arXiv:2602.07120v1 Announce Type: new Abstract: Modern language models (LMs) tend to memorize portions of their training data and emit verbatim spans. When the underlying sources are sensitive or copyright-protected, such reproduction raises issues of consent and compensation for creators and compliance risks for developers. We propose Anchored Decoding, a plug-and-play inference-time method for suppressing verbatim copying: it enables decoding from any risky LM trained on mixed-license data by keeping generation in bounded proximity to a permissively trained safe […]

Ver mais

Federated learning (FL) allows distributed clients to collaboratively train a global model in a privacy-preserving manner. However, one major challenge is domain skew, where clients’ data originating from diverse domains may hinder the aggregated global model from learning a consistent representation space, resulting in poor generalizable ability in multiple domains. In this paper, we argue that the domain skew is reflected in the domain-specific biased features of each client, causing the local model’s representations to collapse into a […]

Ver mais

Learning-to-Defer routes each input to the expert that minimizes expected cost, but it assumes that the information available to every expert is fixed at decision time. Many modern systems violate this assumption: after selecting an expert, one may also choose what additional information that expert should receive, such as retrieved documents, tool outputs, or escalation context. We study this problem and call it Learning-to-Defer with advice. We show that a broad family of natural separated surrogates, which learn […]

Ver mais

Get Started with Rust: Generics Generic programming allows programmers to write general algorithms that work with arbitrary types. It reduces code duplication and provides type safety, which enables us to write more concise and clean code. This approach is also known as parametric polymorphism or templates in other languages. Rust supports two types of generic code: Compile-time generics, similar to C++ templates. Run-time generics, similar to virtual functions in C++ and generics in Java. Serokell has a broad […]

Ver mais

arXiv:2601.14354v1 Announce Type: new Abstract: Joint Embedding Predictive Architectures (JEPA) offer a scalable paradigm for self-supervised learning by predicting latent representations rather than reconstructing high-entropy observations. However, existing formulations rely on textit{deterministic} regression objectives, which mask probabilistic semantics and limit its applicability in stochastic control. In this work, we introduce emph{Variational JEPA (VJEPA)}, a textit{probabilistic} generalization that learns a predictive distribution over future latent states via a variational objective. We show that VJEPA unifies representation learning with Predictive […]

Ver mais

if you’ve ever built an elegant, complex ML pipeline to solve something a 10-line prompt could’ve handled… this is for you. i’ve been thinking about what separates people who do useful research from people who do impressive-looking research. it’s almost always the problems you choose rather than raw technical skill. here’s the mental model i’ve landed on. every problem kind of follows these steps: find a clear problem people actually care about try the dumbest solution first. can […]

Ver mais

arXiv:2601.04327v1 Announce Type: new Abstract: Parallel programming is central to HPC and AI, but producing code that is correct and fast remains challenging, especially for OpenMP GPU offload, where data movement and tuning dominate. Autonomous coding agents can compile, test, and profile on target hardware, but outputs are brittle without domain scaffolding. We present ParaCodex, an HPC-engineer workflow that turns a Codex-based agent into an autonomous OpenMP GPU offload system using staged hotspot analysis, explicit data planning, correctness […]

Ver mais

LLMs are eating specialty skills. There will be less use of specialist front-end and back-end developers as the LLM-driving skills become more important than the details of platform usage. Will this lead to a greater recognition of the role of Expert Generalists? Or will the ability of LLMs to write lots of code mean they code around the silos rather than eliminating them? — Martin Fowler, tidbits from the Thoughtworks Future of Software Development Retreat, via HN) Tags: […]

Ver mais

The crypto market is no longer driven purely by hype. As investors look ahead to 2026, the focus has shifted toward timing, utility, and long-term sustainability. In this article, we take a closer look at Dogecoin (DOGE) and Mutuum Finance (MUTM)—two very different cryptocurrencies at very different stages—to assess which may offer the stronger investment opportunity in today’s market environment. Dogecoin (DOGE) Dogecoin remains one of the most recognizable cryptocurrencies in the market, largely due to its meme-driven […]

Ver mais