This is not a prediction. It is an intellectual exercise. Maybe it is wrong. Maybe it is too early. Maybe it can be discarded as a strange idea. But I think the question is worth asking: if AI-generated text becomes the dominant training substrate, where will the entropy needed for future AI evolution come from? This article continues a line of thought I explored in my previous piece, “More Memory Won’t Fix Your AI Agents”. There, I argued […]

Ver mais

arXiv:2603.10002v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly tasked with producing and manipulating structured artifacts. We consider the task of end-to-end spreadsheet generation, where language models are prompted to produce spreadsheet artifacts to satisfy users’ explicit and implicit constraints, specified in natural language. We introduce SpreadsheetArena, a platform for evaluating models’ performance on the task via blind pairwise evaluations of LLM-generated spreadsheet workbooks. As with other complex, open-ended tasks, relevant evaluation criteria can vary substantially […]

Ver mais

Published on December 4, 2025 6:46 PM GMT Highly capable AI systems might end up deciding the future. Understanding what will drive those decisions is therefore one of the most important questions we can ask. Many people have proposed different answers. Some predict that powerful AIs will learn to intrinsically pursue reward. Others respond by saying reward is not the optimization target, and instead reward “chisels” a combination of context-dependent cognitive patterns into the AI. Some argue that […]

Ver mais

Modern sensing systems generate large volumes of unlabeled multivariate time-series data. This abundance of unlabeled data makes self-supervised learning (SSL) a natural approach for learning transferable representations. However, most existing approaches are optimized for reconstruction or forecasting objectives and often fail to capture the semantic structure required for downstream classification and reasoning tasks. While recent sensor-language alignment methods improve semantic generalization through captioning and zero-shot transfer, they are limited to fixed sensor configurations, such as predefined channel sets, […]

Ver mais

Continual learning, enabling models to acquire new skills and knowledge without degrading existing capabilities, remains a fundamental challenge for foundation models. While on-policy reinforcement learning can reduce forgetting, it requires explicit reward functions that are often unavailable. Learning from expert demonstrations, the primary alternative, is dominated by supervised fine-tuning (SFT), which is inherently off-policy. We introduce Self-Distillation Fine-Tuning (SDFT), a simple method that enables on-policy learning directly from demonstrations. SDFT leverages in-context learning by using a demonstration-conditioned model […]

Ver mais

arXiv:2603.13441v2 Announce Type: replace Abstract: Quantum principal component analysis (qPCA) is commonly formulated as the extraction of eigenvalues and eigenvectors of a covariance-encoded density operator. Yet in many qPCA settings, the practical objective is simpler: projecting data onto the dominant spectral subspace. In this work, we introduce a projection-first framework, the Filtered Spectral Projection Algorithm (FSPA), which bypasses explicit eigenvalue estimation while preserving the essential spectral structure. FSPA amplifies any nonzero warm-start overlap with the leading principal subspace […]

Ver mais

Modern Translation Systems heavily rely on high-quality, large parallel datasets for state-of-the-art performance. However, such resources are largely unavailable for most of the South Asian languages. Among them, Nepali and Tamang fall into such category, with Tamang being among the least digitally resourced languages in the region. This work addresses the gap by developing NepTam20K, a 20K gold standard parallel corpus, and NepTam80K, an 80K synthetic Nepali-Tamang parallel corpus, both sentence-aligned and designed to support machine translation. The […]

Ver mais

Stochastic resetting, where a dynamical process is intermittently returned to a fixed reference state, has emerged as a powerful mechanism for optimizing first-passage properties. Existing theory largely treats static, non-learning processes. Here we ask how stochastic resetting interacts with reinforcement learning, where the underlying dynamics adapt through experience. In tabular grid environments, we find that resetting accelerates policy convergence even when it does not reduce the search time of a purely diffusive agent, indicating a novel mechanism beyond […]

Ver mais

Hey hey. Like the title says, we are currently optimizing our system (hive-mind/swarm-like collective) to be able to create great RL environments. And we are starting with pufferlib envs. It is doing a pretty damn good job atm. We are currently bootstrapped and we are limited on compute. Even a small batch of gpus, of a decent size, would be pretty wild for us. If you have any extra gpus laying around, or would potentially want to sponsor […]

Ver mais