musedivision

ml research

[musedivision @musedivision]

@dhh Seems like correlation, even tho I love the idea

[musedivision @musedivision]

@ylecun @geoffreyhinton Lol it’s just because you think LLMs won’t lead to a AGI breakthrough.

You have not been vindicated on that opinion.

[Simon Willison @simonw]

Being a computer scientist who refuses to find anything about LLMs interesting right now is a bit like being a geneticist who refuses to find anything interesting about the recently opened Jurassic Park

[musedivision @musedivision]

I always thought that as a ML Researcher we would be safe from the advance of AI.

In the past it was said that jobs like truck drivers jobs would be taken first by AGI like systems…

But all fields related to code are being eaten and changed so fast now

[Michael Black @Michael_J_Black]

What is the role of academic computer vision research in the age of increasingly powerful large models? Is GPT-6 Astra a step change? How can a researcher have an impact today in academia?

These are the questions I ask myself as I head off to ECCV 2026, a conference I’ve attended since 1992. One of my papers this year is VIGA, a method that takes an image as input and outputs a 3D Blender scene that represents that image. This is a classical inverse-graphics task and VIGA was the first method to solve it using an agentic approach.

The idea is now several years old and the first version of the paper was rejected. This delayed publication significantly. After it was accepted at ECCV, it was quickly surpassed by people using Claude Code for the same purpose. Today GPT-6 Astra blows away all previous results. But we still head off to ECCV to tell the community about our invention that is now fully out of date.

The way academic work often progresses is that one reads recent papers, notices that they have limitations, comes up with a new idea, explores this, publishes it, etc. Any published paper I read today is based on ideas that are at least a year old. And those ideas were based on the literature of the time, which was also a year old. That means that any paper I see at ECCV is likely two years out of date. In AI today, two years means your work is likely irrelevant.

At CVPR this summer I noticed that many authors have not gotten the message. They continue to work on “old” problems that have a long history. This history is based on assumptions about how the “vision problem” will be “solved”. The truth is that it is being solved in a very different way and many of these problems are no longer relevant. Another group of papers focuses on very niche problems where large models likely fail because of insufficient data or lack of business interest. The impactful papers were largely from industry and had long author lists and massive data+compute behind them. These papers were also out of data, describing systems that had been released months before, but at least they served to provide the community with more complete documentation and analysis of commercial systems.

So what should academics do? First, we need to put aside the tools we’ve used for years and start from scratch. Every project should start by trying really hard to solve the problem with existing tools. I would like to see every paper begin with a detailed experimental analysis of how existing models perform and why they fail (if they do). This gives the kind of insight we need today. Then, assuming current models fail, the solution should provide some fundamental insight that will outlive the next release of such models.

Reviewers today still focus on technical novelty. This pushes people to focus on tweaking architectures rather than clearly moving the field forward. Papers need to be judged based on their novel insight and not their novel technical contribution. This is a real shift in thinking but it focuses us on what matters - progress of the field.

If we want there to be a “field” of computer vision, then it can’t become a marginal backwater, focusing on esoteric problems. If you haven’t tried using Astra (or whatever comes next) to solve your problem, then you have not done your homework. This omission should be seen as negatively as not having a previous work section.

Concretely, I think papers should include a new section analogous to “Related Work” where that related work is current models and how they perform on the task. Reviewers should start asking for this and expecting authors to be able to articulate their insights about the limitations of existing large models.

I'm interested in your thoughts.

[musedivision @musedivision]

@LewisCTech @tsoding @DanielW_Kiwi I can hear the difference but non New Zealanders can’t hear it at all

[musedivision @musedivision]

@thinkingwest The west isn’t Christianity

[musedivision @musedivision]

@ericweinstein But they’ll never put you on the billboard anyways because your theory is too anti establishment 😢

[musedivision @musedivision]

@cmuratori I mean pcs don’t use cd anymore either …

[musedivision @musedivision]

@Y40IFRQTTING Bro sandwich shop, they good

[musedivision @musedivision]

Intelligence is a perpetual motion machine

[musedivision @musedivision]

@dylan522p Sounds like a skill issue

[musedivision @musedivision]

@jukan05 Perfect days

[musedivision @musedivision]

It’s a refreshing take from @dhh

It does feel qualitatively different to use agents to generate code these last 6 months.

In ML research it feels like magic, all the arch, training or inference hypotheses you had can be tested so fast.

It’s so fun

https://t.co/WSd1QLv29S

[musedivision @musedivision]

@tekbog Bad taste bro, lots of game devs worked on hard this had had to go back to the office full time because of the first leak now it’s all for nothing

[musedivision @musedivision]

This feels bit like early 2021 image gen, where the imperfections made it feel actually artistic

[Surya @kickingkeys]

You can just RL a coding model to paint with javascript btw

[musedivision @musedivision]

@paulg Or they are kiwi, Aussie or british and it’s used to downplay things

[musedivision @musedivision]

@ksw_arman Ouch haha

[Dabs🩸 @DabsMalone]

every company eventually becomes 11 people doing actual work and 64 people asking for updates

[musedivision @musedivision]

I can’t stand this gushing over open models from Chinese labs like they represent freedom ideals…

It’s a competitive strategy when compute poor.

Decades of Chinese industrial espionage stealing American tech forgotten. I think china has been more consistent in that position

[Naval @naval]

Frontier Labs are incented to be fearful as it provides a moral justification for staying closed.

Chinese Labs take the consistent position of doing what’s good for all and being open.

So the little guys find themselves in the uncomfortable position of rooting for the Chinese…