Lex

This new image editing model from Black Forest Labs called **FLUX.1 Kontext** is really good. I ran some experiments on photos of Doggo, and couldn't believe how well it could maintain character consistent across multiple turns of editing.

https://notesbylex.com/absurdly-good-doggo-consistency-wit…

Learning to Reason without External Rewards (aka Self-confidence Is All You Need)

Turns out we can just use the LLM's internal sense of confidence as the reward signal to train a reasoning model, no reward model / ground-truth examples / self-play needed.

Amazing.

https://notesbylex.com/learning…

"My new hobby: watching AI slowly drive Microsoft employees insane" https://old.reddit.com/r/ExperiencedDevs/comments/1krttqo/my_new_hobby_watching_ai_slowly_drive_microsoft/

A cool approach to iteratively improving generated images, using o3 as an LLM-judge to generate targeted masks for improvements: https://simulate.trybezel.com/research/image_agent

If Trump was really looking out for Elon, he would have posted that on Twitter.

You're giving them way too much credit. Trump has always had really really bad ideas, but this time around there are no adults in government to shut them down, only sycophantic yes men.

All that pep talk just to make social media content.

Can you link me to it? Surprisingly hard to Google for.

Teaser figures in ACL template papers Teaser figures in ACL template papers. GitHub Gist: instantly share code, notes, and snippets.

ARR deadline is coming up! If you're wondering how to make a beautiful full-width teaser figure on your first page, above the abstract, in LaTeX, check out this gist I made showing how I do it!

gist.github.com/michaelsaxon...

🔥 allenai/Llama-3.1-Tulu-3-8B (trained with PPO) -> allenai/Llama-3.1-Tulu-3.1-8B (trained with GRPO)

We are happy to "quietly" release our latest GRPO-trained Tulu 3.1 model, which is considerably better in MATH and GSM8K!

"As a former tech lead at Meta for 6 years... I got 'meets all' or 'exceeds' every single half except the one in which I took parental leave."

www.reddit.com/r/business/c...

Won't someone please think of the child processes?

Media Watch has Chas derangement syndrome!

Really awesome work and big thank you for sharing it on Bluesky!

Example of injecting Wait token into the model generation.

A hilariously simple repro of OpenAI's test-time scaling paradigm called "Budget Scaling": end the thinking when your token budget is met, or append "Wait" to the model's generation to keep thinking, allowing the model to fix incorrect reasoning steps.

arxiv.org/abs/2501.19393

True.

Abstract and figures from paper R.I.P.: Better Models by Survival of the Fittest Prompts

A method for evaluating data for preference optimisation.

Rejecting Instruction Preferences (RIP) can filter prompts from existing training sets or make high-quality synthetic datasets. They see large performance gains across various benchmarks compared to unfiltered data.

arxiv.org/abs/2501.18578

GitHub - Jiayi-Pan/TinyZero: Clean, accessible reproduction of DeepSeek R1-Zero Clean, accessible reproduction of DeepSeek R1-Zero - Jiayi-Pan/TinyZero

A reproduction of Deepseek R1-Zero.

"The recipe:

We follow DeepSeek R1-Zero alg -- Given a base LM, prompts and ground-truth reward, we run RL.

We apply it to CountDown: a game where players combine numbers with basic arithmetic to reach a target number."

github.com/Jiayi-Pan/Ti...

Reasoning models can be useful for generating high-quality few-shot examples:

1. generate 10-20 examples from criteria in different styles with r1/o1/CoT, etc
2. have a model rate for each example based on quality + adherence.
3. filter/edit top examples by hand

Repeat for each category of output.

My dog Doggo, a stag hound X bull arab, chilling in the grass

Happy dog.

The Illustrated DeepSeek-R1

Spent the weekend reading the paper and sorting through the intuitions. Here's a visual guide and the main intuitions to understand the model and the process that created it.

newsletter.languagemodels.co/p/the-illust...

mlx-examples/llms/mlx_lm/models/deepseek_v3.py at main · ml-explore/mlx-examples Examples in the MLX framework. Contribute to ml-explore/mlx-examples development by creating an account on GitHub.

The DeepSeek V3 model file in ~450 lines of code in MLX LM.

github.com/ml-explore/m...

via awnihannun on Twitter.

So DeepSeek found a way to train a gpt4 quality model for *only* 6M worth of Nvidia hardware, and the market thinks this is bad for Nvidia?

Trump's Achilles heel is that he's a sucker for compliments and flattery. Just tell him how smart and good at business he is, and you'll get whatever you want from him.

You can see the billionaires leaning into that approach this time around, and it really seems to be working for them.

Imo the Google AI summaries are really helpful. The usefulness of an LLM increases substantially when it can reference its sources

It's a chart of OpenAI outages reported in last 24 hours by Down Detector (which is just user reports). In the last few minutes, there's a huge spike.

ChatGPT seems to be down.

Announcing The Stargate Project Announcing The Stargate Project

"The Stargate Project is a new company which intends to invest $500 billion over the next four years building new AI infrastructure for OpenAI in the United States"

$500 billion! For comparison, the 1960s Apollo project, when adjusted for inflation, cost around $250B.

openai.com/index/announ...

"Verified DeepSeek performance on ARC-AGI's Public Eval (400 tasks) + Semi-Private (100 tasks)

DeepSeek V3:
* Semi-Private: 7.3%
* Public Eval: 14%

DeepSeek Reasoner:
* Semi-Private: 15.8%
* Public Eval: 20.5%

Performance is on par, albeit slightly lower, than o1-preview"

x.com/arcprize/sta...

how_many_rs.md GitHub Gist: instantly share code, notes, and snippets.

This is gold. DeepSeek-R1's thought process for "how many 'r's in strawberry."

"So positions 3, 8, and 9 are Rs? No, that can't be right because the word is spelled as S-T-R-A-W-B-E-R-R-Y, which has two Rs at the end...

Wait, maybe I'm overcomplicating it...."

gist.github.com/IAmStoxe/1a1...