Co-authored the Gemma 3 Technical Report.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Pathways-scale language modeling (PaLM)
Co-authored PaLM: Scaling Language Modeling with Pathways.
Open-weight chat and foundation models (Llama 2)
Co-authored Llama 2: Open Foundation and Fine-Tuned Chat Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.
Open-model frontier reports (DeepSeek-V3)
Co-authored the DeepSeek-V3 Technical Report.
Measuring real-world coding ability (SWE-bench)
Co-authored SWE-bench: a key benchmark for whether models can resolve real GitHub issues.
Open-weight LLMs (Qwen2)
Co-authored the Qwen2 Technical Report.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Open, fully-documented language models (OLMo)
Co-authored OLMo: Accelerating the Science of Language Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Frontier model development (GPT-4)
Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Open multimodal models (Gemma 3)
Co-authored the Gemma 3 Technical Report.
Open language models (Gemma 2)
Co-authored Gemma 2: improving open language models at a practical size.
Open language models (Gemma 2)
Co-authored Gemma 2: improving open language models at a practical size.
Open multimodal models (Gemma 3)
Co-authored the Gemma 3 Technical Report.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Generalist agents (Gato)
Co-authored Gato: a key reference for generalist, multi-task agents.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Large-scale training, systems
A high-signal figure for understanding how DeepMind turned ambitious research systems into durable products, especially across reinforcement learning, speech, and code generation.
RWKV and efficient sequence modeling
A useful long-tail page because he is present on both the original RWKV paper and Eagle/Finch, then shows up again on multilingual embedding evaluation work, which makes him more than a one-paper launch contributor.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
RWKV and efficient sequence modeling
A strong page to keep because it connects the original RWKV paper to a later, much clearer research identity in graph representation learning, latent-space geometry, and multi-task adaptation.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Bidirectional transformer pretraining (BERT)
Co-authored BERT: a turning point for transfer learning in NLP.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Sparsely-gated mixture-of-experts
Co-authored the sparsely-gated MoE layer paper: a foundational conditional-computation design.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Open code models (CodeGemma)
Co-authored CodeGemma: open code models based on Gemma.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Co-authored the DeepSeek-V3 Technical Report.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Co-authored “The Llama 3 Herd of Models”.
Efficient MoE scaling (GLaM)
Co-authored GLaM: an influential MoE scaling reference in large language modeling.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Frontier model development (GPT-4)
Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Frontier model development (GPT-4)
Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Co-authored OLMo: Accelerating the Science of Language Models.
Open-source LLMs (EleutherAI)
Worth tracking if you care about the seam between open-model benchmarking and the harder question of what frontier systems should actually be evaluated for.
Open, fully-documented language models (OLMo)
Co-authored OLMo: Accelerating the Science of Language Models.
Self-rewarding post-training
Co-authored Self-Rewarding Language Models: explores self-improvement via internal reward modeling.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Co-authored Imagen: a milestone for photorealistic text-to-image diffusion models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Frontier model development (GPT-4)
Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.
Model-written evaluations for LM behavior
Co-authored model-written evals: a practical technique for discovering and measuring LM behaviors.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Open-source LLMs (EleutherAI)
A good person to follow for the part of evaluation work that goes beyond leaderboard scores and asks how models generalize across cultures, languages, and shifting social context.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Small, capable models (Phi-3)
Co-authored the Phi-3 Technical Report (capable models designed for smaller footprints).
Open language models from Google (Gemma)
Co-authored Gemma: open models based on Gemini research and technology.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Promptable segmentation foundation models (SAM)
Co-authored Segment Anything.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Holistic evaluation of language models (HELM)
Co-authored HELM: a framework for evaluating language models across many axes beyond raw accuracy.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Open language models (Gemma 2)
Co-authored Gemma 2: improving open language models at a practical size.
Frontier model development (GPT-4)
Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.
Open-source LLMs (EleutherAI)
One of the quieter but still important contributors in the open-data and open-evaluation lineage behind The Pile, GPT-NeoX, and later benchmarking infrastructure.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Compute-optimal scaling for LLM training
Important for the DeepMind large-model lineage because his work sits inside the sequence from compute-optimal scaling into Gemini rather than only the headline launch moment.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Co-authored the DeepSeek-V3 Technical Report.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Open code LLMs (StarCoder)
Co-authored StarCoder: a foundational open code model effort (BigCode).
Open-model frontier reports (DeepSeek-V3)
Co-authored the DeepSeek-V3 Technical Report.
Co-authored “The Llama 3 Herd of Models”.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Co-authored the DeepSeek-V3 Technical Report.
Co-authored “The Llama 3 Herd of Models”.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Mixture-of-experts LLMs
Co-authored Mixtral of Experts: a key MoE reference in the open-weights frontier.
Open language models (Gemma 2)
Co-authored Gemma 2: improving open language models at a practical size.
Frontier model development (GPT-4)
Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.
Open-source LLMs (EleutherAI)
Worth tracking for the open-model side of the field, especially where dataset construction, practical training work, and alignment-flavored thinking meet.
Open language models from Google (Gemma)
Co-authored Gemma: open models based on Gemini research and technology.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Fully Sharded Data Parallel training (FSDP)
Co-authored PyTorch FSDP: practical lessons for scaling fully-sharded training workloads.
Open language models (Gemma 2)
Co-authored Gemma 2: improving open language models at a practical size.
Small, capable models (Phi-3)
Co-authored the Phi-3 Technical Report (capable models designed for smaller footprints).
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Co-authored the DeepSeek-V3 Technical Report.
Small, capable models (Phi-3)
Co-authored the Phi-3 Technical Report (capable models designed for smaller footprints).
Pathways-scale language modeling (PaLM)
Co-authored PaLM: Scaling Language Modeling with Pathways.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.
Alignment via AI feedback (Constitutional AI)
A good page for the societal-impacts side of Anthropic’s research, especially where evaluation work turns toward persuasion and real-world downstream effects.
Fully Sharded Data Parallel training (FSDP)
Co-authored PyTorch FSDP: practical lessons for scaling fully-sharded training workloads.
Co-authored “The Llama 3 Herd of Models”.
Open-model frontier reports (DeepSeek-V3)
Co-authored the DeepSeek-V3 Technical Report.
Open-weight frontier models (Llama 3)
Co-authored “The Llama 3 Herd of Models”.
Fast, cheap LLM serving (PagedAttention)
Co-authored vLLM: a widely used serving stack for efficient LLM inference.
Multimodal frontier models (Gemini)
Co-authored Gemini: A Family of Highly Capable Multimodal Models.