What the smart people are recommending. 8874 books, tools, and products endorsed by the thinkers absorb.md tracks. Ranked by how many times each has been recommended across compiled podcasts, papers, posts, and tweets.
“We introduce Perturb-and-Correct (P&C), a post-hoc method for constructing epistemically diverse predictors from a single pretrained network.”
“We introduce Personalized RewardBench, a novel benchmark designed to rigorously assess reward models' capacity to model personalized preferences.”
“We present Basis Selection with Importance (BSI), a principled low-rank compression framework that ranks and prunes bases by directly estimating the expected loss increase incurred when each basis is …”
“_Blog post by Yann LeCun_”
“# Sampling diverse near-optimal solutions via algorithmic quantum annealing”
“Here, we introduce a quantum-inspired family of nonlocal Nonequilibrium Monte Carlo (NMC) algorithms by developing an adaptive gradient-free strategy that can efficiently learn key instance-wise geome…”
“author of the Hill algorithm for linear systems”
“he's actually the author of the first paper on, on, on barren plateaus.”
“there are some, you know, we've recently studied quantum convolutional neural networks, and it looks like this might be an interesting useful architecture in the NISQ era”
“at one point i read alan turing's 1950 paper called computing machinery and intelligence which is the turing test paper”
“and if that sounds interesting to you you may be interested in the abstraction and reasoning corpus which is a kind of intelligence test that's meant to be used by humans and machines that i released …”
“Here we give classical sampling algorithms with better total variation distance and Kullback-Leibler divergence than these experiments and a computational cost quadratic in the number of modes.”
“We observe that the comment of [1, arXiv:2302.07897] is consistent with [2] on key points: i) the microscopic mechanism of the experimentally observed teleportation is size winding and ii) the system …”
“# On the duality between contrastive and non-contrastive self-supervised learning _Blog post by Yann LeCun_”
“In this paper, we develop two metrics for critically examining this assumption: Causal Importance of Reasoning (CIR)... and Sufficiency of Reasoning (SR)...”
“we're introducing the Open Molecules 2025 data set and Meta's universal model for atoms...By making open molecules and universal model available, we're enabling researchers to drive innovation”
“Yann LeCun on a vision to make AI systems learn and reason like animals and humans”
“We formalize this setting as reinforcement learning with rich feedback and introduce Self-Distillation Policy Optimization (SDPO)”
“I encourage you to explore our full blog post for more details and together let's push the boundaries of AI research to solve the big scientific questions about human and machine intelligence.”
“Formation of robust bound states of interacting microwave photons”
“# Noise-resilient Edge Modes on a Chain of Superconducting Qubits _ArXiv paper co-authored by Hartmut Neven_”
“Active Self-Supervised Learning: A Few Low-Cost Relationships Are All You Need”
“just last week the Transformers team wrote a really, really cool blog post about MoEs and how they work. So if anybody watching is interested, that's a great way to get into it.”
“# Augmented Language Models: a Survey _Blog post by Yann LeCun_”
“# The SSL Interplay: Augmentations, Inductive Bias, and Generalization”
“Best of AI Curation Vol. 3, covers: No-gradient architecture, LLM tool making and mastery (3 papers!), RLHF without RL, an uncensored LLM, an open course, and more. Deep dive with me: 🧵”
“5 Myths about Israel and the War in Gaza”
“Best of AI Curation Vol. 3, covers: No-gradient architecture, LLM tool making and mastery (3 papers!), RLHF without RL, an uncensored LLM, an open course, and more. Deep dive with me: 🧵”
“Best of AI Curation Vol. 3, covers: No-gradient architecture, LLM tool making and mastery (3 papers!), RLHF without RL, an uncensored LLM, an open course, and more. Deep dive with me: 🧵”
“Best of AI Curation Vol. 3, covers: No-gradient architecture, LLM tool making and mastery (3 papers!), RLHF without RL, an uncensored LLM, an open course, and more. Deep dive with me: 🧵”
“Facts & Values Clarifying the Moral Landscape”
“# The Sin of Moral Equivalence _Blog post by Sam Harris_”
“this is a a paper that Jake Browning who's a philosopher and and and and I published in the noima magazine which is a philosophy magazine about the fact that a system that is purely”
“# VICRegL: Self-Supervised Learning of Local Visual Features”
“We released a new error correcting code it's a new low density parity check code that we actually call the gross code This new code requires orders and magnitude fewer cubits than the surface code and…”
“I wrote about the argument that this week’s AI regulations represent ‘regulatory capture’ and make the case that if even Elon Musk is asking the government to do something, we have every right to mand…”
“I read the readings of the works of like Ray Kurt and felt like there was something there and that AI was really going to go somewhere”
“In this work, we propose CLERA, a unified model for Cognitive Load and Eye Region Analysis, which achieves precise keypoint detection and spatiotemporal tracking in a joint-learning framework.”
“# Revisiting Feature Prediction for Learning Visual Representations from Video”
“Here is a detailed explanation of the account takeover hack that happened on this account this week. https://avc.xyz/anatomy-of-a-twitterx-account-takeover-hack”
“# Visualizing Dynamics of Charges and Strings in (2+1)D Lattice Gauge Theories”
“In this work, we identify representation collapse in the model's intermediate layers as a key factor limiting their reasoning capabilities. To address this, we propose Sequential Variance-Covariance R…”
“In Ref. [1], Bravyi et al. found examples of Bivariate Bicycle (BB) codes with similar logical performance to the surface code but with an improved encoding rate.”
“# Observation of disorder-free localization using a (2+1)D lattice gauge theory on a quantum processor”
“in 1969 uh F fets from University of Washington published this beautiful paper called oper and conditioning of cortical unit activity where he was able to record a single unit neuron from a monkey and…”
“If you still think AGI is sci-fi or too far away to matter, consider checking out [https://www.safe.ai/work/statement-on-ai-risk](https://www.safe.ai/work/statement-on-ai-risk) and who disagree with y…”
“from a theoretical complexity point of view I actually find this great uh this paper by Scott Aronson actually established the hardness of this problem”
“this is a famous paper by Ed far look at optimization where you can come up with circuits that will be uh potentially useful for these problems”
“There's another Super popular paper that you author call koala cognitive architectures for language agents”
“Let's talk about your newest essay on uh the age of intelligence”
“which is used in arXiv:2511.09371 to establish a 6-functor formalism of complex analytic motivic homotopy theory and produce an analytification map that is compatible with the six operations”
“In this work, we propose Visual-Predictive Instruction Tuning (VPiT) - a simple and effective extension to visual instruction tuning that enables a pretrained LLM to quickly morph into an unified auto…”
“Okay, Scott Aronson has a beautiful sorry beautiful paper that can show random circuits are actually hard to simulate.”
“when people thought Trump was a clown and he was just a clown that's all he was so I wrote a blog post someone remember it it was called clown genius so I reframed him as somebody who was a showan and…”
“this term I heard come out of this paper Alpha codium it basically achieves state-of-the-art kind of like coding performance not necessarily through better models or better prompting strategies but th…”
“This paper presents METIS, the first RAG system that jointly schedules queries and adapts the key RAG configurations of each query, such as the number of retrieved text chunks and synthesis methods, i…”
“We introduce a Navigation World Model (NWM), a controllable video generation model that predicts future visual observations based on past observations and navigation actions.”
“In this paper, we investigate the quality of intermediate representations in various LLM architectures, including Transformers and State Space Models (SSMs).”
“You, Elon, Vivek, David Sacks, you guys are all acting from the highest motives, but I'm playing devil's advocate here. You're moving very fast, you're really moving fast. Those federal workers whose …”
“I'll refer folks to the Gemini 1.5 uh technical report which actually just got released slightly after IO it's like 150 pages”
“https://arxiv.org/abs/2410.03339”
“We present Video JEPA with Variance-Covariance Regularization (VJ-VCR): a joint-embedding predictive architecture for self-supervised video representation learning that employs variance and covariance…”
“you have seen this recent archive paper by Google tee on the demonstration of the äh bre äh of fundamental cor based on S code so they here they use äh to the area of super conducting Cub to implement…”
“Vanguard uh comes out with a report every year they just came out for with their 2023 report which has the 2022 data and this is a this is something I follow as soon as it comes out I look it up”
“I think people should actually read the kind of management discussion section that are in these S1s... it can look quite dense, but it's worth reading.”
“you may have seen when you published an excerpt in The Atlantic, I just surfaced it for readers because I thought it was great.”
“And then there's that 17,000 word profile of Sam Altman by Ronan Farrow. Yeah, I mean, a fascinating piece of journalism for sure.”
“whose complexity is asymptotically quadratically faster than its classical counterpart in [Wang & Ling, IEEE Trans. Inf. Theory 2019]”
“we derive two versions of quantum dual attacks that improve upon the previous ones in [Pouly & Shen, EUROCRYPT 2024]”
“# RAGServe: Fast Quality-Aware RAG Systems with Configuration Adaptation _Blog post by Ravi Netravali_ https://arxiv.org/abs/2412.10543”
“To this end, we present DINO World Model (DINO-WM), a new method to model visual dynamics without reconstructing the visual world.”
“Our quantum Discrete Gaussian sampler can also be used to speed up the algorithm for solving the Short Integer Solution problem, in any norm, of [Bollauf, Pouly & Shen, ePrint 2026/225].”
“You recently turned 30 and you made a wonderful blog post about the last 10 years of being in the space and uh some of the things that you personally cared about mm uh what were kind of would you mind…”
“so you know one of the other posts that I published a few months ago is you know making ethereum Cipher Punk again and one of the themes I have there is this idea of like building this uh like basical…”
“So if you look, for example, there's a nice study by Otis at Berkeley and a bunch of other people where they did a controlled experiment in Kenya.”
“This motivated us to show how the Hamilton-Ostrogradski formalism can be applied it to the Pais-Uhlenbeck oscillator. We hope that the approach presented in this work can serve as a basis for discussi…”
“# X-Sample Contrastive Loss: Improving Contrastive Learning with Sample Similarity Graphs”
“because there's papers like this uh paper by Sergey Bravi where they show that they can actually show for what we call short depth circuits there's actually a separation between uh what classical and …”
“I like this paper. I'll give you some examples before I go a little bit more. This is a paper out of the Rican team where they look at uh time dynamics of a so on the top where you see the white lines…”
“This release represents IceCube's most sensitive and comprehensive publicly available all-sky muon track dataset to date and should be preferred over previous releases.”
“you know that there was this National Quantum Initiative act that was passed in the US in uh 2018 I think uh it was written by then Senator kamla Harris”
“Empirically, using only just three such steps improves sample quality over strong diagonal-covariance baselines, including OCM-DDPM, across standard image benchmarks.”
“# #OccupyDNC Communiqué No. 1”
“146 of the 24,576 features in the layer-8 residual-stream SAE release of Bloom (2024) clear a Holm-corrected significance threshold”
“I think it wasn't until 2017 when I first saw the results from GPT-1 that it clicked for me that language is probably the area in which we can do this.”
“The first one I think, you know, OpenAI was getting on board with. The second one in a way had always been part of OpenAI's messaging, but, you know, over many years of the time that I spent there, I …”
“To address this, we introduce RoboPEPP, a method that fuses information about the robot's physical model into the encoder using a masking-based self-supervised embedding-predictive architecture.”
“Purpose: To explore best-practice approaches for generating synthetic chest X-ray images and augmenting medical imaging datasets to optimize the performance of deep learning models in downstream tasks…”
“At a 16. Z had a great essay, on flock. We can really, really solve crime. And it's just a choice.”
“we'll be talking about your recent paper titled video as the new language for real world decisionmaking”
“I had another paper called learning interactive real world simulators”
“# Non-Abelian braiding of graph vertices in a superconducting processor”
“Agents as a Service”
“# V-JEPA: The next step toward advanced machine intelligence”
“_Blog post by Mustafa Suleyman_”
“MW is Short FTAI Aviation Ltd. (FTAI US) _Blog post by Muddy Waters_”
“This paper introduces the City Physiology: a new conceptual framework to facilitate the integration of city layers when designing holistic simulators.”
“there's an interesting paper that investigates this so you can uh you can check it out it's called Embers of autoaggression”
“there's an interesting paper that investigates how llms handle composition uh so it's titled the limits of Transformers of compositionality”
“a very cool example of how you can uh combine LMS with discret program search is this paper after this search IND resoning with the language model”