Mixture-of-experts variants of parameter-efficient fine-tuning enable per-token specialization, but they introduce additional trainable routers and expert parameters, increasing memory usage and training cost. This undermines the core goal of parameter-efficient fine-tuning. We propose Monkey Jump, a method that brings mixture-of-experts-style specialization to parameter-efficient fine-tuning without introducing extra trainable parameters for experts or routers. Instead of adding new adapters as experts, Monkey Jump treats the adapters already present in each Transformer block (such as query, key, value, up, and […]

Ver mais

We all knew that this year’s CES would be dominated by AI, and the reality is nothing short of that. Major companies and startups have unveiled some of the most unique AI products at the largest consumer tech conference in Las Vegas. When we talk of AI, Meta is among the names that pops up in everyone’s head, and rightly so. Meta teleprompter brings notes right to your eyes Well, Meta showed up at the CES 2026 event, […]

Ver mais

I spent many years working with founders and now I work with researchers. Although there are always individual exceptions, on average it’s surprising to me how different the best people in these groups are (including in some qualities that I had assumed were present in great people everywhere, like very high levels of self-belief). So I’ve been thinking about the ways they’re the same, because maybe there is something to learn about qualities of really effective people in […]

Ver mais

Learned index structures aim to accelerate queries by training machine learning models to approximate the rank function associated with a database attribute. While effective in practice, their theoretical limitations are not fully understood. We present a general framework for proving lower bounds on query time for learned indexes, expressed in terms of their space overhead and parameterized by the model class used for approximation. Our formulation captures a broad family of learned indexes, including most existing designs, as […]

Ver mais

Let’s be honest, code testing is everything but a joyful task. However, a good unit testing framework makes this process as smooth as possible. Eventually, testing becomes a regular and continuous process, accompanied by the assurance that our code will operate just as exact and seamlessly as a Swiss clockwork.

Ver mais

Table of Links Abstract and 1. Introduction Related Works Convex Relaxation Techniques for Hyperbolic SVMs 3.1 Preliminaries 3.2 Original Formulation of the HSVM 3.3 Semidefinite Formulation 3.4 Moment-Sum-of-Squares Relaxation Experiments 4.1 Synthetic Dataset 4.2 Real Dataset Discussions, Acknowledgements, and References A. Proofs B. Solution Extraction in Relaxed Formulation C. On Moment Sum-of-Squares Relaxation Hierarchy D. Platt Scaling [31] E. Detailed Experimental Results F. Robust Hyperbolic Support Vector Machine Abstract Hyperbolic spaces have increasingly been recognized for their outstanding […]

Ver mais

arXiv:2601.08061v1 Announce Type: new Abstract: Language models now provide an interface to express and often solve general problems in natural language, yet their ultimate computational capabilities remain a major topic of scientific debate. Unlike a formal computer, a language model is trained to autoregressively predict successive elements in human-generated text. We prove that chaining a language model’s autoregressive output is sufficient to perform universal computation. That is, a language model can simulate the execution of any algorithm on […]

Ver mais

arXiv:2601.07419v1 Announce Type: new Abstract: Over-smoothing and over-squashing have been extensively studied in the literature on Graph Neural Networks (GNNs) over the past years. We challenge this prevailing focus in GNN research, arguing that these phenomena are less critical for practical applications than assumed. We suggest that performance decreases often stem from uninformative receptive fields rather than over-smoothing. We support this position with extensive experiments on several standard benchmark datasets, demonstrating that accuracy and over-smoothing are mostly uncorrelated […]

Ver mais