Hasil untuk "Style. Composition. Rhetoric"

Bill Nichols

Yanjie Li, Wenxuan Zhang, Xinqi Lyu et al.

Recently, text-to-image diffusion models have been widely used for style mimicry and personalized customization through methods such as DreamBooth and Textual Inversion. This has raised concerns about intellectual property protection and the generation of deceptive content. Recent studies, such as Glaze and Anti-DreamBooth, have proposed using adversarial noise to protect images from these attacks. However, recent purification-based methods, such as DiffPure and Noise Upscaling, have successfully attacked these latest defenses, showing the vulnerabilities of these methods. Moreover, present methods show limited transferability across models, making them less effective against unknown text-to-image models. To address these issues, we propose a novel anti-mimicry method, StyleGuard. We propose a novel style loss that optimizes the style-related features in the latent space to make it deviate from the original image, which improves model-agnostic transferability. Additionally, to enhance the perturbation's ability to bypass diffusion-based purification, we designed a novel upscale loss that involves ensemble purifiers and upscalers during training. Extensive experiments on the WikiArt and CelebA datasets demonstrate that StyleGuard outperforms existing methods in robustness against various transformations and purifications, effectively countering style mimicry in various models. Moreover, StyleGuard is effective on different style mimicry methods, including DreamBooth and Textual Inversion. The code is available at https://github.com/PolyLiYJ/StyleGuard.

Haojun Tang, Qiwei Lin, Tongda Xu et al.

Attention injection-based style transfer has achieved remarkable progress in recent years. However, existing methods often suffer from content leakage, where the undesired semantic content of the style image mistakenly appears in the stylized output. In this paper, we propose V-Shuffle, a zero-shot style transfer method that leverages multiple style images from the same style domain to effectively navigate the trade-off between content preservation and style fidelity. V-Shuffle implicitly disrupts the semantic content of the style images by shuffling the value features within the self-attention layers of the diffusion model, thereby preserving low-level style representations. We further introduce a Hybrid Style Regularization that complements these low-level representations with high-level style textures to enhance style fidelity. Empirical results demonstrate that V-Shuffle achieves excellent performance when utilizing multiple style images. Moreover, when applied to a single style image, V-Shuffle outperforms previous state-of-the-art methods.

Shreya Havaldar, Adam Stein, Eric Wong et al.

Successful communication depends on the speaker's intended style (i.e., what the speaker is trying to convey) aligning with the listener's interpreted style (i.e., what the listener perceives). However, cultural differences often lead to misalignment between the two; for example, politeness is often lost in translation. We characterize the ways that LLMs fail to translate style - biasing translations towards neutrality and performing worse in non-Western languages. We mitigate these failures with RASTA (Retrieval-Augmented STylistic Alignment), a method that leverages learned stylistic concepts to encourage LLM translation to appropriately convey cultural communication norms and align style.

Tengjie Li, Shikui Tu, Lei Xu

Recent advances in vision-language models have facilitated progress in sketch generation. However, existing specialized methods primarily focus on generic synthesis and lack mechanisms for precise control over sketch styles. In this work, we propose a training-free framework based on diffusion models that enables explicit style guidance via textual prompts and referenced style sketches. Unlike previous style transfer methods that overwrite key and value matrices in self-attention, we incorporate the reference features as auxiliary information with linear smoothing and leverage a style-content guidance mechanism. This design effectively reduces content leakage from reference sketches and enhances synthesis quality, especially in cases with low structural similarity between reference and target sketches. Furthermore, we extend our framework to support controllable multi-style generation by integrating features from multiple reference sketches, coordinated via a joint AdaIN module. Extensive experiments demonstrate that our approach achieves high-quality sketch generation with accurate style alignment and improved flexibility in style control. The official implementation of M3S is available at https://github.com/CMACH508/M3S.

Kunxiao Liu, Guowu Yuan, Hao Wu et al.

Artistic style transfer aims to use a style image and a content image to synthesize a target image that retains the same artistic expression as the style image while preserving the basic content of the content image. Many recently proposed style transfer methods have a common problem; that is, they simply transfer the texture and color of the style image to the global structure of the content image. As a result, the content image has a local structure that is not similar to the local structure of the style image. In this paper, we present an effective method that can be used to transfer style patterns while fusing the local style structure into the local content structure. In our method, dif-ferent levels of coarse stylized features are first reconstructed at low resolution using a Coarse Network, in which style color distribution is roughly transferred, and the content structure is combined with the style structure. Then, the reconstructed features and the content features are adopted to synthesize high-quality structure-aware stylized images with high resolution using a Fine Network with three structural selective fusion (SSF) modules. The effectiveness of our method is demonstrated through the generation of appealing high-quality stylization results and a com-parison with some state-of-the-art style transfer methods.

Chon In Leong, I-Ling Chung, Kin-Fong Chao et al.

The goal of music style transfer is to convert a music performance by one instrument into another while keeping the musical contents unchanged. In this paper, we investigate another style transfer scenario called ``failed-music style transfer''. Unlike the usual music style transfer where the content remains the same and only the instrumental characteristics are changed, this scenario seeks to transfer the music from the source instrument to the target instrument which is deliberately performed off-pitch. Our work attempts to transfer normally played music into off-pitch recorder music, which we call ``failed-style recorder'', and study the results of the conversion. To carry out this work, we have also proposed a dataset of failed-style recorders for this task, called ``FR109 Dataset''. Such an experiment explores the music style transfer task in a more expressive setting, as the generated audio should sound like an ``off-pitch recorder'' while maintaining a certain degree of naturalness.

Carolina van Baalen, Laura Alvarez, Robert Style et al.

Active systems comprising micron-sized self-propelling units, also termed microswimmers, are promising candidates for the bottom-up assembly of small structures and reconfigurable materials. Here we leverage field-driven colloidal assembly to induce structural transformations in dense layers of microswimmers driven by an alternating current (AC) electric field and confined in a microfabricated trap under the influence of gravity. By varying the electric field frequency, we realize significant structural transformations, from a gas-like state at high frequencies to dynamically rearranging dense crystalline clusters at lower frequencies, characterized by vorticity in their dynamics. We demonstrate the ability to switch between these states on-demand, showing that the clustering mechanism differs from motility-induced phase separation. Our results offer a valuable framework for designing high-density active matter systems with controllable structural properties, envisioned to advance the development of artificial materials with self-healing and reconfiguration capabilities.

Kai Konen, Sophie Jentzsch, Diaoulé Diallo et al.

This research explores strategies for steering the output of large language models (LLMs) towards specific styles, such as sentiment, emotion, or writing style, by adding style vectors to the activations of hidden layers during text generation. We show that style vectors can be simply computed from recorded layer activations for input texts in a specific style in contrast to more complex training-based approaches. Through a series of experiments, we demonstrate the effectiveness of activation engineering using such style vectors to influence the style of generated text in a nuanced and parameterisable way, distinguishing it from prompt engineering. The presented research constitutes a significant step towards developing more adaptive and effective AI-empowered interactive systems.

Abhishek Saroha, Mariia Gladkova, Cecilia Curreli et al.

3D scene stylization extends the work of neural style transfer to 3D. A vital challenge in this problem is to maintain the uniformity of the stylized appearance across multiple views. A vast majority of the previous works achieve this by training a 3D model for every stylized image and a set of multi-view images. In contrast, we propose a novel architecture trained on a collection of style images that, at test time, produces real time high-quality stylized novel views. We choose the underlying 3D scene representation for our model as 3D Gaussian splatting. We take the 3D Gaussians and process them using a multi-resolution hash grid and a tiny MLP to obtain stylized views. The MLP is conditioned on different style codes for generalization to different styles during test time. The explicit nature of 3D Gaussians gives us inherent advantages over NeRF-based methods, including geometric consistency and a fast training and rendering regime. This enables our method to be useful for various practical use cases, such as augmented or virtual reality. We demonstrate that our method achieves state-of-the-art performance with superior visual quality on various indoor and outdoor real-world data.

Yanxia Feng, Dominic Gerber, Stefanie Heyden et al.

Hydrogels are particularly versatile materials that are widely found in both Nature and industry. One key reason for this versatility is their high water content, which lets them dramatically change their volume and many of their mechanical properties -- often by orders of magnitude -- as they swell and dry out. Currently, we lack techniques that can precisely characterize how these properties change with water content. To overcome this challenge, here we develop Gel-Freezing Osmometry (GelFrO): an extension of freezing-point osmometry. We show how GelFrO can measure a hydrogel's mechanical response to compression and osmotic pressure, while only using small, $O(100μ$L$)$ samples. Because the technique allows measurement of properties over an unusually wide range of water contents, it allows us to accurately test theoretical predictions. We find simple, power-law behavior for both mechanical response to compression, and osmotic pressure, while these are not well-captured by classical Flory-Huggins theory. We interpret this power-law behavior as a hallmark of a microscopic fractal structure of the gel's polymer network, and propose a simple way to connect the gel's fractal dimension to its mechanical and osmotic properties. This connection is supported by observations of hydrogel microstructures using small-angle x-ray scattering. Finally, our results motivate us to propose an updated constitutive model describing hydrogel swelling, and mechanical response.

Manuel Caritto

Centuria, il metaromanzo di Giorgio Manganelli apparso nel 1979, è (usando le parole di Italo Calvino) un’enciclopedia aperta. Scaturita al tempo stesso da un evidente scetticismo epistemologico e da un’idea di elaborazione letteraria più problematica, l’enciclopedia aperta si regge sull’equilibrio precario tra una forza centrifuga, fonte di una pluralità di storie e vicende potenzialmente infinite; e una forza centripeta, che si occupa al contrario della loro sistemazione all’interno di una rigida struttura. Questo saggio si propone di isolare i due poli di questo congegno narrativo paradossale, rispettivamente mettendo a fuoco: per le forze strutturanti, il rapporto con alcune opere di consultazione (dizionario, enciclopedia); per le forze disgreganti, la ricorrenza di specifici ‘meccanismi di pluralizzazione’ narrativa quali il ri-uso di materiale mutuato da opere precedenti, un’istanza enunciativa ipotetica e congetturale, la predilezione per i temi dell’attesa e della potenzialità degli eventi, e infine la ricorrenza dei concetti di elenco, schema e catalogo.

Alfredo Bushby

Entre los numerosos méritos que se pueden atribuir a Entre fuegos. Dramaturgia peruana del periodo de subversión armada y antisubversión, de Percy Encinas, cabe destacar dos que, de muchas formas, abarcan a los demás. En primer lugar, está el hecho de ensanchar el círculo del corpus de estudio del teatro peruano para incluir representaciones y textos poco considerados anteriormente. En segundo lugar, el tratamiento “riesgosamente prudente” de los tiempos de la subversión armada y antisubversión (SAAS), periodo en que se focaliza el estudio, lo cual es otra forma de incluir dichos discursos en el mencionado corpus.

Zheng Lin, Zhao Zhang, Kang-Rui Zhang et al.

Neural style transfer (NST) can create impressive artworks by transferring reference style to content image. Current image-to-image NST methods are short of fine-grained controls, which are often demanded by artistic editing. To mitigate this limitation, we propose a drawing-like interactive style transfer (IST) method, by which users can interactively create a harmonious-style image. Our IST method can serve as a brush, dip style from anywhere, and then paint to any region of the target content image. To determine the action scope, we formulate a fluid simulation algorithm, which takes styles as pigments around the position of brush interaction, and diffusion in style or content images according to the similarity maps. Our IST method expands the creative dimension of NST. By dipping and painting, even employing one style image can produce thousands of eye-catching works. The demo video is available in supplementary files or in http://mmcheng.net/ist.

Tai-Yin Chiu, Danna Gurari

Photorealistic style transfer is the task of synthesizing a realistic-looking image when adapting the content from one image to appear in the style of another image. Modern models commonly embed a transformation that fuses features describing the content image and style image and then decodes the resulting feature into a stylized image. We introduce a general-purpose transformation that enables controlling the balance between how much content is preserved and the strength of the infused style. We offer the first experiments that demonstrate the performance of existing transformations across different style transfer models and demonstrate how our transformation performs better in its ability to simultaneously run fast, produce consistently reasonable results, and control the balance between content and style in different models. To support reproducing our method and models, we share the code at https://github.com/chiutaiyin/LS-FT.

Mireille Fares, Michele Grimaldi, Catherine Pelachaud et al.

Modeling virtual agents with behavior style is one factor for personalizing human agent interaction. We propose an efficient yet effective machine learning approach to synthesize gestures driven by prosodic features and text in the style of different speakers including those unseen during training. Our model performs zero shot multimodal style transfer driven by multimodal data from the PATS database containing videos of various speakers. We view style as being pervasive while speaking, it colors the communicative behaviors expressivity while speech content is carried by multimodal signals and text. This disentanglement scheme of content and style allows us to directly infer the style embedding even of speaker whose data are not part of the training phase, without requiring any further training or fine tuning. The first goal of our model is to generate the gestures of a source speaker based on the content of two audio and text modalities. The second goal is to condition the source speaker predicted gestures on the multimodal behavior style embedding of a target speaker. The third goal is to allow zero shot style transfer of speakers unseen during training without retraining the model. Our system consists of: (1) a speaker style encoder network that learns to generate a fixed dimensional speaker embedding style from a target speaker multimodal data and (2) a sequence to sequence synthesis network that synthesizes gestures based on the content of the input modalities of a source speaker and conditioned on the speaker style embedding. We evaluate that our model can synthesize gestures of a source speaker and transfer the knowledge of target speaker style variability to the gesture generation task in a zero shot setup. We convert the 2D gestures to 3D poses and produce 3D animations. We conduct objective and subjective evaluations to validate our approach and compare it with a baseline.

Camilo Rubén Fernández-Cozman

Javier Sologuren es uno de los más notables autores latinoamericanos del siglo xx porque renovó la poesía peruana durante los años cincuenta, fue traductor literario muy reconocido y reflexionó rigurosamente sobre los poetas simbolistas y vanguardistas franceses. Publicó su poemario Otoño, endechas en 1959 donde se observa el influjo de la obra de Stéphane Mallarmé. El artículo se sustenta en las propuestas de la Retórica Cultural de Tomás Albaladejo en lo que respecta al empleo de la categoría de motor metafórico. Sobre la base del concepto de imaginario cultural de Antonio García Berrio, se señala cómo funciona el motor metafórico en la poesía de Sologuren a partir de la noción de la instancia de la enunciación y de la recepción. Sologuren se nutre de la poética de Mallarmé y reestructura las propuestas del escritor francés con el fin de crear una nueva metáfora. Para fundamentar su hipótesis, además, se aborda la ensayística del poeta peruano donde habla de las particularidades de Mallarmé y sus lazos con creadores como Paul Valéry o Guillaume Apollinaire.

Graziano Krätli

Recensione di Within the Sweet Noise of Life, di Sandro Penna (Seagull Books 2021).

Kai Li, Chenyue Jiao

The data paper is an emerging academic genre that focuses on the description of research data objects. However, there is a lack of empirical knowledge about this rising genre in quantitative science studies, particularly from the perspective of its linguistic features. To fill this gap, this research aims to offer a first quantitative examination of which rhetorical moves-rhetorical units performing a coherent narrative function-are used in data paper abstracts, as well as how these moves are used. To this end, we developed a new classification scheme for rhetorical moves in data paper abstracts by expanding a well-received system that focuses on English-language research article abstracts. We used this expanded scheme to classify and analyze rhetorical moves used in two flagship data journals, Scientific Data and Data in Brief. We found that data papers exhibit a combination of IMRaD- and data-oriented moves and that the usage differences between the journals can be largely explained by journal policies concerning abstract and paper structure. This research offers a novel examination of how the data paper, a novel data-oriented knowledge representation, is composed, which greatly contributes to a deeper understanding of data and data publication in the scholarly communication system.

Eleftheria Briakou, Sweta Agrawal, Ke Zhang et al.

This paper reviews and summarizes human evaluation practices described in 97 style transfer papers with respect to three main evaluation aspects: style transfer, meaning preservation, and fluency. In principle, evaluations by human raters should be the most reliable. However, in style transfer papers, we find that protocols for human evaluations are often underspecified and not standardized, which hampers the reproducibility of research in this field and progress toward better human and automatic evaluation methods.