[Editor’s Note: This class was a part of the 2019 DFL Jane Street Fellowship.] This guide would not have been possible without the help and feedback from many people. Special thanks to Yasaman Bahri for her feedback, support, and mentoring. Thank you to Kumar Krishna Agrawal, Sam Schoenholz, and Jeffrey Pennington for their valuable input and guidance. Finally, thank you to our group members Chris Akers, Brian Friedenberg, Sajel Shah, Vincent Su, Witold Szejgis, for their curiosity, commitment […]

Ver mais

:::info Astounding Stories of Super-Science February, 2026, by Astounding Stories is part of HackerNoon’s Book Blog Post series. You can jump to any chapter in this book here. The Moors and the Fens, volume 1 (of 3) – Chapter X: Dead Men’s Shoes Astounding Stories of Super-Science February 2026: The Moors and the Fens, volume 1 (of 3) – Chapter X Dead Men’s Shoes By J. H. Riddell ::: There are some persons, living in agreeable town or country […]

Ver mais

Been digging into the LLaDA2.1 paper (arXiv:2602.08676) and ran some comparisons that I think are worth discussing. The core claim is that discrete diffusion language models can now compete with AR models on quality while offering substantially higher throughput. The numbers are interesting but the tradeoffs are more nuanced than the headline results suggest. The paper introduces a T2T (Token to Token) editing mechanism on top of the standard M2T (Mask to Token) scheme, controlled by dual thresholds […]

Ver mais

arXiv:2601.05431v1 Announce Type: new Abstract: Accurately assessing the potential for fault slip is essential in many subsurface operations. Conventional model-based history matching methods, which entail the generation of posterior geomodels calibrated to observed data, can be challenging to apply in coupled flow-geomechanics problems with faults. In this work, we implement a variational autoencoder (VAE)-based data-space inversion (DSI) framework to predict pressure, stress and strain fields, and fault slip tendency, in CO${_2}$ storage projects. The main computations required by […]

Ver mais

The ‘uncanny valley’ is the final frontier for generative video. We have seen AI avatars that can talk, but they often lack the soul of human interaction. They suffer from stiff movements and a lack of emotional context. Tavus aims to fix this with the launch of Phoenix-4, a new generative AI model designed for the Conversational Video Interface (CVI). Phoenix-4 represents a shift from static video generation to dynamic, real-time human rendering. It is not just about […]

Ver mais

Partial differential equations (PDEs) are a central tool for modeling the dynamics of physical, engineering, and materials systems, but high-fidelity simulations are often computationally expensive. At the same time, many scientific applications can be viewed as the evolution of spatially distributed fields, making data-driven forecasting of such fields a core task in scientific machine learning. In this work we study autoregressive deep-learning surrogates for two-dimensional PDE dynamics on periodic domains, focusing on generalization to out-of-distribution initial conditions within […]

Ver mais

arXiv:2603.00629v1 Announce Type: new Abstract: Large time series models (LTMs) have emerged as powerful tools for universal forecasting, yet they often struggle with the inherent diversity and nonstationarity of real-world time series data, leading to an unsatisfactory trade-off between forecasting accuracy and generalization. Rather than continually finetuning new LTM instances for each domain, we propose a data-centric framework, time-series adaptive transformation optimization (TATO), that enables a single frozen pre-trained LTM to adapt to diverse downstream domains through an […]

Ver mais

Built a text-only baseline: trained a Random Forest on ~90,000 resolved Polymarket questions (YES/NO). Features: TF-IDF (word ngrams, optional char ngrams) + a few cheap flags (date/number/%/currency, election/macro/M&A keywords). Result: ~80% accuracy on 15.000 held-out data/questions (plus decent Brier/logloss after calibration). Liked the idea played a bit more with differnt data sets and did some cross validation with Kalshi data and saw similar results. Now having this running with paper money and competing with stat of the art […]

Ver mais

Robotic imitation learning typically requires models that capture multimodal action distributions while operating at real-time control rates and accommodating multiple sensing modalities. Although recent generative approaches such as diffusion models, flow matching, and Implicit Maximum Likelihood Estimation (IMLE) have achieved promising results, they often satisfy only a subset of these requirements. To address this, we introduce PRISM, a single-pass policy based on a batch-global rejection-sampling variant of IMLE. PRISM couples a temporal multisensory encoder (integrating RGB, depth, tactile, […]

Ver mais