Adriano D'Alessandro

What is it about Marxist philosophers that makes them active into their 90s? Fredric Jameson was the same way. I swear to God, the modern biohackers just need to study Marx if they want to live forever.

I think there's two things happening.

1. AI hype looks (and is) a lot like the crypto hype cycle (huge energy footprint, lots of GPUs, negative use cases)

2. People don't want to admit how powerful these systems have gotten. A lot of academics grew their audiences on criticizing these systems.

Perhaps, a better example is intimate deepfakes. It's so pervasive, now, that it's essentially impossible to stop. The researchers who created the technology cannot do anything to mitigate the downstream harm. It's simply out of their hands.

I only turn to Oppenheimer because it's an example of a scientist responsible for creating a consequential technology losing authority over it. I'm moreso just pointing out that we, as researchers, can't guarantee anything about how our work gets used if there are existing incentives to misuse it.

person die from a denial of care. And there is no technical solution to this. There are only political ones. But we're playing wack-a-mole if we try to mitigate every bad outcome by criminalizing it. The real problem is that our society is optimizing for capital accumulation.

to mitigate the bad at a systemic level when the person "pressing the button" gets an enormous benefit and has distance from the person facing the consequences. A better example is the US healthcare system and algorithmic denial of coverage. The person who makes the money does not have to witness a

When we ask "how might we use this technology for good, and mitigate the bad?", it assumes this is possible in our current system. The reason I used the analogy is because we exist in a complex economic system with many interests, where cause and effect are far separated. It becomes very challenging

And this is just one single task. I just don't see radiologists being replaced in the short term. We've also been claiming radiologists will be replaced every year for like a decade.

I'm working on a project with some radiologists. We're trying to do early rheumatoid arthritis detection in wrist radiographs. It's so damn hard to get any model to understand the very subtle variation in bone density loss and joint space narrowing in noisy radiographs when you have limited data.

My filesystem at any given point:

2503.02821v2.pdf
2503.02821v2 (1).pdf
...
2503.02821v2 (6).pdf

I would argue that the images in crowd counting datasets are high entropy! You have a variety of scales, ethnicities, ages, roles, actions, etc. (an Asian woman soldier is very semantically different from a South American boy playing soccer)

It's the labels that are low entropy (i.e. just person).

question of doing good or bad with AI isn't one we ever get to make. No more than Oppenheimer got to decide how the bomb was used. Our society is built around capital accumulation and the ones with all the capital will press the capital accumulation button until they are the only ones left

There's this old moral question: "If you could press a button that would give you a bunch of money, but it would cause someone you don’t know in a distant part of the world to die, would you do it?".

The problem with AI is that we are not the ones who get to decide if the button is pressed. The

🚨 New paper out!
"VisualOverload: Probing Visual Understanding of VLMs in Really Dense Scenes"
👉 arxiv.org/abs/2509.25339
We test 37 VLMs on 2,700+ VQA questions about dense scenes.
Findings: even top models fumble badly—<20% on the hardest split and key failure modes in counting, OCR & consistency.

I've seen a few papers now evaluating counting performance in VLMs using novel datasets with a counting split. I'm curious why nobody uses crowd counting and few-shot datasets like JHU++, SHA, FSC147, REC8K, etc.

Sean "Spaghetti" Orr

How does AI interact with culture?

We’re thrilled to have Dr. Maria Antoniak (@mariaa.bsky.social) join us at FGVC! Her interdisciplinary work on AI in the humanities brings a new perspective to our workshop: the fine-grained categorization of the intangible parts of culture. See you at #CVPR2026!

Too much competition, which drivers down innovation. Everyone takes small incremental steps rather than chasing neat but risky ideas.

🚨 Updated Deadline Alert!

We have extended the deadline for the Proceedings from Feb 27th to Mar 3rd.

#CVPR #CV #AI

😒 real happy for ya

diversity with FLUX is using image-to-image translation as an augmentation strategy. If you want diverse images of "cows", just generate images for random scenes, and use them as a structural prior. Here's an image from a paper I'm working on where we're generating structural hard negatives:

especially in fine-grained settings. So there is a trade-off between the diversity of categories you can represent and the diversity for a single category.

2. They don't seem to investigate whether something as trivial as augmentation can solve these problems.

3. One hack I've used to get more

A few things pop into mind when reading the results.

1. While recent models have lower distributional diversity in their output for a single category, they do also correctly generate significantly more categories. The FLUX models can correctly depict substantially more categories than SD1.5,

I started noticing that more recent models did not exhibit this property, despite the outputs being more aesthetic. I figured there was a trade-off that was being made during training to prioritize object fidelity and identity over other properties in the prompt.

I had a paper at ECCV that investigated whether you could get object counting data out of text-to-image models by prompting them with "An image of {N} oranges" or similar. The output for early LDMs would always be a bit wrong, but if you averaged the count in enough of them, you would get N. But,

🚨 Reminder: The Proceedings Track deadline is February 27.

Don't miss out on this chance to share your research 😁 We're very excited to see what you've been working on!

#CVPR2026 #AI #ML

someone had maybe passed my paper around to them.

It's for sure collusion, but I don't know who they are. This wasn't during CVPR, though. In the most recent example, they were indirect, but made it clear they had read my recently submitted paper (they mentioned things that were only in the submitted version and not on Arxiv). But it sounded like

We are excited to announce that we are co-hosting the AnimalCLEF25 challenge with LifeCLEF at #CVPR2026! 🐾

Individual animal re-identification is a major challenge in conservation. Help us build evidence-based conservation tools that work!

Find out more: www.kaggle.com/competitions...

reviewing my paper or know who is, and they are probing to see what papers I'm reviewing.

I'm rambling now, but all of this is to say, I've been working on a different paper for ECCV which is hopefully a bit more bulletproof. It sucks to abandon a paper, but alas.