✨ Lessons of a first time founder ✨

I wrote down a bunch of important lessons Daniel Smilkov and I learned over the last year building Lilac, joining Databricks!.

We had a short ride, but we learned a ton!

This blog is mostly for technical folks who know how to build product, and are trying to…

endoftext is live on Product Hunt 😸

Alex Cabrera and I have been building for the past ~2 months and would love to read your comments.

producthunt.com/posts/endoftext

We are happy to announce our survey and taxonomy on text to image synthesis evaluation!
There are interesting differences between SOTA metrics and their capabilities of measuring text-conditioned image quality.
Check it out at: huggingface.co/papers/2403.11…

You can use the prompts you've optimized with endoftext.app directly in your projects. Simply use our API endpoint.

If you export your chat history from ChatGPT, you get the system prompt(s) for free, no jailbreaking or similar needed

Excited to share a tool Alex and I have been building!
<| 𝗲𝗻𝗱𝗼𝗳𝘁𝗲𝘅𝘁 |> is an AI-powered prompt editor that helps you write better prompts with suggested edits, automatic rewrites, and test case generation.

We'd love to get your feedback: endoftext.app

Kasper Hornbæk and I have a fully funded 3-year PhD position open in #HCI and #XR in Copenhagen DIKU - Department of Computer Science, UCPH & Pioneer Centre for AI. The call DL is March 31: candidate.hr-manager.net/ApplicationIni…! Please reach out to me or Kasper with any questions 🙂

The recording of this session with Alex is up!
Watch here: youtu.be/z8GSTYqvTPQ

We discussed:
- OSS AI Evaluation
- Look at your data
- Quantify behaviors
- Tell stories

Enjoy!

ML friends! I will be on a panel about the role of visualization in an AI-driven world and would be really interested in YOUR view. It would be really cool if you could fill out this little questionnaire until Jan 30. Also, feel free to share the link.

forms.gle/aHRfpn6aYmjBPT…

There is still time to register for their webinar this Friday, January 26th, 1:30 - 2:30pm PST. Registration info in the thread below.

Language Technologies Institute | @CarnegieMellon Zeno Alex Alex Cabrera

In the 30th session of #MultimodalWeekly , we welcome Alex, a researcher at CMU Human-Computer Interaction Institute.

Check out this really cool project for your next paper, email, or marketing copy.

(Tweet writing improved by Wordflow)

in the end, the closest AI analogy to the 'source' in 'open-source' is still the training data and a bit of recipe. not the few hundred lines of pytorch code (generic), nor the trained weights (a compiled executable). pushes the threshold quite high in today's AI world though

Optimism is the only reasonable strategy, for everything in life.

If you're interested in the hottest evaluation projects, give this a read!

Lots of predictions of synthetic data for AI being big this year. Decided to look at the OG Alpaca dataset:

hub.zenoml.com/project/f192ed…

Impressive for being GPT-4 generated w/ 1 prompt, but begs the question of how to generate diverse, OOD data

Hi!
Following some suggestions and comments, I updated the 2023 LLMs year review with more resources!
huggingface.co/blog/2023-in-l…

Researchers from Carnegie Mellon University, BerriAI explore the translation capabilities of Google’s Gemini and suggest Gemini Pro could be a valuable tool for MT.
Syeda Nahida Akter Zichun Yu Aashiq Muhamed Tianyue Alex Alex Cabrera Krrish Chenyan Xiong
slator.com/is-google-gemi…