Diffusion large language models (dLLMs) have emerged as a new architecture following auto regressive models. Their denoising process offers a powerful generative advantage, but they present significant challenges in learning and understanding numerically sensitive mathematical and order-sensitive logical tasks. Current training methods, including pre-training, fine-tuning, and reinforcement learning, focus primarily on improving general knowledge retention and reasoning abilities, but lack a comprehensive understanding of mathematical and logical patterns. We propose DSFT, a simple yet effective Diffusion SFT strategy, by adjusting the masking strategy and loss function, guiding models to understand mathematical and logical patterns. This strategy can be flexibly combined with pre-training, reinforcement learning, and other training methods. Validated on models such as LLaDA and Dream series, we prove that DSFT on small-scale data can achieve improvements of 5-10% and approximately 2% on mathematical and logical problems, respectively. This inspiring masking approach offers insights for future learning of specific patterns, which can be easily and efficiently combined with other tr
The evolution of colour vision is captivating, as it reveals the adaptive strategies of extinct species while simultaneously inspiring innovations in modern imaging technology. In this study, we present a simplified model of visual transduction in the retina, introducing a novel opsin layer. We quantify evolutionary pressures by measuring machine vision recognition accuracy on colour images shaped by specific opsins. Building on this, we develop an evolutionary conservation optimisation algorithm to reconstruct the spectral sensitivity of opsins, enabling mutation-driven adaptations to to more effectively spot fruits or predators. This model condenses millions of years of evolution within seconds on GPU, providing an experimental framework to test long-standing hypotheses in evolutionary biology , such as vision of early mammals, primate trichromacy from gene duplication, retention of colour blindness, blue-shift of fish rod and multiple rod opsins with bioluminescence. Moreover, the model enables speculative explorations of hypothetical species, such as organisms with eyes adapted to the conditions on Mars. Our findings suggest a minimalist yet effective approach to task-specific
The smart textile and wearables sector is looking towards advancing technologies to meet both industry, consumer and new emerging innovative textile application demands, within a fast paced textile industry. In parallel inspiration based on the biological neural workings of the human brain is driving the next generation of artificial intelligence. Artificial intelligence inspired hardware (neuromorphic computing) and software modules mimicking the processing capabilities and properties of neural networks and the human nervous system are taking shape. The textile sector needs to actively look at such emerging and new technologies taking inspiration from their workings and processing methods in order to stimulate new and innovative embedded intelligence advancements in the etextile world. This emerging next generation of Artificial intelligence(AI) is rapidly gaining interest across varying industries (textile, medical, automotive, aerospace, military). How such properties can inspire and drive advancements within the etextiles sector needs to be considered. This paper will provide an insight into current nanotechnology and artificial intelligence advancements in the etextiles domain
Inspiration is linked to various positive outcomes, such as increased creativity, productivity, and happiness. Although inspiration has great potential, there has been limited effort toward identifying content that is inspiring, as opposed to just engaging or positive. Additionally, most research has concentrated on Western data, with little attention paid to other cultures. This work is the first to study cross-cultural inspiration through machine learning methods. We aim to identify and analyze real and AI-generated cross-cultural inspiring posts. To this end, we compile and make publicly available the InspAIred dataset, which consists of 2,000 real inspiring posts, 2,000 real non-inspiring posts, and 2,000 generated inspiring posts evenly distributed across India and the UK. The real posts are sourced from Reddit, while the generated posts are created using the GPT-4 model. Using this dataset, we conduct extensive computational linguistic analyses to (1) compare inspiring content across cultures, (2) compare AI-generated inspiring posts to real inspiring posts, and (3) determine if detection models can accurately distinguish between inspiring content across cultures and data sou
Data visualization practitioners routinely invoke inspiration, yet we know little about how it is constructed in public conversations. We conduct a discourse analysis of 31 episodes from five popular data visualization podcasts. Podcasts are public-facing and inherently performative: guests manage impressions, articulate values, and model "good practice" for broad audiences. We use this performative setting to examine how legitimacy, identity, and practice are negotiated in community talk. We show that "inspiration talk" is operative rather than ornamental: speakers legitimize what counts, who counts, and how work proceeds. Our analysis surfaces four adjustable evaluation criteria by which inspiration is judged-novelty, authority, authenticity, and affect-and three operative metaphors that license different practices-spark, muscle, and resource bank. We argue that treating inspiration as a boundary object helps explain why these frames coexist across contexts. Findings provide a vocabulary for examining how inspiration is mobilized in visualization practice, with implications for evaluation, pedagogy, and the design of galleries and repositories that surface inspirational examples.
Anti-Ramsey theory was initiated in 1975 by Erdős, Simonovits and Sós, inspiring hundreds of publications since then. The present work is the third and last piece of our trilogy in which we introduce a far-reaching generalization via the following two functions for any graph $G$ and family ${\cal F}$ of graphs: If $K_2 \in {\cal F}$, let $f(n,G|{\cal F})$ be the smallest integer $k$ such that every edge coloring of $K_n$ with at least $k$ colors forces a copy of $G$ in which all color classes are members of ${\cal F}$. If $K_2 otin {\cal F}$, let $g(n,G|{\cal F})$ be the largest integer $k$ for which there exists an edge coloring of $K_n$ using exactly $k$ colors, such that every copy of $G$ contains an induced color class which is a member of ${\cal F}$. We develop methods suitable for deriving asymptotically tight results for the $f$-function and the $g$-function for many combinations of $G$ and ${\cal F}$. The preceding parts of the trilogy are arXiv: 2405.19812 and 2408.04257, published in Discrete Applied Math. Vol. 363 and Mathematics Vol. 12:23, respectively.
The Noisy Intermediate-Scale Quantum (NISQ) era of technology in which we currently find ourselves is defined by non-universality, susceptibility to errors and noise, and a search for useful applications. While demonstrations of practical quantum advantage remain elusive in this era, it provides space to develop and analyze the advantages and limitations of systems and their ability to solve problems. In this work, we critically assess a proposed quantum algorithm for the graph isomorphism problem, implemented on a photonic quantum device. Inspired by the nature of this quantum algorithm, we formulate a necessary condition for the isomorphism of graphs encoded in Gaussian boson samplers and a classical algorithm to test for it. Our classical algorithm makes use of efficiently computable statistical properties of a quantum sampling system to show a pair of graphs fail to meet our necessary condition and thus cannot be isomorphic. We analyze our algorithm in the context of the inspiring, sampler-based quantum algorithm of Bràdler et. al., the classical color refinement algorithm, and the state-of-the-art quasi-polynomial Babai algorithm.
Recent large language model (LLM) research has undergone an architectural shift from encoder-decoder modeling to nowadays the dominant decoder-only modeling. This rapid transition, however, comes without a rigorous comparative analysis especially \textit{from the scaling perspective}, raising concerns that the potential of encoder-decoder models may have been overlooked. To fill this gap, we revisit encoder-decoder LLM (RedLLM), enhancing it with recent recipes from decoder-only LLM (DecLLM). We conduct a comprehensive comparison between RedLLM, pretrained with prefix language modeling (LM), and DecLLM, pretrained with causal LM, at different model scales, ranging from $\sim$150M to $\sim$8B. Using RedPajama V1 (1.6T tokens) for pretraining and FLAN for instruction tuning, our experiments show that RedLLM produces compelling scaling properties and surprisingly strong performance. While DecLLM is overall more compute-optimal during pretraining, RedLLM demonstrates comparable scaling and context length extrapolation capabilities. After instruction tuning, RedLLM achieves comparable and even better results on various downstream tasks while enjoying substantially better inference effic
Online, visual artists have more places than ever to routinely share their creative work and connect with other artists. These interactions support the routine enactment of creative identity in artists and provide inspirational opportunities for artists. As creative work shifts online, interactions between artists and routines around how these artists get inspired to do creative work are mediated by and through the logics of the online platforms where they take place. In an interview study of 22 artists, this paper explores the interplay between the development of artists' creative identities and the, at times, contradictory practices they have around getting inspired. We find platforms which support the disciplined practice of creative work while supporting spontaneous moments of inspiration, play an increasing role in passive approaches to searching for inspiration, and foster numerous small community spaces for artists to negotiate their creative identities. We discuss how platforms can better support and embed mechanisms for inspiration into their infrastructures into their design and platform policy.
Metamaterials benefit from unique architected patterns to achieve lightweight with exceptional mechanical properties inaccessible to conventional materials. Typical mechanical metamaterials are inspired by crystal-like lattice structures, whose closely packed frameworks often exhibit a rigid mechanical nature. Here, we present polymer-inspired metamaterials (PIMs) by programming deformation and strengthening mechanisms that mimic the mechanical roles of key constituent elements in polymer networks. By combining metamaterial programmability with polymer-inspired structures, we design crosslinking, proto-crystalline order, and entanglement in PIMs to enable macroscale strengthening mechanisms inspired by crosslink, molecular-density, and pre-stretch strengthening in polymers, expanding the metamaterial structure-property design space. This macroscale polymer-inspired programmability also suggests that PIMs could serve as a design platform incorporating the programmability strategies to achieve desired deformation and strengthening responses, holding a potential for applications in soft robotic joints and compliant connectors.