My name is Amir Saffari and this is my website and blog. I’ve a PhD in Computer Vision and Machine Learning and work at Alexa AI, Amazon as a Principal ML Scientist. Currently, I focus on generative AI, training large language models (LLMs), teaching LLMs to use thousands of tools and APIs to accomplish personalised and complex tasks in real-time, reasoning using weak supervision, and reinforcement learning based program synthesis.

Amir

Recent Papers

Complete list of papers

News

  • 2022, Mintaka dataset is now available, see our paper for more information.
  • 2021, Our two new papers at EMNLP extend differentiable knowledge graphs for complex QA, E2EQA and Rigel.

Recent Posts

CVPR 2012 Paper

So Sam and I got a paper on Efficient Online Structured Output Learning for Keypoint-Based Object Tracking accepted at CVPR 2012.
Read more →

Struck at ICCV 2011

Our paper, Struck: Structured Output Tracking with Kernels, got accepted at ICCV 2011. You can get more information about this project from Sam Hare’s website. {{youtube gnT34hJwdjM}} {{youtube jaI1m1vZEx8}}
Read more →

MIForest Source Code

So Christian finally cleaned up the code for our ECCV paper, the famous MILForest paper. So you can find it by following this link: MIForest
Read more →

ECCV 2010 Papers

We managed to get two papers accepted at ECCV, one being an ORAL: Christian Leistner, Amir Saffari, Horst Bischof, MILForests: Multiple-Instance Learning with Randomized Trees, Oral Presentation Amir Saffari, Christian Leistner, Martin Godec, Horst Bischof, Robust Multi-View Boosting with Priors. More to come soon.
Read more →