Researchers — page 3

Co-authored the Codex evaluation paper: an early anchor for code LLM capability measurement.

Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.

Co-authored Llama 2: Open Foundation and Fine-Tuned Chat Models.

Co-authored “The Llama 3 Herd of Models”.

Co-authored Flamingo: an influential multimodal model for few-shot vision-language tasks.

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.

Co-authored Gemma 2: improving open language models at a practical size.

Co-authored Gemma 2: improving open language models at a practical size.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored the sparsely-gated MoE layer paper: a foundational conditional-computation design.

One of the earlier Anthropic contributors worth tracking if you care about the transition from RLHF-style assistant training into scaling and evaluation work.

Co-authored MMLU: a widely used benchmark for general LLM capability across many subjects.

Co-authored PaLI: a key reference for scaling multilingual vision-language models.

Co-authored Llama 2: Open Foundation and Fine-Tuned Chat Models.

Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.

A high-signal researcher for grounded language and retrieval-heavy systems, especially if you want to understand how language models stay useful as the world changes around them.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored the Phi-3 Technical Report (capable models designed for smaller footprints).

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored the Gemma 3 Technical Report.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored “The Llama 3 Herd of Models”.

Useful to follow if you care about the practical evaluation layer of open models, especially where benchmark tooling and reproducible comparisons actually shape what the ecosystem measures.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored “The Llama 3 Herd of Models”.

Co-authored the Gemma 3 Technical Report.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemma: open models based on Gemini research and technology.

A high-signal person to follow for the part of alignment research that asks whether a model’s stated reasoning can actually be trusted and measured.

A strong person to follow for the point where machine learning research starts shaping the compute stack itself, especially in chip placement and systems-aware optimization.

Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.

Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.

Co-authored “The Llama 3 Herd of Models”.

Co-authored “The Llama 3 Herd of Models”.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored PaLM: Scaling Language Modeling with Pathways.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Llama 2: Open Foundation and Fine-Tuned Chat Models.

Co-authored Gemma 2: improving open language models at a practical size.

Co-authored the Hugging Face Transformers paper that helped standardize modern NLP workflows.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Flamingo: an influential multimodal model for few-shot vision-language tasks.

Co-authored Mixtral of Experts: a key MoE reference in the open-weights frontier.

Co-authored the Gemma 3 Technical Report.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemma 2: improving open language models at a practical size.

Co-authored Gemma: open models based on Gemini research and technology.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored “The Llama 3 Herd of Models”.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored “The Llama 3 Herd of Models”.

Co-authored “The Llama 3 Herd of Models”.

An important open-model researcher for understanding how early public LLM efforts, scaling heuristics, and open data work fed into the broader modern model ecosystem.

Co-authored the Phi-3 Technical Report (capable models designed for smaller footprints).

Co-authored “The Llama 3 Herd of Models”.

Co-authored “The Llama 3 Herd of Models”.

A useful person to follow for the bridge between reinforcement-learning instincts and later alignment methods like DPO, especially where preference optimization is treated as a core learning problem rather than a bolt-on finetuning trick.

Important because he helped define how people think about language-model decoding quality, and his work keeps showing up where practical generation behavior matters more than benchmark theater.

Co-authored GPT-3: Language Models are Few-Shot Learners.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored the Phi-3 Technical Report (capable models designed for smaller footprints).

Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.

Co-authored StarCoder: a foundational open code model effort (BigCode).

Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.

Co-authored “The Llama 3 Herd of Models”.

Co-authored OLMo: Accelerating the Science of Language Models.

A strong bridge figure between the older fastText and self-supervision era and the newer open-weight LLaMA wave at Meta.

Co-authored StarCoder: a foundational open code model effort (BigCode).

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Llama 2: Open Foundation and Fine-Tuned Chat Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored AlphaZero: a canonical reference for self-play + search in RL.

Co-authored “The Llama 3 Herd of Models”.

One of the clearest people to track if you want to understand how frontier open-weight labs balance model quality, deployment speed, and product ambition.

Co-authored QLoRA: made high-quality fine-tuning feasible on modest hardware.

Co-authored Code Llama: a key open-model reference for code generation and coding assistants.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored “The Llama 3 Herd of Models”.

Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.

Co-authored GPT-3: Language Models are Few-Shot Learners.

Co-authored ARC: an influential reasoning benchmark for question answering.

Co-authored the Gemma 3 Technical Report.

Co-authored PaLI: a key reference for scaling multilingual vision-language models.

A foundational figure in modern sequence modeling whose work on the Transformer changed the technical direction of language and multimodal systems.

Co-authored Gato: a key reference for generalist, multi-task agents.

Co-authored “The Llama 3 Herd of Models”.

Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.

Co-authored the GPT-4 Technical Report: a key reference for the GPT-4-era frontier.

Co-authored “The Llama 3 Herd of Models”.

Co-authored Reflexion: a practical pattern for improving agents via self-critique and memory.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored “The Llama 3 Herd of Models”.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Co-authored “The Llama 3 Herd of Models”.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Worth following because he brings a real theory background into the model-systems layer, especially where structured linear algebra and sequence methods end up mattering for practical modern architectures.

Useful because he connects an earlier line of conversational-AI work at Nextremer with later authorship on both the original RWKV paper and Eagle/Finch, which makes this page more than a stray coauthor stub.

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.

A useful profile for the research layer behind DeepMind’s large-model program, especially across the line from Gopher and Chinchilla into Gemini.

Co-authored Gemini: A Family of Highly Capable Multimodal Models.

Useful to follow for the scaling and productization layer of the LLaMA line, especially as it moved from the first paper into the broader Llama 3 release wave.