This article investigates the evolution of divinatory practices in Korea from the premodern era to the contemporary period, focusing on the shift from collective rituals to individualized forms of esoteric practice. Drawing on historical, anthropological, and sociological sources, it examines the Jeomchal ceremony and sojae doryang rituals of the Silla and Goryeo periods, ritual adaptations in the Joseon era, and modern practices with particular focus on Tarot readings. The study employs a theoretical framework integrating esotericism and social theories to analyze how engagement in divinatory rituals reflects broader social and cultural transformations. Findings indicate that, while early divinatory practices were largely collective and aimed at societal or karmic balance, contemporary forms prioritize personal insight, autonomy, and psychological well-being. Despite this individualization, modern practices maintain subtle forms of social connection through shared symbolic frameworks and digital networks. The article argues that the historical tendency of Korean divination illustrates a dynamic interplay between continuity and change, collective and individual orientations, and traditional and globalized practices. This analysis contributes to Korean studies by situating divination within a broader socio-cultural and historical perspective, offering insights into the persistence and adaptation of esoteric practices in the face of modernization and social challenges.
This study investigates the translation and recognition of Han Kangs' literary works in the Chinese context. It traces the chronological development of Chinese translations of her writings since 2007. Combining close textual reading, media analysis, and reader-based data mining, the research further explores how Han Kang's image as a "feminist writer" has been constructed, reinforced, and contested across literary, academic, and popular discourses in China. While gender-based interpretations dominate, this study argues that this framing risks oversimplifying the philosophical and ethical complexity of Han Kangs' work, particularly her engagement with themes such as trauma, historical violence, and the human condition. Through an in-depth analysis of her later novels, especially We Do Not Part, this study reveals the multidimensionality of Han Kang's literary vision, showing how her use of bodily imagery, natural symbols, and doubled narrative voices bridges personal memory and collective trauma. Finally, the study reflects on the mechanisms of global literary dissemination, drawing attention to the central role of translation institutions, translator agency, and reception dynamics in shaping crosscultural literary recognition.
This study examines the political status and identities of sources and responsibility attributions of fake news in media articles covering the two recent Korean presidential elections, which saw the election of candidates of different political orientations. Results of quantitative content analysis for 746 stories from four nationwide news outlets showed that power ownership (ruling vs. opposing), rather than political orientation (conservative vs. progressive), better explained the variations in source appearance. In general, regardless of ideological orientation, sources associated with ruling parties blamed fake news more than those from opposing parties. The presidents and nominatives from the ruling party, as well as electives from opposing parties, were major sources blaming fake news. Experts, media, and the private sector were occasionally found to be sources blaming fake news from neutral zones. In terms of responsibility attribution for fake news, while unidentified targets blamed for fake news outnumbered any other identified objects, the president, nominatives, and electives were more frequently blamed when sources were from opposing parties. Theoretical and practical applications of the study results are then discussed to better understand the nature of fake news.
This article examines the 1952 WHO/UNKRA health planning mission in South Korea within the broader framework of United Nations-led reconstruction after the Korean War. Three medical professionals-George national health and medical conditions. They delivered a comprehensive system. The report detailed important challenges and offered potential solutions encompassing all aspects of rehabilitation in the health and medical system, which were closely tied to the developmental project of Western intervention in newly independent countries after World War II. This article argues the WHO/UNKRA mission was an exercise in making society legible to the state, that is, rendering it measurable and manageable as a precondition for intervention. Situating the mission and its proposals within postwar development strategies, this article shows that meaningful state involvement required first creating standardized, visible units of administration and knowledge. These included building a medical administrative apparatus, training scientific and technical personnel, and establishing mechanisms for monitoring and managing the broader population.
The Korean Cheontae order presents an apparent tension: though founded as a new Buddhist movement in the post-liberation period, it has successfully established itself as the legitimate heir to the historical Tiantai/Cheontae tradition. This article examines how the order navigated this tension by constructing historical legitimacy through transnational exchange with Chinese and Japanese Tiantai communities, reinterpreting its central practice of Guanyin prayer in light of Tiantai doctrine and practice, and re-transmitting its tradition to China. The Korean Cheontae order is one of the most successful new Buddhist movements to emerge following Korean independence in 1945. This study analyzes how the order appropriated the name "Cheontae" by defining its relationship to that tradition and specifying which elements of Tiantai thought and practice it claims to inherit, even as discrepancies between the order's official narrative and internal sources complicate that claim. The findings reveal how a postcolonial Korean Buddhist organization has adapted and transformed Chinese Buddhist traditions through importation, reinterpretation, and re-transmission, shedding light on what, according to the order itself, makes this movement meaningfully "Cheontae."
This study explores how Korean Seon Buddhism has adapted its meditation practices in response to modern influences, particularly through the integration of mindfulness-based approaches within the Templestay program. While rooted in traditional Ganhwa Seon, which emphasizes hwadu contemplation for self-inquiry and master-disciple transmission, contemporary Seon practice has evolved to accommodate a broader audience, including international and secular practitioners. However, the push for greater accessibility in Seon meditation has sparked debates about whether these modern adaptations dilute traditional Seon training or make it more relevant to contemporary practitioners. Focusing on English conducted Templestay programs targeting foreigners, this study aims to analyze how these adaptations shape the experience of Seon meditation and make it accessible to a global audience. Through first-hand experiences at four Korean Buddhist temples offering Templestay programs-Hwagyesa, Golgulsa, Lotus Lantern International Meditation Center, and the International Seon Center-the study examines the distinct meditation approaches used, ranging from hwadu practice to Seonmudo, breath awareness, and mantra meditation. The findings suggest that rather than homogenizing with global mindfulness, Templestay programs constitute sites of pragmatic negotiation, in which the degree of adherence to traditional Ganhwa Seon varies significantly according to the specific institutional history and pedagogical orientation of each temple.
The Seokjeon ceremony (Confucian Memorial Rite), a major Confucian ritual since the Joseon period, has long carried significant symbolic and socio-political weight. Its form changed considerably across Sungkyunkwan and local hyanggyo under modernization, colonial rule, and postwar political change. Under colonial rule, Confucianism was deliberately removed from the category of religion and its functions reduced to that of education, thus turning it into an instrument for promoting imperial ideology. Although Sungkyunkwan regained its formal status after liberation, disputes over officiating the Seokjeon ceremony became a key arena for competing Confucian factions. Even amid this turbulence, Confucian groups actively sought to preserve the ritual's religious character, most notably through prolonged debates over shifting its date between the lunar and solar calendars-an effort to reaffirm its spiritual legitimacy in the face of Christianitys' growing influence. In recent decades, however, the ceremony has come increasingly to reflect broader East Asian trends that reframe Confucian rituals as cultural rather than religious practices. Achieving heritage designation in 1986, the Seokjeon ceremony today exemplifies the transformation of Confucianism from a lived moral-religious system into cultural tradition and national heritage.
Radiance fields, including their recent efficient forms such as 3D Gaussian Splatting and Sparse Voxels, have revolutionized photorealistic 3D scene visualization by enabling high-fidelity reconstruction of complex environments, making them a natural match for light field displays. However, integrating these technologies presents significant computational challenges, as light field displays require many high-resolution renderings from slightly shifted viewpoints, while radiance fields rely on computationally intensive volume rendering, which is intractable to achieve real-time speeds even with efficient scene representations. In this paper, we propose a unified and efficient framework for real-time radiance field rendering on light field displays. Rather than re-rendering each view independently, our method converts the input radiance field into shared intermediate sweeping planes that can be efficiently composited into dense light-field views in a single pass. Our method prioritizes shared, non-directional plane caching for real-time performance, trading fine view-dependent color effects for a modest increase in intermediate memory usage. Our framework generalizes across different scene representations without retraining and avoids repeated computation across views. We further demonstrate a real-time interactive application on a Looking Glass display, achieving 200+ FPS at 512p across 45 rendered views and enabling seamless, immersive 3D interactive viewing experiences. On standard benchmarks, our method achieves up to 22× speedup compared to independently rendering each view, while largely preserving image quality.
Machine Civilization is a multi-screen mechanical installation that imagines an autonomous system observing the world through diverse machine perspectives and generating scripts independent of human semantic frameworks. The installation implements a real-time system of observation, interpretation, collective deliberation, and script inscription, producing formal familiarity and semantic untranslatability. Taking this artwork as a case, this study starts from machines’ autonomous cognitive features and takes “machine writing” as its entry point to explore three issues in the intelligence era: the gap between interpretability and understandability, the flow of meaning between humans and machines, and the ethical encounter with the machine as the Other. Inspired by Levinas’s concept of alterity, the work challenges anthropocentric perspectives by emphasizing machine cognition’s autonomy beyond human semantics. This study contributes to expanding the epistemological and ethical discussion on non-human intelligence through artistic practice, and exploring more open modes of interaction with the technological Other.
3D Aesthetics is significant in digital design, shaping how users experience real-time 3D content in games, VR, and product design. However, creating aesthetically pleasing shapes remains challenging due to diverse subjective standards and the lack of tools that support aesthetics-driven editing. Users often rely on intuition without explicit guidance on visual appeal, making aesthetics refinement slow, inconsistent, and cognitively demanding, particularly in fast-paced, iterative workflows. To address this challenge, we conducted in-depth interviews with design experts to identify challenges in aesthetics-oriented modeling workflows. Based on the findings, we developed Aesthetic3D , a 3D modeling interface that provides real-time aesthetics scores learned from human perceptual data. Furthermore, Aesthetic3D seamlessly integrates the learned aesthetics measures into intuitive editing operations, enabling aesthetics-driven exploration and refinement of shape geometry. We evaluated Aesthetic3D through an ablation study, an open-ended study, and three generalization evaluations. Comprehensive experiments show that with Aesthetic3D , users can easily and effectively enhance the aesthetics appeal of 3D shapes.
The application of machine learning techniques to 3D Gaussian Splatting (3DGS) has enabled high-fidelity interactive rendering of complex 3D scenes on modest compute resources, offering significant potential for scientific visualization. However, the memory demands of reconstructing large-scale scientific volumes at high fidelity exceed the capacity of individual GPUs. While distributed 3DGS frameworks exist for urban scenes, they rely on global scene analysis or frequent all-to-all communication, inherently limiting their scalability on High-Performance Computing (HPC) clusters. In this paper, we present a scalable distributed 3DGS training and rendering framework designed for large-scale scientific volumetric data. Leveraging the native domain decomposition of HPC simulations, our approach treats each spatial partition as an independent optimization task that avoids device-to-device communication. We further exploit temporal coherence in simulation data to accelerate training on subsequent timesteps. We perform experimental studies demonstrating that our approach achieves near-ideal weak scaling by eliminating synchronization overhead. Additionally, we evaluate the trade-offs of the standard 3DGS optimization strategy against a Markov Chain Monte Carlo (MCMC) approach. We find that while standard 3DGS produces higher quality reconstructions with larger Gaussian footprints, MCMC effectively bounds the primitive count at scale in exchange for a minor reduction in fidelity. Furthermore, we demonstrate that our temporal fine-tuning strategy for time-series data remains robust at scale, establishing a reliable path towards in situ applications of 3DGS for massive scientific simulations.
Understanding how humans interact with charts is crucial for designing effective visualization systems. While identifying a user’s task from gaze is fundamental, traditional methods rely on labor-intensive feature engineering, showing limited performance and high task-specificity. In this work, we introduce a framework for MLLM-based gaze-to-task inference, including an automatic few-shot sample generation process that creates structured demonstrations for in-context learning. We present the first systematic investigation into gaze-based task inference on charts, benchmarking five MLLMs and rigorously exploring gaze encoding and prompting strategies to establish optimal design principles. Our findings identify the heatmap representation as the optimal visual gaze encoding, and we demonstrate the necessity of Chain-of-Thought prompting. Notably, MLLMs can autonomously decode cognitive intent without manual AOI definitions, exceeding traditional baseline performance. This study offers actionable insights for integrating human gaze into MLLMs, guiding the design of future systems that adapt to a user’s analytical focus.
Implicit surfaces offer distinct advantages over traditional boundary representations, including infinite resolution, low memory footprint, smooth geometry by construction, and support for non-destructive modeling. In this work, we introduce a method for localizing geometric detail in a way that preserves the mathematical properties required for accurate and efficient rendering using sphere tracing. Our contributions include novel procedural modeling techniques that expand the range of repetition patterns achievable in implicit surfaces; an interpolation-based approach that maintains field correctness while remaining computationally efficient; and a cache-based acceleration strategy that significantly improves the rendering performance of domain-repeated implicit geometries.
Welcome to the 2026 I3D PACMCGIT issue. PACMCGIT is a specialized journal in the collection of Proceedings of the ACM (PACM) that is specifically focused on publishing research of the highest quality within the broad domains of computer graphics and interactive techniques. This issue is a collection of the papers selected and presented at the 30th ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games (I3D) 2026. Continuing the tradition, the I3D collection of papers constitutes the first PACMCGIT issue of the year, this time marking the beginning of PACMCGIT’s ninth year presenting the publications of I3D. I3D 2026 received 41 submissions. Following a rigorous double-blind peer-reviewing process with 122 reviews, including a second review cycle by a panel of experts from the I3D 2026 international program committee, 16 papers were accepted to be presented at the conference, resulting in an acceptance rate of \(39\%\) . This year, one paper was desk-rejected as it was clearly incomplete. Paper topics span various exciting areas, from physical simulation and animation to real-time rendering, ray-tracing, VR, and neural rendering. These papers appeared in talk presentations at the ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games (I3D) in May 2026 at Lucasfilm in San Francisco (CA).
The generation of consistent multi-view images is essential for diffusion-based 3D texture synthesis, a key component of modern 3D content pipelines. However, existing methods remain weakly coupled across viewpoints, often exhibiting inconsistencies between views that result in artifacts such as ghosting, misalignment, and texture distortions. We introduce PixelSync, a training-free mechanism that explicitly enforces multi-view consistency during the diffusion process. Our method operates in two stages. First, we compute visibility-aware pixel correspondences between known views in a fast pre-computation step by utilizing geometric information obtained from a given 3D model. Then, during the diffusion process, we enforce consistency through two complementary algorithms: (i) attention synchronization, by modifying the attention scores in the U-Net’s down- and mid-blocks, and (ii) latent synchronization, by progressively aligning features on their corresponding shared pixels. PixelSync can be seamlessly integrated into pre-trained diffusion models without retraining, offering a lightweight and effective solution. We demonstrate the effectiveness of our approach for generative texture synthesis by plugging PixelSync into established multi-view diffusion backbones, showcasing reduced cross-view artifacts and improved global coherence across standard quantitative and qualitative evaluations.
Eye-Tracking (ET) has recently gained popularity in Recommender Systems (RS) as a source of implicit feedback, effectively linking user attention to product saliency. However, leveraging user visual behaviour for recommendations does not scale due to the large volumes of ET data required to train RS models and the need for specialized equipment. This work investigates the extent to which state-of-the-art ET generative models can effectively mimic user information foraging behaviour in query-driven, search-recommendation tasks, as well as augment organic datasets with synthetic data for use in ET-based RS. We benchmark saliency and scan-path generators under task context, evaluating aggregate and individual saliency metrics and testing whether synthetic ET preserves category-sensitive gaze differences observed in humans. Our findings demonstrate that synthetic ET reproduces aggregate attention but fails to capture the search dynamics observed in organic data. Individual-level predictive accuracy remains low, with moderate improvements from fine-tuning and leave-one-out training, indicating data limitations.
This paper addresses the problem of unpaired shape translation on 3D point clouds. While prior methods typically rely on global latent vectors or spatially structured grids, such representations often lack the flexibility to capture both semantic structures and fine-grained geometric details. To address this, we propose operating directly in a structured token space, where tokens are pretrained through masked autoencoding. Unlike rigid spatial grids that impose fixed layouts, our tokens naturally adapt to geometric variations while maintaining semantic coherence. This structured yet flexible latent space enables semantically meaningful and geometrically precise transformations. A transformer-based translator is proposed to manipulate these tokens. This gated dual-branch translator enables detail-preserving and topology-aware shape translation across categories. Experiments on challenging tasks, such as chair-to-table transformations, demonstrate that our approach outperforms existing methods in preserving both global structure and part-level details.
We introduce Canvas3D, a novel framework for learning an editable 3D instance field from sparse 2D observations. Our approach utilizes depth information from observed views to warp sparse inputs into previously unobserved viewpoints. Each missing view is warped from the two nearest observed views in different directions, effectively filling occlusion-caused gaps. These warped views significantly enhance the learning of a geometrically precise 3D instance field by leveraging multi-view consistency. Additionally, we propose an effective method for manipulating the learned 3D instance field through 2D interactions in a geometrically coherent manner. This allows for object-level manipulation in 3D space by editing 2D instance maps from arbitrary viewpoints. The instance field also serves as an intermediate representation for 3D-aware editable image synthesis. This bypasses the complexities of direct 3D manipulation by editing a robust instance field before translating it into consistent images using off-the-shelf semantic image synthesis models. Extensive experiments demonstrate that our method accurately reconstructs a 3D instance field from sparse 2D observations and enables 2D-driven object-level manipulation of the field. We also explore the potential of our method in 3D-aware controllable image synthesis, showcasing synthesis results guided by edited semantic maps.
Algorithms leveraging ReSTIR-style spatiotemporal reuse have recently proliferated, hugely increasing effective sample count for light transport in real-time ray and path tracers. Many papers have explored novel theoretical improvements, but algorithmic improvements and engineering insights toward optimal implementation have largely been neglected. We demonstrate enhancements to ReSTIR PT that make it 2-3 & times; faster, decrease both visual and numerical error, and improve its robustness, making it closer to production-ready. We halve the spatial reuse cost by reciprocal neighbor selection, robustify shift mappings with new footprint-based reconnection criteria, and reduce spatiotemporal correlation with duplication maps. We further improve both performance and quality by extensive optimization, unifying direct and global illumination into the same reservoirs, and utilizing existing techniques for color noise and disocclusion noise reduction.
