Antonio Tejero-de-Pablos
@toni-tiler.bsky.social
Research scientist in computer vision / Samurai / Rapper
Reposted by Antonio Tejero-de-Pablos
The Ilya Sutskever episode with Dwarkesh Patel is now available
www.youtube.com/watch?v=aR20...
www.youtube.com/watch?v=aR20...
November 25, 2025 at 7:45 PM
The Ilya Sutskever episode with Dwarkesh Patel is now available
www.youtube.com/watch?v=aR20...
www.youtube.com/watch?v=aR20...
Am I the only one who when vibe-codes feels like "the LLM whisperer"?
November 17, 2025 at 8:42 AM
Am I the only one who when vibe-codes feels like "the LLM whisperer"?
Reposted by Antonio Tejero-de-Pablos
Transfer learning and self-supervised learning have changed how we build neural architectures. Reusing representations from one task saves time and data, but tuning for a new problem is never plug-and-play. Always check what actually transfers. 🤔 #ItzikThoughtLoop
October 30, 2025 at 2:29 AM
Transfer learning and self-supervised learning have changed how we build neural architectures. Reusing representations from one task saves time and data, but tuning for a new problem is never plug-and-play. Always check what actually transfers. 🤔 #ItzikThoughtLoop
Reposted by Antonio Tejero-de-Pablos
Since it resonated with the audience, I’ll recap my main argument against AGI here. ‘General intelligence’ is like phlogiston, or the aether. It’s an outmoded scientific concept that does not refer to anything real. Any explanatory work it did can be done better by a richer scientific frame. 1/3
This was a truly heartening day, with deeply thoughtful challenges to the dominant narrative framed around AGI, coming from across disciplines and perspectives. Felt like the tide might finally be turning a bit, at least among the scientific community. Thanks @royalsociety.org!
1/2 I'm looking forward to taking part in a panel on AGI and the Turing Test, tomorrow afternoon (Thurs 2nd Oct) at the @royalsociety.org, w/ Dame Wendy Hall, Shannon Vallor, William Isaac, & Sir Nigel Shadbolt. royalsociety.org/science-even...
October 2, 2025 at 10:09 PM
Since it resonated with the audience, I’ll recap my main argument against AGI here. ‘General intelligence’ is like phlogiston, or the aether. It’s an outmoded scientific concept that does not refer to anything real. Any explanatory work it did can be done better by a richer scientific frame. 1/3
Reposted by Antonio Tejero-de-Pablos
How AI Taught Itself to See
Self-supervised learning is fascinating! How can AI learn from images only without labels?
In this video, we’ll build the method from first principles and uncover the key ideas behind CLIP, MAE, SimCLR, and DINO (v1–v3).
Video link: youtu.be/oGTasd3cliM
Self-supervised learning is fascinating! How can AI learn from images only without labels?
In this video, we’ll build the method from first principles and uncover the key ideas behind CLIP, MAE, SimCLR, and DINO (v1–v3).
Video link: youtu.be/oGTasd3cliM
How AI Taught Itself to See [DINOv3]
YouTube video by Jia-Bin Huang
youtu.be
September 16, 2025 at 11:13 PM
How AI Taught Itself to See
Self-supervised learning is fascinating! How can AI learn from images only without labels?
In this video, we’ll build the method from first principles and uncover the key ideas behind CLIP, MAE, SimCLR, and DINO (v1–v3).
Video link: youtu.be/oGTasd3cliM
Self-supervised learning is fascinating! How can AI learn from images only without labels?
In this video, we’ll build the method from first principles and uncover the key ideas behind CLIP, MAE, SimCLR, and DINO (v1–v3).
Video link: youtu.be/oGTasd3cliM
In a sort-of social experiment, instead of assigning "reviewers" to the submitted papers, we assigned "mentors". Although the job content was the same, the quality of the reviews and the satisfaction of the authors improved quite significantly (even those whose paper was rejected). Think about it.
September 3, 2025 at 7:18 AM
In a sort-of social experiment, instead of assigning "reviewers" to the submitted papers, we assigned "mentors". Although the job content was the same, the quality of the reviews and the satisfaction of the authors improved quite significantly (even those whose paper was rejected). Think about it.
Reposted by Antonio Tejero-de-Pablos
The system is completely broken.
blog.neurips.cc/2021/12/08/t...
Review results are random and affect the achievement record of thousands of researchers globally. If review quality cannot be controlled the responsibility would fall on ACs to recheck flagged papers, which doesn't seem feasible either
blog.neurips.cc/2021/12/08/t...
Review results are random and affect the achievement record of thousands of researchers globally. If review quality cannot be controlled the responsibility would fall on ACs to recheck flagged papers, which doesn't seem feasible either
July 4, 2025 at 7:13 AM
The system is completely broken.
blog.neurips.cc/2021/12/08/t...
Review results are random and affect the achievement record of thousands of researchers globally. If review quality cannot be controlled the responsibility would fall on ACs to recheck flagged papers, which doesn't seem feasible either
blog.neurips.cc/2021/12/08/t...
Review results are random and affect the achievement record of thousands of researchers globally. If review quality cannot be controlled the responsibility would fall on ACs to recheck flagged papers, which doesn't seem feasible either
Reposted by Antonio Tejero-de-Pablos
Reinforcement Learning Teachers of Test Time Scaling
arxiv.org/abs/2506.08388
We introduce a new way to teach LLMs how to reason by learning to teach, not solve! Here, a teacher model is rewarded based on how effectively its explanations help the student model recover correct solutions.
arxiv.org/abs/2506.08388
We introduce a new way to teach LLMs how to reason by learning to teach, not solve! Here, a teacher model is rewarded based on how effectively its explanations help the student model recover correct solutions.
Introducing Reinforcement-Learned Teachers (RLTs): Transforming how we teach LLMs to reason with reinforcement learning (RL).
Blog: sakana.ai/rlt
Paper: arxiv.org/abs/2506.08388
Code: github.com/SakanaAI/RLT
We introduce a new way to teach LLMs how to reason by learning to teach, not solve.
Blog: sakana.ai/rlt
Paper: arxiv.org/abs/2506.08388
Code: github.com/SakanaAI/RLT
We introduce a new way to teach LLMs how to reason by learning to teach, not solve.
June 23, 2025 at 2:37 AM
Reinforcement Learning Teachers of Test Time Scaling
arxiv.org/abs/2506.08388
We introduce a new way to teach LLMs how to reason by learning to teach, not solve! Here, a teacher model is rewarded based on how effectively its explanations help the student model recover correct solutions.
arxiv.org/abs/2506.08388
We introduce a new way to teach LLMs how to reason by learning to teach, not solve! Here, a teacher model is rewarded based on how effectively its explanations help the student model recover correct solutions.
Reposted by Antonio Tejero-de-Pablos
Guys … I’m all for critiquing the way LLMs are implemented in society and debunking folks who make wild claims that they’re sentient. But definitively claiming the brain doesn’t use statistical learning is not the right take. Don’t be armchair neuroscientists please. Seeing some wild opinions today.
June 19, 2025 at 7:47 PM
Guys … I’m all for critiquing the way LLMs are implemented in society and debunking folks who make wild claims that they’re sentient. But definitively claiming the brain doesn’t use statistical learning is not the right take. Don’t be armchair neuroscientists please. Seeing some wild opinions today.
As I see valuable work being rejected all the time, I tend to think that:
- Good reviewers feel like accepting the papers, and try to find enough reasons to do so
- Bad reviewers feel like rejecting the papers, and try to find enough reasons to do so
They may seem equivalent but oh they are not
- Good reviewers feel like accepting the papers, and try to find enough reasons to do so
- Bad reviewers feel like rejecting the papers, and try to find enough reasons to do so
They may seem equivalent but oh they are not
May 16, 2025 at 12:12 PM
As I see valuable work being rejected all the time, I tend to think that:
- Good reviewers feel like accepting the papers, and try to find enough reasons to do so
- Bad reviewers feel like rejecting the papers, and try to find enough reasons to do so
They may seem equivalent but oh they are not
- Good reviewers feel like accepting the papers, and try to find enough reasons to do so
- Bad reviewers feel like rejecting the papers, and try to find enough reasons to do so
They may seem equivalent but oh they are not
I wish reviewers would stop seeing ablation studies as a chance to reject a paper, emphasizing the impracticability of the proposed method since its accuracy is unstable when conditions vary... it's an ablation study you dummy, that's what's supposed to happen! 🤦♂️
#only_in_topcvconf #bully_reviewers
#only_in_topcvconf #bully_reviewers
May 12, 2025 at 4:20 AM
I wish reviewers would stop seeing ablation studies as a chance to reject a paper, emphasizing the impracticability of the proposed method since its accuracy is unstable when conditions vary... it's an ablation study you dummy, that's what's supposed to happen! 🤦♂️
#only_in_topcvconf #bully_reviewers
#only_in_topcvconf #bully_reviewers
Reposted by Antonio Tejero-de-Pablos
These boxes are not moving. A mind-bending optical illusion by Japanese artist Jagarikin.
May 7, 2025 at 1:03 AM
These boxes are not moving. A mind-bending optical illusion by Japanese artist Jagarikin.
Reposted by Antonio Tejero-de-Pablos
[Please share🔄]
Did you read blog posts / preprint about #ICLR2025 LLM experiment?
Were you in the #ICLR2025 review process?
As I was not in the process, I'm asking for opinions to improve review systems
Thank you for your help🙏
🤗 huggingface.co/blog/yoshito...
Did you read blog posts / preprint about #ICLR2025 LLM experiment?
Were you in the #ICLR2025 review process?
As I was not in the process, I'm asking for opinions to improve review systems
Thank you for your help🙏
🤗 huggingface.co/blog/yoshito...
Personal thoughts on a randomized study of LM-based review feedback agent at ICLR 2025
A Blog post by Yoshitomo Matsubara on Hugging Face
huggingface.co
April 22, 2025 at 3:59 PM
[Please share🔄]
Did you read blog posts / preprint about #ICLR2025 LLM experiment?
Were you in the #ICLR2025 review process?
As I was not in the process, I'm asking for opinions to improve review systems
Thank you for your help🙏
🤗 huggingface.co/blog/yoshito...
Did you read blog posts / preprint about #ICLR2025 LLM experiment?
Were you in the #ICLR2025 review process?
As I was not in the process, I'm asking for opinions to improve review systems
Thank you for your help🙏
🤗 huggingface.co/blog/yoshito...
Reposted by Antonio Tejero-de-Pablos
Verified my Bluesky account🦋
I was hesitant to do that, but TIL the following update
> Update as of December 12, 2024: If you change your default Bluesky username (with the .bsky.social suffix) to a website URL, your old .bsky.social username will be reserved for you.
bsky.social/about/blog/4...
I was hesitant to do that, but TIL the following update
> Update as of December 12, 2024: If you change your default Bluesky username (with the .bsky.social suffix) to a website URL, your old .bsky.social username will be reserved for you.
bsky.social/about/blog/4...
How to verify your Bluesky account - Bluesky
Here's how to verify your Bluesky account by setting your website as your username.
bsky.social
April 22, 2025 at 12:23 AM
Verified my Bluesky account🦋
I was hesitant to do that, but TIL the following update
> Update as of December 12, 2024: If you change your default Bluesky username (with the .bsky.social suffix) to a website URL, your old .bsky.social username will be reserved for you.
bsky.social/about/blog/4...
I was hesitant to do that, but TIL the following update
> Update as of December 12, 2024: If you change your default Bluesky username (with the .bsky.social suffix) to a website URL, your old .bsky.social username will be reserved for you.
bsky.social/about/blog/4...
In the multi-object setting, objects that appear earlier in the caption or that are larger in size are prioritized by the CLIP encoder.
arxiv.org/abs/2502.19842
arxiv.org/abs/2502.19842
April 20, 2025 at 2:38 PM
In the multi-object setting, objects that appear earlier in the caption or that are larger in size are prioritized by the CLIP encoder.
arxiv.org/abs/2502.19842
arxiv.org/abs/2502.19842
Reposted by Antonio Tejero-de-Pablos
I got to test the LLM "help" they provided and it was quite useful! It suggested being more specific on key areas and I thought I genuinely helped me improve the quality of my reviews, without changing my original assessment in any way. Imho it was very well tailored.
April 17, 2025 at 2:45 PM
I got to test the LLM "help" they provided and it was quite useful! It suggested being more specific on key areas and I thought I genuinely helped me improve the quality of my reviews, without changing my original assessment in any way. Imho it was very well tailored.
Reposted by Antonio Tejero-de-Pablos
Credibility, not capability.
The most important thing we build in technology and academia is not capability, but credibility. It does not matter how fast we calculate, how smart we are, or the number of products or papers we make, if we cannot answer "Why should anybody believe anything we say?"
The most important thing we build in technology and academia is not capability, but credibility. It does not matter how fast we calculate, how smart we are, or the number of products or papers we make, if we cannot answer "Why should anybody believe anything we say?"
April 13, 2025 at 2:21 PM
Credibility, not capability.
The most important thing we build in technology and academia is not capability, but credibility. It does not matter how fast we calculate, how smart we are, or the number of products or papers we make, if we cannot answer "Why should anybody believe anything we say?"
The most important thing we build in technology and academia is not capability, but credibility. It does not matter how fast we calculate, how smart we are, or the number of products or papers we make, if we cannot answer "Why should anybody believe anything we say?"
Reposted by Antonio Tejero-de-Pablos
🧠 Keeping LLMs factually up to date is a common motivation for knowledge editing.
But what would it actually take to support this in practice at the scale and speed the real world demands?
We explore this question and really push the limits of lifelong knowledge editing in the wild.
👇
But what would it actually take to support this in practice at the scale and speed the real world demands?
We explore this question and really push the limits of lifelong knowledge editing in the wild.
👇
April 8, 2025 at 3:32 PM
🧠 Keeping LLMs factually up to date is a common motivation for knowledge editing.
But what would it actually take to support this in practice at the scale and speed the real world demands?
We explore this question and really push the limits of lifelong knowledge editing in the wild.
👇
But what would it actually take to support this in practice at the scale and speed the real world demands?
We explore this question and really push the limits of lifelong knowledge editing in the wild.
👇
Reposted by Antonio Tejero-de-Pablos
Reposted by Antonio Tejero-de-Pablos
Visual Geometry Grounded Transformer (VGGT) predicts cameras, point maps, depth maps, and point tracks for up to hundreds of images in less than a second on a H100 GPU.
github.com/facebookrese...
github.com/facebookrese...
March 16, 2025 at 5:44 PM
Visual Geometry Grounded Transformer (VGGT) predicts cameras, point maps, depth maps, and point tracks for up to hundreds of images in less than a second on a H100 GPU.
github.com/facebookrese...
github.com/facebookrese...
I don't think I can go back to surveying papers without LLM summaries for quick filtering. But the fact that there's people who've never done a survey without an LLM kind of worries me. Like people who've never made a sum without a calculator or spoken a foreign language without a translation tool.
March 13, 2025 at 12:33 PM
I don't think I can go back to surveying papers without LLM summaries for quick filtering. But the fact that there's people who've never done a survey without an LLM kind of worries me. Like people who've never made a sum without a calculator or spoken a foreign language without a translation tool.
Reposted by Antonio Tejero-de-Pablos
Reposted by Antonio Tejero-de-Pablos
"our politics is split between a left that defends government even when it doesn't work, and a right that wants to destroy government even when it is working"
www.youtube.com/watch?v=Vwjx...
www.youtube.com/watch?v=Vwjx...
There Is a Liberal Answer to Elon Musk | The Ezra Klein Show
YouTube video by The Ezra Klein Show
www.youtube.com
March 10, 2025 at 5:00 AM
"our politics is split between a left that defends government even when it doesn't work, and a right that wants to destroy government even when it is working"
www.youtube.com/watch?v=Vwjx...
www.youtube.com/watch?v=Vwjx...
IMHO, AI doesn't do jack. People who want to profit from it do LOL Really, we need to stop using the lazy word AI and talk about specific people, technologies and business models. AI doesn't mean anything, I don't use it on my CV anymore and even stop listening when it pops up in a convo.
March 4, 2025 at 9:13 AM
IMHO, AI doesn't do jack. People who want to profit from it do LOL Really, we need to stop using the lazy word AI and talk about specific people, technologies and business models. AI doesn't mean anything, I don't use it on my CV anymore and even stop listening when it pops up in a convo.