ULMFiT is a 2018 transfer-learning method by Jeremy Howard and Sebastian Ruder for fine-tuning language models across NLP tasks; it beat the state of the art on six text-classification tasks and performed well with few labels.

OpenAI's 2019 paper on GPT-2, a 1.5-billion-parameter Transformer trained on web text. It showed that a model trained only to predict the next word could perform several tasks when prompted, and its staged release sparked public debate.

CLIP is an OpenAI model trained on 400 million image-text pairs to map pictures and sentences into one shared space, so it can classify images zero-shot from a written description.

A DeepMind study showing that, for a fixed computing budget, language models should be trained on far more data than was then standard; its 70-billion-parameter Chinchilla beat larger contemporaries.

Meta's safety and preparedness report for Muse Spark, the model behind Meta AI, sets out the catastrophic-risk evaluations behind its launch and argues that, with mitigations applied, residual risk is acceptable.