I was also looking for a video. The concept sounds good, but feels like I need to learn a lot of new commands, or have a cheat sheet next to me to be able to be able to use the framework.
Wondering how average users can benefit from this platform with Claude Code and the relation to Vending Bench that tracks how much money LLM's can make.
I can confirm. When trying convert simple Word sentences and tables to e.g. Markdown/HTML from a Word XML you need a PhD in XML edge cases and nested garbage.
I was amazed when I realised that Word docs were just zip files and you could poke around in the xml files embedded inside of them.
I almost implemented a working React -> Word document renderer back in 2017, but it didn't have support for creating the xml tags with : inside of them (which OOXML documents use).
Even though markitdown is a Microsoft project, it's just a thin wrapper around a bunch of 3rd party Python packages. For example, to go from docx to Markdown, it uses mammoth to convert docx to HTML[0], then uses markdownify to convert the HTML into Markdown[1].
Technically, they're a bit more than just zip files (they're OPC containers [0]), but if you're hand editing the file content it doesn't really matter.
Well, it is not pretty to see how the sausage gets made, but extracting formatted text from docx is absolutely doable, no PhD involved. Source: I have done it as a little sidequest because it was useful to audit a set of word documents.
Building a AI-assisted aviation regulation compliance tool for aviation professionals: https://aviation.bot
At the moment just a RAG with EU aviation regulations. Soon FAA, UK CAA, and more complex AI agent features that do more complex deep research.
Wondering how this compares to the Gemini (preview) embeddings as they seem to perform significantly better than OpenAI embeddings 3 large. I don't see any MTEB scores so hard to compare.
While we benchmarked internally, on BEIR, we opted not to report our model onto MTEB for the following reason:
1) MTEB has been gamed - if you look at this model (https://huggingface.co/voyageai/voyage-3-m-exp) on the MTEB leaderboard, its an intermediate checkpoint of Voyage-3-Large where they finetuned it on datasets that represent MTEB datasets.
2) If you look at the recent datasets in MMTEB, you'll find that it has quite a lot of machine translated or "weird" datasets that are quite noisy
In general, for our Search Models, we benchmark on these public academic datasets but we definitely do not try to hillclimb in this direction as we find it has little correlation with real use-cases
Recommendations? I'm seeking a robotics course akin to prompt engineering, LangChain, or LangGraph that prepares me for the 'ChatGPT' moment in robotics. While SLAM navigation and similar topics are vital, it feels reminiscent of five years ago when one needed to master machine learning to create their own ML models. Nowadays, most LLM models are API-based and don't require an ML background.
reply