Paper Review: DINOv3
25 August 2025
Meta's self-supervised vision model trained on 17 billion images, introducing Gram anchoring to prevent feature degradation at scale and achieving state-of-the-art across diverse vision tasks without fine-tuning.
Blogposts
Data science, career and other topics
Browse by category:
My experience of searching for a job in 2024 as an MLE
14 August 2025
My experience of searching for a job in 2024 as an MLE across the globe
Paper Review: Group Sequence Policy Optimization
04 August 2025
My review of the paper Group Sequence Policy Optimization
Paper Review: Subliminal Learning: Language models transmit behavioral traits via hidden signals in data
28 July 2025
My review of the paper Subliminal Learning Language models transmit behavioral traits via hidden signals in data
Paper Review: ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models
30 June 2025
My review of the paper ProRL Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models
Paper Review: V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning
23 June 2025
A self-supervised video model trained on 1M+ hours of video that understands motion, anticipates actions, and — with just 62 hours of robot data — performs zero-shot robotic pick-and-place planning.