PinnedPublished inTowards Data ScienceA long-term Data Science roadmap which WON’T help you become an expert in only several monthsFrom time to time I’m asked: how does one become Data scientist? What courses are necessary? How long will it take? How did you become DS…Dec 2, 201819Dec 2, 201819
Paper Review: Reverse Thinking Makes LLMs Stronger ReasonersTraining student LLM by making teacher LLM generate questions, forward reasoning, backward questions, and backward reasoning4d ago4d ago
Paper Review: Project Sid: Many-agent simulations toward AI civilizationMany-agent simulations for collaborationNov 25Nov 25
Paper Review: Large Language Models Orchestrating Structured Reasoning Achieve Kaggle Grandmaster…LLM performing at Kaggle Grandmaster Level! Or maybe not.Nov 11Nov 11
Paper Review: Unbounded: A Generative Infinite Game of Character Life SimulationAn AI-driven “infinite game” where real-time language and vision models power open-ended character simulations with high consistency and…Oct 29Oct 29
Paper Review: Contextual Document EmbeddingsDocument embedding: using context and getting better performance!Oct 21Oct 21
Paper Review: Differential TransformerSubtracting two attention maps significantly increases transformer performance!Oct 14Oct 14
Paper Review: Depth Pro: Sharp Monocular Metric Depth in Less Than a SecondA new SOTA in CV: Monocular Metric Depth in 0.3 seconds!Oct 7Oct 7
Book Review: Unlocking Data with Generative AI and RAGLearn how to build RAGs using LLM, evaluate them and improve each component!Sep 29Sep 29
Paper Review: Training Language Models to Self-Correct via Reinforcement LearningMulti-turn Reinforcement Learning improving self-correction without multiple models or STF!Sep 23Sep 23