Originally in 2019, GPT-2 was trained by OpenAI on 32 TPU v3 chips for 168 hours (7 days), with $8/hour/TPUv3 back then, for a total cost of approx. $43K. It achieves 0.256525 CORE score, which is an ensemble metric introduced in the DCLM paper over 22 evaluations like ARC/MMLU/etc. As of the last few improvements merged into nanochat (many of them originating in modded-nanogpt repo), I can now reach a higher CORE score in 3.04 hours (~$73) on […]

Ver mais

arXiv:2601.17109v1 Announce Type: new Abstract: Health information seeking has fundamentally changed since the onset of Large Language Models (LLM), with nearly one third of ChatGPT’s 800 million users asking health questions weekly. Understanding the sources of those AI generated responses is vital, as health organizations and providers are also investing in digital strategies to organically improve their ranking, reach and visibility in LLM systems like ChatGPT. As AI search optimization strategies are gaining maturity, this study introduces an […]

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

Graph-structured data is foundational to numerous web applications, and watermarking is crucial for protecting their intellectual property and ensuring data provenance. Existing watermarking methods primarily operate on graph structures or entangled graph representations, which compromise the transparency and robustness of watermarks due to the information coupling in representing graphs and uncontrollable discretization in transforming continuous numerical representations into graph structures. This motivates us to propose DRGW, the first graph watermarking framework that addresses these issues through disentangled representation […]

Ver mais

arXiv:2601.10804v1 Announce Type: new Abstract: Large Language Models (LLMs) exhibit strong multilingual capabilities, yet remain fundamentally constrained by the severe imbalance in global language resources. While over 7,000 languages are spoken worldwide, only a small subset (fewer than 100) has sufficient digital presence to meaningfully influence modern LLM training. This disparity leads to systematic underperformance, cultural misalignment, and limited accessibility for speakers of low-resource and extreme-low-resource languages. To address this gap, we introduce Bring Your Own Language (BYOL), […]

Ver mais

We invite you to read an article on how C++ is used in modern game development and why the industry is still not ready to move away from it. The author explores how C++ works at different levels of game engines and how performance requirements, legacy code, and platform constraints make an impact on the industry. We published and translated this article with the copyright holder’s permission. The author is Sergey Kushnirenko. I intended to write a follow-up […]

Ver mais

Built a reservoir computing system (Liquid State Machine) as a learning experiment. Instead of a standard static reservoir, I added biological simulation layers on top to see how constraints affect behavior. What it actually does (no BS): – LSM with 2000+ reservoir neurons, Numba JIT-accelerated – Hebbian + STDP plasticity (the reservoir rewires during runtime) – Neurogenesis/atrophy reservoir can grow or shrink neurons dynamically – A hormone system (3 floats: dopamine, cortisol, oxytocin) that modulates learning rate, reflex […]

Ver mais

Mobile Software Engineering has emerged as a distinct subfield, raising questions about the transferability of its research findings to general Software Engineering. This paper addresses the challenge of evaluating the generalizability of mobile-specific research, using Green Computing as a representative case. We propose a systematic method that combines a mapping study to identify potentially overlooked mobile-specific papers with a focused literature review to assess their broader relevance. Applying this approach, we find that several mobile-specific studies offer insights […]

Ver mais

arXiv:2601.22365v1 Announce Type: new Abstract: The Gilbert-Pollak Conjecture citep{gilbert1968steiner}, also known as the Steiner Ratio Conjecture, states that for any finite point set in the Euclidean plane, the Steiner minimum tree has length at least $sqrt{3}/2 approx 0.866$ times that of the Euclidean minimum spanning tree (the Steiner ratio). A sequence of improvements through the 1980s culminated in a lower bound of $0.824$, with no substantial progress reported over the past three decades. Recent advances in LLMs have […]

Ver mais