I’m interested in NLP&IR, with the long-term goal of figuring out the perfect way to do information retrieval. You may know me from my frequent twitter rants about using late-interaction over dense vectors, or my love for encoders.
I do R&D at Mixedbread, where I contribute to building ever-better multimodal retrieval systems.
Please feel free to check out my About page if you’d like a very quick overview of stuff I’ve worked on!
You can also reach out if you’d like to chat!
I never really write online about personal things (except a few times, generally about my wife, who is awesome). Not that I don’t want to, I’m not particularly privacy-minded, but more that I never feel like it’s that interesting to share a ton about my pretty mundane day-to-day. There’s one thing that I do off-handedly publicly mention, maybe more than I realise: health. Generally in the context of “sorry about RAGatouille, I’ve had less time to spend on it than I thought because of health”, or some other similar phrasing....
Say hello to answerai-colbert-small-v1, a tiny ColBERT model that punches well above its weight.
Introducing JaColBERTv2.5🇯🇵, the new best Japanese retrieval model. Through this release, we present a thorough analysis to better understand what helps in training a good multi-vector retrieval model.
Blog post offering a quick overview of ColBERT and how it works, and introducing an efficient pooling trick to alleviate the issues it faces.
If you’re old school, you can find a raw HTML version of this post here This is a fairly long, stream-of-thought post about how I currently (Sunday Feb 4, 2024) feel about the broader ML/NLP/IR ecosystem and its future. Everything here’s on-the-fly opinion and can & will change. In summary: I think ML developments are incredibly exciting, and we need to continue to work on bridging the gap between ML-as-a-commodity-for-ML-practitioners to ML-as-a-commodity-for-everyone....