World models provide a powerful framework for simulating environment dynamics conditioned on actions or instructions, enabling downstream tasks such as action planning or policy learning. Recent approaches leverage world models as learned simulators, but its application to decision-time planning remains computationally prohibitive for real-time control. A key bottleneck lies in latent representations: conventional tokenizers encode each observation into hundreds of tokens, making planning both slow and resource-intensive. To address this, we propose CompACT, a discrete tokenizer that compresses each observation into as few as 8 tokens, drastically reducing computational cost while preserving essential information for planning. An action-conditioned world model that occupies CompACT tokenizer achieves competitive planning performance with orders-of-magnitude faster planning, offering a practical step toward real-world deployment of world models.
Video generation models have rapidly progressed, positioning themselves as video world models capable of supporting decision-making applications like robotics and autonomous driving. However, current benchmarks fail to rigorously evaluate these claims, focusing only on general video quality, ignoring important factors to world models such as physics adherence. To bridge this gap, we propose WorldModelBench, a benchmark designed to evaluate the world modeling capabilities of video generation models in application-driven domains. WorldModelBench offers two key advantages: (1) Against to nuanced world modeling violations: By incorporating instruction-following and physics-adherence dimensions, WorldModelBench detects subtle violations, such as irregular changes in object size that breach the mass conservation law - issues overlooked by prior benchmarks. (2) Aligned with large-scale human preferences: We crowd-source 67K human labels to accurately measure 14 frontier models. Using our high-quality human labels, we further fine-tune an accurate judger to automate the evaluation procedure, achieving 8.6% higher average accuracy in predicting world modeling violations than GPT-4o with 2B parameters. In addition, we demonstrate that training to align human annotations by maximizing the rewards from the judger noticeably improve the world modeling capability. The website is available at https://worldmodelbench-team.github.io.
We introduce Matrix-Game, an interactive world foundation model for controllable game world generation. Matrix-Game is trained using a two-stage pipeline that first performs large-scale unlabeled pretraining for environment understanding, followed by action-labeled training for interactive video generation. To support this, we curate Matrix-Game-MC, a comprehensive Minecraft dataset comprising over 2,700 hours of unlabeled gameplay video clips and over 1,000 hours of high-quality labeled clips with fine-grained keyboard and mouse action annotations. Our model adopts a controllable image-to-world generation paradigm, conditioned on a reference image, motion context, and user actions. With over 17 billion parameters, Matrix-Game enables precise control over character actions and camera movements, while maintaining high visual quality and temporal coherence. To evaluate performance, we develop GameWorld Score, a unified benchmark measuring visual quality, temporal quality, action controllability, and physical rule understanding for Minecraft world generation. Extensive experiments show that Matrix-Game consistently outperforms prior open-source Minecraft world models (including Oasis and MineWorld) across all metrics, with particularly strong gains in controllability and physical consistency. Double-blind human evaluations further confirm the superiority of Matrix-Game, highlighting its ability to generate perceptually realistic and precisely controllable videos across diverse game scenarios. To facilitate future research on interactive image-to-world generation, we will open-source the Matrix-Game model weights and the GameWorld Score benchmark at https://github.com/SkyworkAI/Matrix-Game.
Designing adaptive mechanisms to align individual and collective interests remains a central challenge in artificial social intelligence. Existing methods often struggle with modeling heterogeneous agents possessing persistent latent traits (e.g., skills, preferences) and dealing with complex multi-agent system dynamics. These challenges are compounded by the critical need for high sample efficiency due to costly real-world interactions. World Models, by learning to predict environmental dynamics, offer a promising pathway to enhance mechanism design in heterogeneous and complex systems. In this paper, we introduce a novel method named SWM-AP (Social World Model-Augmented Mechanism Design Policy Learning), which learns a social world model hierarchically modeling agents' behavior to enhance mechanism design. Specifically, the social world model infers agents' traits from their interaction trajectories and learns a trait-based model to predict agents' responses to the deployed mechanisms. The mechanism design policy collects extensive training trajectories by interacting with the social world model, while concurrently inferring agents' traits online during real-world interactions to further boost policy learning efficiency. Experiments in diverse settings (tax policy design, team coordination, and facility location) demonstrate that SWM-AP outperforms established model-based and model-free RL baselines in cumulative rewards and sample efficiency.
Federico Baldassarre, Marc Szafraniec, Basile Terver
et al.
We present DINO-world, a powerful generalist video world model trained to predict future frames in the latent space of DINOv2. By leveraging a pre-trained image encoder and training a future predictor on a large-scale uncurated video dataset, DINO-world learns the temporal dynamics of diverse scenes, from driving and indoor scenes to simulated environments. We show that DINO-world outperforms previous models on a variety of video prediction benchmarks, e.g. segmentation and depth forecasting, and demonstrates strong understanding of intuitive physics. Furthermore, we show that it is possible to fine-tune the predictor on observation-action trajectories. The resulting action-conditioned world model can be used for planning by simulating candidate trajectories in latent space.
Alex Lewandowski, Adtiya A. Ramesh, Edan Meyer
et al.
Continual learning is often motivated by the idea, known as the big world hypothesis, that "the world is bigger" than the agent. Recent problem formulations capture this idea by explicitly constraining an agent relative to the environment. These constraints lead to solutions in which the agent continually adapts to best use its limited capacity, rather than converging to a fixed solution. However, explicit constraints can be ad hoc, difficult to incorporate, and may limit the effectiveness of scaling up the agent's capacity. In this paper, we characterize a problem setting in which an agent, regardless of its capacity, is constrained by being embedded in the environment. In particular, we introduce a computationally-embedded perspective that represents an embedded agent as an automaton simulated within a universal (formal) computer. Such an automaton is always constrained; we prove that it is equivalent to an agent that interacts with a partially observable Markov decision process over a countably infinite state-space. We propose an objective for this setting, which we call interactivity, that measures an agent's ability to continually adapt its behaviour by learning new predictions. We then develop a model-based reinforcement learning algorithm for interactivity-seeking, and use it to construct a synthetic problem to evaluate continual learning capability. Our results show that deep nonlinear networks struggle to sustain interactivity, whereas deep linear networks sustain higher interactivity as capacity increases.
We introduce the WorldScore benchmark, the first unified benchmark for world generation. We decompose world generation into a sequence of next-scene generation tasks with explicit camera trajectory-based layout specifications, enabling unified evaluation of diverse approaches from 3D and 4D scene generation to video generation models. The WorldScore benchmark encompasses a curated dataset of 3,000 test examples that span diverse worlds: static and dynamic, indoor and outdoor, photorealistic and stylized. The WorldScore metrics evaluate generated worlds through three key aspects: controllability, quality, and dynamics. Through extensive evaluation of 19 representative models, including both open-source and closed-source ones, we reveal key insights and challenges for each category of models. Our dataset, evaluation code, and leaderboard can be found at https://haoyi-duan.github.io/WorldScore/
What are the essential components of world models? How far do we get with world models that are not employing RNNs, transformers, discrete representations, and image reconstructions? This paper introduces SGF, a Simple, Good, and Fast world model that uses self-supervised representation learning, captures short-time dependencies through frame and action stacking, and enhances robustness against model errors through data augmentation. We extensively discuss SGF's connections to established world models, evaluate the building blocks in ablation studies, and demonstrate good performance through quantitative comparisons on the Atari 100k benchmark.
The year 2024 sees Lesotho celebrate 200 years since the formation of Basotho as a nation. In these 200 years, the church has played a significant role in nation building and development. This article explores and seeks to locate the voice and presence of the church both as an ecumenical body and as a prophetic voice within the tumultuous history of Lesotho’s politics with a specific timeframe from 1833-2002. A literature review of relevant academic articles as well as historical works is used to ensure that the accounts recorded are factually correct. The formation of Basotho in the 18th century is a direct result of violence and war in the form of Lifaqane. King Moshoeshoe I, founder of the Basotho nation believed that peace could be procured via the presence of missionaries, therefore denoting the establishment of the church as an important factor. Within Lesotho’s mainstream history, the role played by the church has often been referred to in passing. Common sayings like “Baruti ba Moshoeshoe (Moshoeshoe’s pastors/ministers), and Lesotho ke naha ea Bokresete (Lesotho is a Christian nation)” are often quoted, and yet the actual role that the church has played and continues to play in Lesotho’s history and development has been an after-thought. In light of this, this article locates the role that the church has played in different eras of Lesotho’s history. From the establishment of the 1 st church (Protestant Church) in Morija, to the formation of ecumenical bodies such as The Heads of the Churches and Church Council of Lesotho to the current emerging coalition democracies under the constitutional monarchy.
Abrahamic faiths, which are among the divine revelations, consider science to be the knowledge of the universe and humans. The proper understanding of science is the study of the natural world through the lens of religious metaphysics which takes into account all relevant factors from a religious viewpoint. This study aims to survey the effect of religious worldview on the metaphysical principles that govern science, and deals with the following questions:1-What effect do religious metaphysical foundations have on the interpretation of scientific theories?2-How scientific work, in the light of the divine worldview, ensures the material and spiritual needs of mankind?3-Is experimental science influenced by religious and philosophical values and insights?4-Are various human and natural sciences far from value judgments?In this paper, we use the fact that all scientific theories are based on some metaphysical assumptions which are not taken from these sciences but are rooted in philosophies or religions, and provide a worldview for the scientist . Furthermore, these worldviews affect scientists’ motivations and their choices of theories. We argue that the worldview of monotheistic religions puts the totality of scientific issues in the context of religious metaphysics.
Philosophy of religion. Psychology of religion. Religion in relation to other subjects
World models are progressively being employed across diverse fields, extending from basic environment simulation to complex scenario construction. However, existing models are mainly trained on domain-specific states and actions, and confined to single-modality state representations. In this paper, We introduce WorldGPT, a generalist world model built upon Multimodal Large Language Model (MLLM). WorldGPT acquires an understanding of world dynamics through analyzing millions of videos across various domains. To further enhance WorldGPT's capability in specialized scenarios and long-term tasks, we have integrated it with a novel cognitive architecture that combines memory offloading, knowledge retrieval, and context reflection. As for evaluation, we build WorldNet, a multimodal state transition prediction benchmark encompassing varied real-life scenarios. Conducting evaluations on WorldNet directly demonstrates WorldGPT's capability to accurately model state transition patterns, affirming its effectiveness in understanding and predicting the dynamics of complex scenarios. We further explore WorldGPT's emerging potential in serving as a world simulator, helping multimodal agents generalize to unfamiliar domains through efficiently synthesising multimodal instruction instances which are proved to be as reliable as authentic data for fine-tuning purposes. The project is available on \url{https://github.com/DCDmllm/WorldGPT}.
This article examines and explores the importance of interreligious dialogue within the multireligious and multicultural communities. The dialogical tool to be used in this endeavor is Comparative Theology. Through the literature review, the scenario of religiously diverse
communities, especially in South Africa, is painted both historically and sociologically. Comparative Theology is defined and historically identified. Theology of dialogue is brought forth as a method that Comparative Theology can use towards social harmony in pluralistic South Africa. Social harmony as a sociological anticipation is defined and related to theology. Social harmony is presented as the form of collective philosophy, which means working together for the greater good, and putting the societal or organisational needs above and ahead of the needs of an individual. The intention is not to convert but to understand and see how harmonious coexistence can be initiated and achieved. Interreligious dialogue is not an initiative of weighing error from truth. It is an engagement with the other, and as dialogical partners coming from
different traditions, Comparative Theologians provide guidelines towards this dialogue. Reflective suggestions are highlighted, and these are networking and partnership development for human survival. Christ’s example of crossing the cultural and religious barriers is elaborated as a model to follow. Contact without contamination is the driving force. Disengagement and personal insulation are not an ideal option. Christian theologians are encouraged to engage dialogically with people of different religions, as dialogue is the better option towards understanding of the other.
World models are a powerful tool for developing intelligent agents. By predicting the outcome of a sequence of actions, world models enable policies to be optimised via on-policy reinforcement learning (RL) using synthetic data, i.e. in "in imagination". Existing world models are autoregressive in that they interleave predicting the next state with sampling the next action from the policy. Prediction error inevitably compounds as the trajectory length grows. In this work, we propose a novel world modelling approach that is not autoregressive and generates entire on-policy trajectories in a single pass through a diffusion model. Our approach, Policy-Guided Trajectory Diffusion (PolyGRAD), leverages a denoising model in addition to the gradient of the action distribution of the policy to diffuse a trajectory of initially random states and actions into an on-policy synthetic trajectory. We analyse the connections between PolyGRAD, score-based generative models, and classifier-guided diffusion models. Our results demonstrate that PolyGRAD outperforms state-of-the-art baselines in terms of trajectory prediction error for short trajectories, with the exception of autoregressive diffusion. For short trajectories, PolyGRAD obtains similar errors to autoregressive diffusion, but with lower computational requirements. For long trajectories, PolyGRAD obtains comparable performance to baselines. Our experiments demonstrate that PolyGRAD enables performant policies to be trained via on-policy RL in imagination for MuJoCo continuous control domains. Thus, PolyGRAD introduces a new paradigm for accurate on-policy world modelling without autoregressive sampling.
One of the biggest challenges to modern deep reinforcement learning (DRL) algorithms is sample efficiency. Many approaches learn a world model in order to train an agent entirely in imagination, eliminating the need for direct environment interaction during training. However, these methods often suffer from either a lack of imagination accuracy, exploration capabilities, or runtime efficiency. We propose Hieros, a hierarchical policy that learns time abstracted world representations and imagines trajectories at multiple time scales in latent space. Hieros uses an S5 layer-based world model, which predicts next world states in parallel during training and iteratively during environment interaction. Due to the special properties of S5 layers, our method can train in parallel and predict next world states iteratively during imagination. This allows for more efficient training than RNN-based world models and more efficient imagination than Transformer-based world models. We show that our approach outperforms the state of the art in terms of mean and median normalized human score on the Atari 100k benchmark, and that our proposed world model is able to predict complex dynamics very accurately. We also show that Hieros displays superior exploration capabilities compared to existing approaches.
Śrī Nāth is one of the most important images of Krishna being worshipped at the temple of Nathdwara in Rajasthan. His devotees consider him to be a living god, he appears in their dreams, and according to their sayings they are in direct contact with him. Śrī Nāth, originally a local deity, is equated with the major Hindu god Krishna. However, while Krishna may be one of the most important gods in India, he is also ambiguous through his acts and words, if not bluntly unreliable. This double nature of Krishna is reflected in the cult of Śrī Nāth. There is an interesting interaction between Śrī Nāth (implying Krishna himself), the main gurus of his cult, i.e., Vallabha (Vallabhācārya) and the latter’s son and main successor Viṭṭhalnāth and his devoted disciples. At times, Śrī Nāth feels the need to stick to the official Brahmanical cult of the temple rituals, on other occasions, there is no problem in transgressing any given official rule. The same is true for the primary teachers, who are often put on par with Krishna himself or one of the celestials closely connected to him. Additionally, the disciples can apparently do anything in their frenzies. All of this reinforces the idea that this entire cult belongs to another world (alaukik). It is part of the everyday world (laukik) of Hindu India, but meanwhile, each and every rule can be ignored if the supernatural breaks through. Even the distinction between Hinduism and Islam at times simply does not seem to be of importance anymore. Muslims can become addicted to the passionate love for Krishna through the form of Śrī Nāth, so it is sometimes stated. Each and every partaker in the cult may share the visions of the initiated devotee, at times even without proper initiation. This all adds to the experience of the supermundane and supernatural in this particular cult.
A family is a basic building block of society created by God and it is crucial for a healthy state and church. In traditional African societies, a family is responsible for the spiritual, emotional, moral, intellectual, and physical empowerment of the child. Parents play a significant role in transmitting knowledge and skills for the survival of the child in their own context. Other structures in society educate the children based on the foundation laid by the family. This article proposes that a family should not be ignored in the formation of ministerial priesthood in the Roman Catholic Church (RCC) seminary. The delay in involving a family hinders the holistic development of the priest candidate. Involvement of the family in the process of formation challenges the relationship between a seminary and a family and encourages an imbalance of male and female formators in a seminary. This article aims to highlight the significance of the family as an indispensable body in the formation process of an African priest candidate, particularly for the RCC in South Africa.
In public health history, vaccinations have been seen as very helpful as they help protect societies from preventable diseases and save many lives around the world. Despite the fact that few religious groups oppose vaccination for a variety of reasons, almost all major religions in the world support vaccination as long as it complies with religious precepts for the preservation, protection, or well-being of livelihoods. However, during the surge of Covid-19
outbreak in South Africa and elsewhere, governments witnessed unprecedented hesitancy and opposition to the Covid-19 vaccination by the general public. Those who opposed the Covid-19 vaccine cited multiple concerns or reasons, ranging from possible side effects, adverse events, vaccine safety, vaccine effectiveness, conspiracy theories, and religious or
cultural reasons. Based on a literature search, this paper attempts to discuss various religious views on the subject of vaccination in general and subsequently make use of such perspectives to support calls for a compulsory Covid-19 vaccination, especially for South African churches. This is a timely topic of central concern as it seeks to politely dispel religious
misunderstandings and confusion that could arise as a result of debates on mandatory vaccines for Covid-19. Health and religion will be brought together to shape the ongoing discussions about the moral urgency of a mandatory Covid-19 vaccination in South Africa.
Acre is a port city in the north-western part of Israel, with a history that goes back more than 4000 years. Being inscribed on the World Heritage List, the Old City of Acre preserves the urban and architectural elements of a historic town. Its outstanding value relies on the Crusader remnants preserved under the Ottoman city, showcasing the dynamism and continuous change of Mediterranean port cities. Moreover, the presence of various religions: Judaism, Christianity, Islam, and Bahai, adds to its complexity, expressed as monuments and religious sites that enrichen Acre’s cultural heritage. The dramatic change in values over the past decades has a direct impact on the built environment and the citizen’s lifestyles, in some cases jeopardising the physical elements and drastically influencing people’s lives. This paper aims to analyse the changes linked to the sea: livelihoods, tourism, and recreational use; and the change of use of the khan, as both the sea and the khan are constant elements in the city. The analysis of these processes serves as the starting point to identify changes in values which can enhance development or promote gentrification, and in the case of Khan Al-Umdan and its vicinity, we aim to recognise the lights and shadows that followed the adaptive reuse evaluation procedure, and the influence of the multiple narratives in its development. The conclusions will provide a solid base on which to develop a methodology on the one hand, identify changing processes, such as gentrification; and on the other, to evaluate adaptive reuse alternatives of cultural heritage in contested societies and changing values.
ABSTRACT: There are very few sources, other than material remains and spatial arrangements revealed by archaeological excavation, that can give modern researchers access to the thought-world of pre-Christian Scandinavian religion. Some skaldic poetry presumed to have been composed before the Conversion may offer a window onto this thought-world. This article investigates how a single kenning from a stanza composed by the tenth-century Icelander Egill Skallagrímsson conceptualised the relationship between the dominant Viking-Age deity Óðinn and the conduct of religious ritual.
RESUME: Der findes meget få kilder, bortset fra materiel kultur og spatiale organiseringer afdækket gennem arkæologiske udgravning, der kan give moderne forskere adgang til førkristen nordisk religions tankeverden. Nogle skjaldedigte, som antages at være komponeret før konverteringen til Kristendom, kan give et indblik i denne tankeverden. Denne artikel undersøger hvordan en enkelt kenning fra en strofe komponeret af islændingen Egill Skallagrímsson i det tiende århundrede konceptualiserer forholdet mellem den dominerende vikingetidsguddom, Óðinn, og udførelsen af religiøse ritualer
The high rate of divorce among both young and older couples is on an increase with several unhappy families living in frustration and uncertainty. This is tied to their unpreparedness before going into marriage and their incompatibility as destiny may not permit the union of both couples in the first place. Though, physical or psychological counselling are important
these have not proved effective enough to solve marital problems. The spiritual dimension of views, the unseen, as a way of guidance before going into marriage has becomes imperative in solving marital situations that have atrophied the family settings and the society at large. It is on this basis, that the essence of Ifa divination in enquiring into the fate of a couple coming
together as one in holy matrimony cannot be undervalued. In other words, it is a structural foundation towards building a happy marriage. The Christian Orthodox worldview on counselling is also touched upon as a means of comparison with the oldest Christian denomination. The spiritual compatibility of the couple is important, and far more than the
physical understanding of blood group genotype, temperament, sexual compatibility, communication skill etc. This paper, therefore, aims at establishing the place of Ifa divination in a couple’s pre-marital counselling, as an important requirement before going into marriage. Twenty (20) Ifa priests were interviewed across southwestern Nigeria. Collections of
information were content analysed, resulting in a recommendation for spiritual enquiries into the compatibility of couples before going into marriage