PinnedAI Papers AcademyDINOv2 from Meta AI — Finally a Foundational Model in Computer VisionDINOv2 is a computer vision model from Meta AI that claims to finally provide a foundational model in computer vision, closing some of the…Aug 15, 2023Aug 15, 2023
AI Papers AcademyMixture of Nested Experts | AI Paper ExplainedIn this post we dive into Mixture of Nested Experts, a new method presented by Google that can dramatically reduce AI computational costAug 12Aug 12
AI Papers AcademyIntroduction to Mixture-of-Experts (MoE)In this post we go back to the original Google’s paper which presented the Mixture-of-Experts layer, or MoE in shortJul 17Jul 17
AI Papers AcademyMixture-of-Agents (MoA): Can open-source LLMs unite to win GPT-4o?In this post we explain the Mixture-of-Agents method, which shows a way to unite open-source LLMs to win GPT-4o on AlpacaEval 2.0Jun 12Jun 12
AI Papers AcademyArithmetic Transformers with Abacus Positional EmbeddingsIn this post we dive into Abacus Embeddings, which dramatically enhance the arithmetic capabilities of Transformers!Jun 1Jun 1
AI Papers AcademyCLLMs: Consistency Large Language Models | AI Paper ExplainedIn this post we dive into Consistency Large Language Models (CLLMs), a new family of models which can dramatically speedup LLMs inference!May 24May 24
AI Papers AcademyReFT: Representation Finetuning for Language ModelsIn this post we dive into ReFT: Representation Finetuning for Language Models, which let us finetune LLMs with 10–50x less params than LoRAApr 15Apr 15
AI Papers AcademyStealing Part of a Production Language Model | AI Paper ExplainedWhat if we could discover OpenAI models internal weights? In this post we dive into a paper which presents an attack that steals LLMs dataApr 4Apr 4
AI Papers AcademyHow Meta AI ‘s Human-Like V-JEPA Works?In this post, we dive into V-JEPA paper, another step by Meta AI towards Yann LeCun’s vision of a more human-like AI, this time for videos.Mar 18Mar 18
AI Papers AcademyThe Era of 1-bit LLMs by Microsoft | Paper ExplainedIn this post we dive into the era of 1-bit LLMs paper by Microsoft, which shows a promising direction for low cost large language modelsMar 21Mar 21