Jason Corso

Corso is a Professor at U Michigan and Co-Founder of Voxel51 who makes the category-defining data+model codevelopment ML Tool: FiftyOne

[Jason Corso @_JasonCorso_]

https://t.co/ZGJYlxItQC

[Jason Corso @_JasonCorso_]

Data is the cause of all of your pain.

[Jason Corso @_JasonCorso_]

@CSProfKGD @oier_mees Trust me bro. 🫥

[Jason Corso @_JasonCorso_]

In my field, all good technical conferences are double blind. I really don’t see the point. Perhaps you are in a different field that doesn’t have this practice which could be a shame. CVPR aaai ICLR NeurIPS icml iccv eccv….

But I’d like to think academic review is meritocratic

[Jason Corso @_JasonCorso_]

@CryptoNobler Probably bc trump just stole 1.8 billion from taxpayers.

[375ai @375ai_]

World models don't run on scraped text. They need continuous, multimodal observation of the physical world: labeled, time-stamped, geo-grounded.

Voxel51's @_JasonCorso_ writes for @FortuneMagazine, "AI models are choking on junk data."

[Jason Corso @_JasonCorso_]

The 15-minute high-level board update is a trap. Without the details, advisors just pattern-match onto other companies and give generic, parrot-like advice.

My take: Boardrooms need to operate like university labs. Block out 90+ mins for 1 hard topic. Detail matters.

TAF Series #18: https://t.co/f06E8tJGLz

[Jason Corso @_JasonCorso_]

@nazanin_ashrafi I prefer a walk in the woods

[Jason Corso @_JasonCorso_]

@CSProfKGD First thing: where’s the scissor
Second thing: I miss Blue Max.

[Jason Corso @_JasonCorso_]

@CSProfKGD Yes this is a key point.

But what I would like to see more of these days is actual discussion amongst reviewers, especially those in disagreement.

[Jason Corso @_JasonCorso_]

@CaimingXiong @Recursive_SI Excited to see how this evolves @CaimingXiong Curious tho what the domain is....no free lunch.

[Jason Corso @_JasonCorso_]

"Disagree and Commit" is the hardest shift for technical founders who are used to being "right." If you're still second-guessing a team decision 2 weeks later, you're the friction.

My take: Conviction > Certainty.

Read the full essay on navigating constructive confrontation: https://t.co/BSMRIlo1Kn

[Jason Corso @_JasonCorso_]

@taiyasaki @overleaf Let’s see, what’s the prompt? Hey Claude, help me reimplement an overleaf like web based collaborative editor for latex and make integration with you and other LLMs first class. Let me know if you have questions. I’ll be back in 48 hours.

[Jason Corso @_JasonCorso_]

@MedasanX In aggregate, why can't we wait? In individual cases, rtfp. All the source post does is rile up people with mostly meaningless data.

[Jason Corso @_JasonCorso_]

Show me the impact not the paper count. Paper mills outputting LPUs are a cancer to the field.

And the data seems incomplete. In aggregate what about all of the institutions less than paper threshold here.

[ℏεsam @Hesamation]

someone analyzed all 5000+ accepted papers at ICLR 2026, and it's a good signal who's pushing the research of AI:
> China has surpassed the US with 43.7% of the papers
> Europe's contribution is surprisingly small (5.3% including UK)

[Jason Corso @_JasonCorso_]

@mmbronstein Ain't that the truth.

[Jason Corso @_JasonCorso_]

@MedasanX No number of papers in top conferences is not a proxy for impact.

[Jason Corso @_JasonCorso_]

@MedasanX yes and……

[Jason Corso @_JasonCorso_]

@adamnemecek1 and github should kill stars and just replace them with the number of other projects that depend on a certain project. stars are a microeconomy. evil

[Jason Corso @_JasonCorso_]

@adamnemecek1 no god no