Tiago (@Tiago477) header
Tiago (@Tiago477) profile picture
Joined March 2011
Photo by Tiago (@Tiago477)Photo by Tiago (@Tiago477)Photo by Tiago (@Tiago477)Photo by Tiago (@Tiago477)Photo by Tiago (@Tiago477)Photo by Tiago (@Tiago477)Photo by Tiago (@Tiago477)Photo by Tiago (@Tiago477)Photo by Tiago (@Tiago477)Photo by Tiago (@Tiago477)
🚨: IBM released the World’s Smallest Movie by moving 65 atoms frame by frame, back in 2013. The cast: 65 atoms. The camera: A 2-ton Scanning Tunneling Microscope. The temperature: -268°C.
121
864
56
12,297
1,184,131
GPT-5 just casually did new mathematics. Sebastien Bubeck gave it an open problem from convex optimization, something humans had only partially solved. GPT-5-Pro sat down, reasoned for 17 minutes, and produced a correct proof improving the known bound from 1/L all the way to 1.5/L. This wasn’t in the paper. It wasn’t online. It wasn’t memorized. It was new math. Verified by Bubeck himself. Humans later closed the gap at 1.75/L, but GPT-5 independently advanced the frontier. A machine just contributed original research-level mathematics. If you’re not completely stunned by this, you’re not paying attention. We’ve officially entered the era where AI isn’t just learning math, it’s creating it. @sama @OpenAI @kevinweil @gdb @markchen90
Claim: gpt-5-pro can prove new interesting mathematics. Proof: I took a convex optimization paper with a clean open problem in it and asked gpt-5-pro to work on it. It proved a better bound than what is in the paper, and I checked the proof it's correct. Details below.
Readers added context they thought people might want to know
Unlike the commenter's claims, the 1.75/L paper was released before ChatGPT came up with the 1.5/L proof. While no paper for the 1.5/L bound existed, no human would've attempted to write a paper on it, because it is strictly worse than the existing 1.75/L bound. x.com/sebastienbubec… arxiv.org/pdf/2503.10138…
924
3,208
785
24,207
4,908,375
A new theoretical model suggests that the combination of gravity and quantum fluctuations alone may explain the origin of structure in the early universe, potentially eliminating the need for an inflaton field. This minimalist approach proposes that the interaction between quantum ripples and gravity could have triggered the rapid expansion and seeded the cosmic structures we observe today. With fewer assumptions and strong predictive power, the model opens the door to clearer tests through future observations, such as gravitational wave data, which could confirm or disprove its validity. 👉 share.google/PvoxhunoRybve6q…
25
64
5
335
10,813
Absolutely stunning performance!✨❤️
114
1,827
143
11,192
531,356
WE'RE PREMIER LEAGUE CHAMPIONS! 🏆
5,650
92,866
14,066
332,449
20,021,487
🇨🇳China cures paralysis!!! In 4, yes 4 patients, Chinese doctors have completed minimally invasive “brain-spinal interface” surgery allowing completely paralyzed patients to regain motor control over their bodies. #China Source: scmp.com/news/china/science/…
85
807
128
3,375
132,686
DESI has made the largest 3D map of our Universe to date Earth is at the center in this animation, each blue point is a galaxy: > 14 million galaxies in total up to 11 billion light years away The gaps are the regions where our own galaxy blocks the view
63
506
67
2,767
155,159
A 17-year-old student built a mind-controlled robotic arm using AI and 3D printing for just $300.
223
790
166
6,183
728,687
So beautiful. 👏
235
1,206
125
6,608
297,658
This little lady plays like an angel. She relies only on her ears, as she’s blind.. Her talent is breathtaking and deserves endless retweets 👍
668
5,399
312
24,242
422,761
Musical art 📹Animation by 73boxofficial
238
2,588
198
13,614
2,394,364
The man who healed what doctors can't: Carl Jung. Studying his ideas will make you immune to depression. These are his 6 Laws for Finding your Life's Purpose:
126
1,613
112
8,576
1,639,064
Tiago retweeted
#OnThisDay twelve years ago, two experiments at CERN announced the existence of a particle with characteristics consistent with those of the #HiggsBoson. Named after the late theoretical physicist Peter Higgs, its discovery crowned the #StandardModel of particle physics. In 2013, the @NobelPrize for Physics was awarded to François Englert and Peter Higgs for their 1964 proposal of the Brout–Englert–Higgs mechanism. To mark 60 years since Peter Higgs published his now famous paper predicting the existence of the boson named after him, here we see 60 event displays from @ATLASexperiment and @CMSExperiment, showing collisions where a Higgs boson is produced and then decays into various particles. #FunFact: in the audio, you can hear Peter Higgs speak about the discovery in his own words. Find out more about the #Higgsboson home.cern/science/physics/hi…
87
185
19
572
38,058
Sterile neutrinos are hypothetical particles that represent an intriguing extension to the Standard Model of particle physics. Unlike the three known types of neutrinos, electron, muon, and tau neutrinos, sterile neutrinos do not interact via the weak nuclear force. This lack of interaction with ordinary matter makes them incredibly difficult to detect but also positions them as potential key players in some of the unresolved puzzles of modern physics. The concept of sterile neutrinos emerged as physicists grappled with the perplexing nature of neutrino masses and oscillations. Neutrinos are known to oscillate between their three flavors, a phenomenon that implies they have mass, contrary to the original postulations of the Standard Model. However, the exact mechanism by which neutrinos acquire mass remains elusive. Introducing sterile neutrinos offers a possible explanation. They could mix with the active neutrino flavors, providing a source of mass through their interactions in a broader theoretical framework. Sterile neutrinos could also shed light on the observed discrepancies in neutrino experiments. Certain anomalies in the results of experiments like the Liquid Scintillator Neutrino Detector (LSND) (bnl.gov/nns/docs/presentatio…) and MiniBooNE at @Fermilab suggest the presence of an additional type of neutrino beyond the three known flavors. These anomalies could be indicative of sterile neutrinos, detectable indirectly through their impact on the oscillation patterns of active neutrinos. The search for sterile neutrinos involves a range of experimental approaches. One such effort is the Short-Baseline Neutrino (SBN) program at Fermilab, which aims to conclusively determine whether the anomalies seen in past experiments can be attributed to sterile neutrinos. By examining neutrino oscillations over short distances with high precision, experiments like SBN could either uncover evidence for sterile neutrinos or rule out their existence within certain parameters. The discovery of sterile neutrinos could transform our understanding of the universe on the largest scales. In the early universe, sterile neutrinos could have influenced the formation of large-scale structures by altering the dynamics of dark matter and dark energy. They might also play a role in the observed matter-antimatter asymmetry in the universe, contributing to a deeper understanding of why the cosmos is dominated by matter rather than an equal mix of matter and antimatter. Astrophysical observations could also benefit from the discovery of sterile neutrinos. For example, sterile neutrinos might be linked to the phenomenon of fast radio bursts (FRBs), extremely energetic pulses of radio waves whose origins remain mysterious. Some theories suggest that sterile neutrinos could be produced in the environments around neutron stars or black holes, potentially offering an explanation for these enigmatic cosmic events. In addition to their potential cosmological impact, sterile neutrinos are considered candidates for dark matter, the mysterious substance that makes up about 27% of the universe's mass-energy content but does not emit, absorb, or reflect light, making it invisible and detectable only via its gravitational effects. Further theoretical work has proposed that sterile neutrinos could interact through previously unknown forces or gauge bosons, leading to phenomena such as self-interacting dark matter. This could help reconcile various astrophysical observations with dark matter models, providing a richer framework for understanding both neutrino physics and cosmology. Despite their hypothetical status, sterile neutrinos embody a crucial frontier in the quest to expand the Standard Model and unravel the deeper mysteries of the universe. Their discovery would mark a significant breakthrough in our understanding of particle physics, providing a new dimension to the complex behavior of neutrinos and offering insights into the fundamental nature of matter and the cosmos. As experiments continue and technology advances, the pursuit of sterile neutrinos remains a vibrant and dynamic field, promising to yield profound discoveries about the building blocks of our universe. 👉 arxiv.org/pdf/2106.05913 👉 symmetrymagazine.org/article…
4
36
1
132
5,356
Anything that potentially challenges the Standard Model is a very interesting experiment. Wild that Germanium-76 takes 100 trillion times the age of the universe to decay, yet it is observable in a year in you have ~1 ton of it! newscenter.lbl.gov/2012/05/1…
4,380
5,752
677
47,578
22,719,812
Tiago retweeted
CERN pays tribute to Peter Higgs Peter Higgs’ scientific legacy will extend far beyond the scope of current discoveries. The Higgs boson is linked to some of the most intriguing and crucial outstanding questions in fundamental physics. This still quite mysterious particle therefore represents a uniquely promising pathway to physics beyond the Standard Model of particle physics.
34
622
58
2,314
116,880
Q* Leaked info: Source: An unspecified PasteBin.(L-I-N-K in next post) Can't confirm the authenticity as it's from unknown source but you can have a look. Q* is a dialog system conceptualized by OpenAI, designed to enhance the traditional dialog generation approach through the implementation of an energy-based model (EBM). Distinct from the prevalent autoregressive token prediction methods, Q* aims to mimic a form of internal deliberation akin to human thought processes during complex problem-solving, such as chess playing, where a deeper analysis of potential moves leads to better decision-making compared to rapid, less considered responses. This model shifts focus towards the inference of latent variables, reminiscent of constructs in probabilistic models and graphical models, fundamentally altering how dialog systems operate. Energy-Based Model for Dialog Generation At the core of Q* is the EBM, which operates by assessing the compatibility of an answer to a given prompt through a scalar output. This output signifies the "energy" of the response, where a lower value indicates a high compatibility (a better answer) and a higher value suggests low compatibility (a poor answer). This mechanism allows Q* to evaluate potential responses holistically, moving beyond the sequential prediction of tokens to understand the underlying relevance and appropriateness of an answer to the prompt. Optimization in Abstract Representation Space The innovation in Q* lies in its optimization process, conducted not within the space of possible text strings but in an abstract representation space. Here, thoughts or ideas are represented in a form that allows for the computational minimization of the EBM's scalar output, akin to finding the path of least resistance in a landscape. This process involves gradient descent, a method for finding the minimum of a function, applied to iteratively refine these abstract representations towards those that yield the lowest energy in relation to the prompt. From Abstract Thought to Textual Response Once an optimal abstract representation — one that minimizes the EBM's output — is identified, Q* employs an autoregressive decoder to transform this abstract thought into a coherent textual response. This step bridges the gap between the non-linguistic, conceptual understanding of the dialog system and the linguistic output required for human interaction. Training the System The EBM within Q* is trained using pairs of prompts and responses, adjusting the system's parameters to minimize the energy for compatible pairs while ensuring that incompatible pairs result in higher energy levels. This training process can incorporate contrastive methods, where the system learns to differentiate between compatible and incompatible pairs, and non-contrastive methods, which involve regularization techniques to control the distribution of low-energy responses across the space of all possible answers. Implications for Dialog Systems Q*'s approach, leveraging EBMs for dialog generation, represents a significant departure from traditional language modeling techniques. By optimizing over an abstract representation space and utilizing gradient-based inference, Q* introduces a more efficient, reasoned, and potentially more powerful method for generating dialog responses. This system not only promises improvements in the quality of generated text but also offers a blueprint for future advancements in AI's ability to engage in human-like reasoning and conversational interactions. Technical Considerations Q*'s effectiveness hinges on the intricacies of its EBM, the optimization landscape it navigates, and the accuracy of its abstract representations. The model's capacity to simulate deep reasoning, akin to human deliberation, sets a new benchmark for dialog systems. Furthermore, the method of training Q*—balancing the need for specificity in correct responses while avoiding the collapse of energy levels across diverse inputs—poses unique challenges and opportunities for AI research.
108
176
83
954
708,673
Could AI therapists have a major positive effect on the future? Imagine if everyone on Earth had free, 24/7 access to an endlessly knowledgeable psychologist who had worked with millions of patients and had a superhuman understanding of human psychological patterns.
452
167
134
2,379
642,021
Beautiful world.
12
528
81
3,081
112,205
River Nile from space....😯
29
432
19
3,500
89,033