Hasil untuk "Style. Composition. Rhetoric"

Sami-ul Ahmed

Large language models (LLMs) are typically personalized via prompt engineering or parameter-efficient fine-tuning such as LoRA. However, writing style can be difficult to distill into a single prompt, and LoRA fine-tuning requires computationally intensive training and infrastructure. We investigate a possible lightweight alternative: steering a frozen LLM with n-gram style priors injected in logit space at decoding time. We train an n-gram model on stylistically distinct corpora -- including Don Quixote, CNN/DailyMail news headlines, and arXiv abstracts -- constructing an interpolated 1-to-3-gram prior over next-token probabilities. During generation we modify the LLM's logits by adding a weighted sum of style log-probabilities from each n-gram order that matches the current context, scaled by a control parameter lambda in [0, 1]. We sweep lambda and style corpora and report style perplexity under the n-gram model, base-model perplexity as a proxy for fluency, Jensen-Shannon (JS) divergence between the original and steered token distributions, and token-overlap statistics. On TinyLlama-1.1B we identify a single narrow regime (for the Don Quixote corpus at lambda=0.1) where style perplexity improves by 24.7% and base-model perplexity improves by 51.4% relative to the frozen model. Outside this regime, and for multi-author corpora such as CNN/DailyMail and arXiv abstracts, even small nonzero lambda values generally result in worse style and fluency, and larger lambda values lead to collapse with extreme perplexities and incoherent text. Logit-space injection of n-gram style priors provides lightweight, tunable style control, but it is fragile: it operates effectively only within a narrow range of low lambda values and is consistently outperformed by prompting and LoRA.

Ziba Mosadegh Mehrjardi, Azizallah Tavakoli kafi Abad, Hadi Heydari Niya

Skilled poets employ a range of linguistic and rhetorical techniques to persuasively engage their audience and harness the expressive power of language in artistic and literary ways. Effective writers strategically use various sentence types—interrogative, imperative, prohibitive, exclamatory, and emphatic—to bridge grammatical forms with semantic meaning and communicate complex ideas. For instance, rhetorical questions serve not only to convey meaning but also to broaden the interpretive scope of the text, allowing the poet to suggest multiple layers of intention and stimulate reflection in the reader. The dynamic interplay between speaker and audience is essential to achieving effective verbal communication. Rumi, a masterful poet and mystic, is renowned for using poetic composition to convey profound spiritual and didactic insights. In the Masnavi-ye Ma’navi, the frequent use of interrogative forms directed at the reader creates a continuous flow of thought, guiding the audience through layered philosophical and mystical concepts.IntroductionThe rhetorical functions of questioning, particularly those involving denial, play a crucial role in this work. This article explores the rhetorical theme of denial and its artistic implications, including eternal negation, exclusion, bowing, contempt, praise, rejection, despair, mockery, support, regret, and more. It emphasizes the significance of employing questioning techniques to generate aesthetic motivation, highlights the distinctiveness of “negation” in questioning, and offers a “prescription” for its use.The spiritual Masnavi contains numerous rhetorical interrogative sentences, which can be broadly categorized into two groups. The second book of the Masnavi-ye Ma’navi primarily focuses on the first category: negative interrogative statements. Below, these interrogative sentences are listed in order of their frequency, each serving a unique rhetorical purpose:Eternal negation: This interrogative form emphasizes the perpetual exclusion or denial of a subject.Insult: A question that intentionally includes denial to embarrass or criticize the listener, another person, or a concept.Exclusion: The speaker uses this type of interrogative to stress the improbability of a situation by posing a question that implies negation.Disability statement: This form often involves expressing reverence and admiration for the subject or individual under discussion.Ignorance: Refers to a state free from attachment or sensual cravings; the role of this interrogative is conceptual or virtual.Hasr wa Qasr: This involves examining or attributing a unique characteristic to a specific attribute.Despair and disappointment: The speaker expresses disillusionment or rejection of a subject explicitly through denial.Tahweel and scare: A rhetorical style intended to frighten or intimidate the audience.Facilitating and easy to say: This style highlights the relative simplicity of performing an action that may seem difficult for the audience.Bragging: This form employs rhetorical questioning to emphasize the speaker’s pride and self-promotion regarding a topic, effectively conveying the essence of the argument.This book may also include other subdivisions, such as those found in colloquial language. Molana’s primary objective in this didactic collection is clearly to advise and guide the audience, a goal he achieves through the strategic use of questions fulfilling various rhetorical roles.Literature ReviewPrimary references on this subject include classical Arabic texts such as Miftah al-Uloom (Sakkaki, 1407: 350), Al-Aydhah (Qazvini, n.d.: 55), Motuwal (Taftazani, 1416: 237), Mughni al-Labib (Ansari, 1406: 26), and Khatshar al-Ma’ani. These seminal works provide a foundational basis for scholarly discussion, as highlighted by Irfan (1372: 320), which marks the beginning of an important chapter inviting academics to engage deeply with the vast knowledge of meanings derived from the miraculous source of the Holy Quran.Persian rhetorical scholars have further developed this field by drawing upon these Arabic sources, while also integrating Qur’anic examples and the rhetorical richness found in vernacular language traditions. Among these influential works is Jawahar al-Balagha, a collection that encompasses key texts such as Mukhtasar al-Ma’ani, Muftah al-Uloom, Darr al-Adab, Hanjar-e-Giftar, Taraz-e-Sukhan, and Rite of Rhetoric, among others.Research MethodologyThis study employs an analytical-descriptive approach, relying primarily on library research as its methodology. The report offers a comprehensive analysis of the findings, including detailed statistics, the purpose behind each rhetorical method, and the classification of virtual meanings. Furthermore, it provides examples for each category and offers thorough explanations of the verses containing these rhetorical sentences.The author strategically employs various rhetorical techniques throughout the presentation. Each section begins with a relevant passage from the Holy Quran that illustrates the targeted rhetorical style. This is followed by examples drawn from classical language resources as well as colloquial expressions to enhance clarity and relevance. Finally, the author presents a verse from the second book of Masnavi-ye Ma’navi that exemplifies the particular rhetorical form under discussion.DiscussionArticles like this serve as valuable resources for students of Persian literature and others interested in the study of semantics. This article, in particular, focuses on the compositional effects of words within the field of semantics. Through detailed explanations, it aims to help readers understand why writers often express broader concepts than those typically encountered in conventional usage.We anticipate that the author will continue to produce papers exploring the various rhetorical functions of compositional phrases, such as commands, restrictions, calls, and desires. Students interested in Persian literature now have an enhanced opportunity to appreciate the elegance of this field, along with the artistry of communication and originality. They are encouraged to engage deeply with the works of poets and writers who exhibit a distinctive approach to rhetoric and semantics.For a comprehensive understanding of rhetorical connotations, it is advisable to analyze Quranic interrogatives, colloquial expressions, and the literary works of eloquent authors. Unfortunately, much of the existing literature on rhetorical sciences lacks a thorough and all-encompassing treatment of this important subject.ConclusionRhetoricians widely recognize the concept of the "negative question," with its various artistic interpretations, as a crucial and frequently discussed topic. Its importance has prompted many writers to explore this subject extensively. Rumi’s style of expression, drawing on the language of the Holy Quran, effectively conveys mystical concepts to his readers through poetic composition. The foundational principle of the educational framework in the spiritual Masnavi is to guide the audience toward the recognition of ultimate truth and the negation of falsehoods, including the rejection of superficial or worldly attachments.In the second book of the Masnavi, 59.5% of rhetorical questions are negative interrogatives. Among these, 66.4% serve the artistic function of "eternal negation." This high frequency reflects Rumi’s profound stance on the rejection of pleasure and transient desires.

Piotr F. Piekutowski

This article introduces characterisation of the tender narrator concept by Polish writer Olga Tokarczuk, which was a central point in her Nobel Prize lecture (2019). During the identification, three key elements of Tokarczuk’s project are specified: the bond of diegetic forms with climate and environmental crisis of the Anthropocene; dynamically changing, fragmented collective and individual perspectives; and the titular narrative tenderness manifested in sensitivity to more-than-human voices, networks, and relations. Through this, the potential of this idea is included in the repertoire of econarratological research and, more broadly, non-anthropocentric narrative theories. To detail the manifestations of the fourth-person narrative, as the tender narrative is also called, this paper problematises spatiotemporal experiences based on the example of Tokarczuk’s novel The Empusium (2024). In the analysis of how representations of time and space are mediated in the tender story, aspects such as interdependencies, despatialisation and fragmentation are brought to the fore.

Guido Scaravilli

Recensione di Concetta Maria Pagliuca e Filippo Pennacchio (a cura di), Tempora, i tempi verbali nel racconto Vol.1. Biblion, 2023; Francesco De Cristofaro, Paolo Giovannetti e Giovanni Maffei (a cura di), Tempora, i tempi verbali nel racconto Vol. 2. Biblion, 2024.

Nam Anh Dinh, Itai Lang, Hyunwoo Kim et al.

We present Geometry in Style, a new method for identity-preserving mesh stylization. Existing techniques either adhere to the original shape through overly restrictive deformations such as bump maps or significantly modify the input shape using expressive deformations that may introduce artifacts or alter the identity of the source shape. In contrast, we represent a deformation of a triangle mesh as a target normal vector for each vertex neighborhood. The deformations we recover from target normals are expressive enough to enable detailed stylizations yet restrictive enough to preserve the shape's identity. We achieve such deformations using our novel differentiable As-Rigid-As-Possible (dARAP) layer, a neural-network-ready adaptation of the classical ARAP algorithm which we use to solve for per-vertex rotations and deformed vertices. As a differentiable layer, dARAP is paired with a visual loss from a text-to-image model to drive deformations toward style prompts, altogether giving us Geometry in Style. Our project page is at https://threedle.github.io/geometry-in-style.

Jaehyun Lee, Wonhark Park, Wonsik Shin et al.

Diffusion-based text-to-image models have achieved remarkable results in synthesizing diverse images from text prompts and can capture specific artistic styles via style personalization. However, their entangled latent space and lack of smooth interpolation make it difficult to apply distinct painting techniques in a controlled, regional manner, often causing one style to dominate. To overcome this, we propose a zero-shot diffusion pipeline that naturally blends multiple styles by performing style composition on the denoised latents predicted during the flow-matching denoising process of separately trained, style-specialized models. We leverage the fact that lower-noise latents carry stronger stylistic information and fuse them across heterogeneous diffusion pipelines using spatial masks, enabling precise, region-specific style control. This mechanism preserves the fidelity of each individual style while allowing user-guided mixing. Furthermore, to ensure structural coherence across different models, we incorporate depth-map conditioning via ControlNet into the diffusion framework. Qualitative and quantitative experiments demonstrate that our method successfully achieves region-specific style mixing according to the given masks.

Ruth Amossy

Roberta Martina Zagarella, Marco Annoni

This paper aims to explore how institutions may counteract conspiracy theories using appropriate discursive resources. We use a rhetorical approach to analyze the first European information campaign launched in 2020 to counteract conspiracy theories about COVID-19 vaccines. On this basis, we advance a series of practical recommendations for institutions to counteract conspiracy theories through information campaigns.

Lingchen Yang, Gaspard Zoss, Prashanth Chandran et al.

3D facial animation is often produced by manipulating facial deformation models (or rigs), that are traditionally parameterized by expression controls. A key component that is usually overlooked is expression 'style', as in, how a particular expression is performed. Although it is common to define a semantic basis of expressions that characters can perform, most characters perform each expression in their own style. To date, style is usually entangled with the expression, and it is not possible to transfer the style of one character to another when considering facial animation. We present a new face model, based on a data-driven implicit neural physics model, that can be driven by both expression and style separately. At the core, we present a framework for learning implicit physics-based actuations for multiple subjects simultaneously, trained on a few arbitrary performance capture sequences from a small set of identities. Once trained, our method allows generalized physics-based facial animation for any of the trained identities, extending to unseen performances. Furthermore, it grants control over the animation style, enabling style transfer from one character to another or blending styles of different characters. Lastly, as a physics-based model, it is capable of synthesizing physical effects, such as collision handling, setting our method apart from conventional approaches.

Stefan Kindermann, Bernd Hofmann

We consider the composition of operators with non-closed range in Hilbert spaces and how the nature of ill-posedness is affected by their composition. Specifically, we study the \mbox{Hausdorff-,} Cesàro-, integration operator, and their adjoints, as well as some combinations of those. For the composition of the Hausdorff- and the Cesàro-operator, we give estimates of the decay of the corresponding singular values. As a curiosity, this provides also an example of two practically relevant non-compact operators, for which their composition is compact. Furthermore, we characterize those operators for which a composition with a non-compact operator gives a compact one.

Alex Reinhart, Ben Markey, Michael Laudenbach et al.

Large language models (LLMs) are capable of writing grammatical text that follows instructions, answers questions, and solves problems. As they have advanced, it has become difficult to distinguish their output from human-written text. While past research has found some differences in surface features such as word choice and punctuation, and developed classifiers to detect LLM output, none has studied the rhetorical styles of LLMs. Using several variants of Llama 3 and GPT-4o, we construct two parallel corpora of human- and LLM-written texts from common prompts. Using Douglas Biber's set of lexical, grammatical, and rhetorical features, we identify systematic differences between LLMs and humans and between different LLMs. These differences persist when moving from smaller models to larger ones, and are larger for instruction-tuned models than base models. This observation of differences demonstrates that despite their advanced abilities, LLMs struggle to match human stylistic variation. Attention to more advanced linguistic features can hence detect patterns in their behavior not previously recognized.

Younes Poorghorban, Bakhtiar Sadjadi

The present inquiry endeavors to scrutinize the process of identity formation with regard to the Culture/Nature dichotomy within the milieu of Lois Lowry's post-apocalyptic dystopian narrative, The Giver. The antipodal forces of Culture and Nature are instrumental in shaping the social subjectivities of individuals. Lowry's post-apocalyptic dystopia portrays a society in which these antitheses are comprehensively epitomized. Our objective is to explicate the genesis of post-apocalyptic identities and to elucidate the representation of Nature/Culture within the social context of the aforementioned literary work. Furthermore, the polarity between power and resistance, which is of notable import to cultural studies, is nonexistent within this post-apocalyptic dystopia. Consequently, the establishment of identities transpires not at the site of contention between power and resistance, but exclusively through the ascendency of the imperializing power. As a corollary, the elimination of the recollections of those individuals who are unable to oppose the imperializing power is integral to the construction of homogeneous identities.

Zhanghan Ke, Yuhao Liu, Lei Zhu et al.

In this paper, we present a Neural Preset technique to address the limitations of existing color style transfer methods, including visual artifacts, vast memory requirement, and slow style switching speed. Our method is based on two core designs. First, we propose Deterministic Neural Color Mapping (DNCM) to consistently operate on each pixel via an image-adaptive color mapping matrix, avoiding artifacts and supporting high-resolution inputs with a small memory footprint. Second, we develop a two-stage pipeline by dividing the task into color normalization and stylization, which allows efficient style switching by extracting color styles as presets and reusing them on normalized input images. Due to the unavailability of pairwise datasets, we describe how to train Neural Preset via a self-supervised strategy. Various advantages of Neural Preset over existing methods are demonstrated through comprehensive evaluations. Notably, Neural Preset enables stable 4K color style transfer in real-time without artifacts. Besides, we show that our trained model can naturally support multiple applications without fine-tuning, including low-light image enhancement, underwater image correction, image dehazing, and image harmonization. Project page with demos: https://zhkkke.github.io/NeuralPreset .

Olga Bogdanova

The article analyzes the novel of the modern Russian writer A. Potyomkin Man is canceled (2007), which received a wide public response. The main idea of the work is the need for a radical change of the “mass man” of the turn of the XX-XXI centuries at the psychosomatic level. The ideological and compositional center is the specially built Rimushkino estate in the Oryol province, where the serf spirit of the Russian Empire at the turn of the XVIII-XIX centuries is reproduced. To answer the question of why the estate space of Russia is becoming the most representative field for anthropological experiments of the beginning of the XXI century, we consider the estate neo-myths of the Silver Age (the lost paradise on earth) and the Soviet period (the camp hell living in the mentality), as well as the imperial-colonial concept of the postmodern era (the estate as a frontier in the process of class-oriented internal colonization of the country). The multidimensional semiotics of the estate sets a new relationship between “metropolitan” and “provincial” concerning the other loci of the novel.

Abhimanyu Sethia, Aayush

Much like most of cognition research, music cognition is an interdisciplinary field, which attempts to apply methods of cognitive science (neurological, computational and experimental) to understand the perception and process of composition of music. In this paper, we first motivate why music is relevant to cognitive scientists and give an overview of the approaches to computational modelling of music cognition. We then review literature on the various models of music perception, including non-computational models, computational non-cognitive models and computational cognitive models. Lastly, we review literature on modelling the creative behaviour and on computer systems capable of composing music. Since a lot of technical terms from music theory have been used, we have appended a list of relevant terms and their definitions at the end.

Zixuan Ye, Yutong Dai, Chaoyi Hong et al.

We study the composition style in deep image matting, a notion that characterizes a data generation flow on how to exploit limited foregrounds and random backgrounds to form a training dataset. Prior art executes this flow in a completely random manner by simply going through the foreground pool or by optionally combining two foregrounds before foreground-background composition. In this work, we first show that naive foreground combination can be problematic and therefore derive an alternative formulation to reasonably combine foregrounds. Our second contribution is an observation that matting performance can benefit from a certain occurrence frequency of combined foregrounds and their associated source foregrounds during training. Inspired by this, we introduce a novel composition style that binds the source and combined foregrounds in a definite triplet. In addition, we also find that different orders of foreground combination lead to different foreground patterns, which further inspires a quadruplet-based composition style. Results under controlled experiments on four matting baselines show that our composition styles outperform existing ones and invite consistent performance improvement on both composited and real-world datasets. Code is available at: https://github.com/coconuthust/composition_styles

Simon Moser, Yanxia Feng, Oncay Yasa et al.

Inspired by the cellular design of plant tissue, we present a new approach to make versatile, tough, highly water-swelling composites. We embed highly swelling hydrogel particles inside tough, water-permeable, elastomeric matrices. The resulting composites, which we call \emph{hydroelastomers}, show little softening as they swell, and have excellent fracture properties that match those of the best-performing, tough hydrogels. Our composites are straightforward to fabricate, based on commercial materials, and can easily be molded or extruded to form shapes with complex swelling geometries. Furthermore, there is a large design space available for making hydroelastomers, since one can use any hydrogel as the dispersed phase in the composite, including hydrogels with stimuli-responsiveness. These features should make hydroelastomers excellent candidates for use in soft robotics and swelling-based actuation, or as shape-morphing materials, while also being useful as hydrogel replacements in a wide range of other fields.

Simone Carati

Recensione di Babbi, Anna Maria e Comparini, Alberto (a cura di). Letteratura e altri saperi. Influssi, scambi, contaminazioni. Carocci, 2020.

Mathew Schwartz, Tomer Weiss, Esra Ataer-Cansizoglu et al.

Matching and recommending products is beneficial for both customers and companies. With the rapid increase in home goods e-commerce, there is an increasing demand for quantitative methods for providing such recommendations for millions of products. This approach is facilitated largely by online stores such as Amazon and Wayfair, in which the goal is to maximize overall sales. Instead of focusing on overall sales, we take a product design perspective, by employing big-data analysis for determining the design qualities of a highly recommended product. Specifically, we focus on the visual style compatibility of such products. We build off previous work which implemented a style-based similarity metric for thousands of furniture products. Using analysis and visualization, we extract attributes of furniture products that are highly compatible style-wise. We propose a designer in-the-loop workflow that mirrors methods of displaying similar products to consumers browsing e-commerce websites. Our findings are useful when designing new products, since they provide insight regarding what furniture will be strongly compatible across multiple styles, and hence, more likely to be recommended.

Yukai Shi, Sen Zhang, Chenxing Zhou et al.

Non-parallel text style transfer has attracted increasing research interests in recent years. Despite successes in transferring the style based on the encoder-decoder framework, current approaches still lack the ability to preserve the content and even logic of original sentences, mainly due to the large unconstrained model space or too simplified assumptions on latent embedding space. Since language itself is an intelligent product of humans with certain grammars and has a limited rule-based model space by its nature, relieving this problem requires reconciling the model capacity of deep neural networks with the intrinsic model constraints from human linguistic rules. To this end, we propose a method called Graph Transformer based Auto Encoder (GTAE), which models a sentence as a linguistic graph and performs feature extraction and style transfer at the graph level, to maximally retain the content and the linguistic structure of original sentences. Quantitative experiment results on three non-parallel text style transfer tasks show that our model outperforms state-of-the-art methods in content preservation, while achieving comparable performance on transfer accuracy and sentence naturalness.