arXiv:2607.05303v1 Announce Type: new Abstract: At equilibrium, the ratio of two steady-state probabilities is a Boltzmann factor, set by a free-energy difference. Such ratios are the natural, normalization-independent readouts of both thermodynamics and information processing, and are often used as a measure of fidelity in biophysical systems. What becomes of these ratios once a system is driven from equilibrium, where the Boltzmann factor no longer holds? Representing Markov processes as graphs and their steady states as averages over a distribution on spanning trees, the \emph{arboreal distribution}, we track the ratio $\pi_i/\pi_j$ under driving along \emph{energetic edges}, where detailed balance is broken, relative to its equilibrium value. Our central finding is that a chosen ratio can stay exactly locked to its equilibrium value arbitrarily far from equilibrium, a phenomenon we call \emph{energetic protection}, whenever an algebraic equality between spanning-tree weights holds. Just as detailed balance constrains rates around cycles, energetic protection constrains weights across trees, providing a new mechanism for robustness against fluctuations in the driving force, such as variations in ATP concentration. Away from this equality, single-edge driving collapses the response onto two arboreal coefficients and forces it to be monotonic, so nonmonotonic single-edge control is impossible at any strength. With two energetic edges, protection and monotonicity combine into a \emph{thermodynamic switch} that holds a function at its equilibrium value for as long as desired and releases it sharply. Equilibrium is known for the limits it places on information processing. We show that new constraints, both no-go principles and exact invariances, survive far from equilibrium. These results reveal how the localization of energy expenditure governs the functional logic of nonequilibrium systems in physics and biology.
Science Journals
arXiv:2607.04428v1 Announce Type: new Abstract: Diffusion large language models (dLLMs) generate text by iteratively denoising a masked sequence, offering a parallel alternative to autoregressive models, but eliciting strong reasoning through post-training remains difficult: supervised fine-tuning is off-policy and suffers from exposure bias, while reinforcement learning gives only sparse, sequence-level rewards and is hard to apply without tractable sequence likelihoods. On-policy self-distillation (OPSD) offers a promising alternative, using one model as both student and teacher to provide dense, token-level, on-policy supervision, but its effectiveness hinges on giving the teacher privileged information (PI) - typically an instance-specific ground-truth reference unavailable at inference - so the student ends up distilling a weak PI-free consensus policy that yields little improvement on dLLM reasoning. We introduce dOPSD, which instead derives the teacher's privilege directly from the student's own denoising trajectory, evaluating masked positions using later, more-decoded steps of that same trajectory rather than an external label, so the teacher's advantage emerges from the model's own decoding process; on Dream and LLaDA, dOPSD improves both in-domain math reasoning and out-of-domain code generation, outperforming supervised and on-policy baselines.
arXiv:2607.05362v1 Announce Type: new Abstract: We study the problem of reconfiguring a crossing-free embedding of a graph on a surface, with edges represented as curves, into another crossing-free embedding of the same graph on the same surface with the same fixed vertex positions. In this process, we reroute one edge at a time while maintaining crossing-free intermediate embeddings. This problem was introduced by Ito et al. [TALG 2025], who showed that even if the graph is a matching of two edges, reconfiguration is not always possible in the plane, but is always possible on the torus. For matchings of two or more edges, they gave a necessary and sufficient condition for reconfigurable embeddings in the plane, but not on the torus. Our main result is that for matchings, trees and forests, reconfiguration is always possible on the torus, and consequently, on any orientable surface of genus at least one. In addition, we provide sufficient conditions for reconfiguration on orientable surfaces of genus at least one and in the projective plane. For more general graphs, we show that reconfiguration is not always possible.
arXiv:2509.21541v3 Announce Type: replace Abstract: Hair simulation and rendering are challenging due to complex strand dynamics, diverse material properties, and intricate light-hair interactions. Recent video diffusion models can generate high-quality videos, but they lack fine-grained control over hair dynamics. We present ControlHair, a hybrid framework that integrates a physics simulator with conditional video diffusion to enable precise and controllable dynamic hair rendering. ControlHair adopts a three-stage pipeline: it first encodes physics conditions into per-frame geometry using a simulator, then extracts per-frame control signals, and finally feeds control signals into a video diffusion model to generate videos with desired hair dynamics. This cascaded design decouples physics reasoning from video generation, supports diverse physics, and makes training the video diffusion model easy. Trained on a curated 10K video dataset, ControlHair outperforms text- and pose-conditioned baselines, delivering precisely controlled hair dynamics. We also demonstrate three use cases of ControlHair, including dynamic hairstyle try-on, bullet-time effects, and cinemagraphic. Project page: https://linwk20.github.io/controlhair-web.
arXiv:2509.26306v5 Announce Type: replace Abstract: Existing multi-agent learning approaches have developed interactive training environments to explicitly promote collaboration among multiple Large Language Models (LLMs), thereby constructing stronger multi-agent systems (MAS). However, during inference, they require re-executing the MAS to obtain final solutions, which diverges from human cognition that individuals can enhance their reasoning capabilities through interactions with others and resolve questions independently in the future. To investigate whether multi-agent interaction can enhance LLMs' independent problem-solving ability, we introduce ILR, a novel co-learning framework for MAS that integrates two key components: Dynamic Interaction and Perception Calibration. Specifically, Dynamic Interaction first adaptively selects either cooperative or competitive strategies depending on question difficulty and model ability. LLMs then exchange information through Idea3, an innovative interaction paradigm designed to mimic human discussion, before deriving their respective final answers. In Perception Calibration, ILR employs Group Relative Policy Optimization (GRPO) to train LLMs while integrating one LLM's reward distribution characteristics into another's reward function, thereby enhancing the cohesion of multi-agent interactions. We evaluate the effectiveness of ILR across three LLMs from two model families of varying scales on five mathematical, one coding, one general question answering, and one scientific reasoning benchmarks. Experimental results show that ILR consistently outperforms single-agent learning, yielding an improvement of up to 5% over the strongest baseline. We further discover that Idea3 can enhance the robustness of stronger LLMs during multi-agent inference, and dynamic interaction types can boost multi-agent learning compared to pure cooperative or competitive strategies.
arXiv:2510.01038v2 Announce Type: replace Abstract: Perturbation-based explainability methods face criticism due to their reliance on out-of-distribution mutants. This raises doubts about the quality of the explanations. In this paper, we introduce a novel forward pass paradigm, Activation-Deactivation (AD), which obviates the need for perturbation of the input. AD replaces perturbation of input features with switching off parts of the model corresponding to to the intended perturbations. We implement ConvAD, an AD approximation algorithm for CNNs. ConvAD is a drop-in mechanism that can be easily added to any trained CNN and, without any additional training, generates more robust and more transferable explanations. We provide evaluation results across multiple architectures, datasets, methods and perturbation strategies, demonstrating the superior quality of ConvAD compared to the SOTA.
arXiv:2510.10355v2 Announce Type: replace Abstract: The semi-implicit (partly decoupled, also called staggered or fraction-step) time discretization is applied to compressible nonlinear dynamical models of viscoelastic solids in the Eulerian description, i.e.\ in the actual deforming configuration, formulated fully in terms of rates. The Kelvin-Voigt rheology and also, in the deviatoric part, the Jeffreys rheology are considered. The numerical stability and, considering the Stokes-type viscosity multipolar of the 2nd-grade, also convergence towards weak solutions are proved in three-dimensional situations, exploiting the convexity of the kinetic energy when written in terms of linear momentum. No (poly)convexity of the stored energy is required and some enhancements (specifically towards damage and diffusion models) are briefly outlined, too.
arXiv:2510.14214v2 Announce Type: replace Abstract: In 5G networks, base station disaggregation and new services pose challenges for radio access network (RAN) configuration, particularly in meeting bandwidth and latency constraints. The BS disaggregation is enabled by functional splitting (FS), which distributes the RAN functions in processing nodes and alleviates latency and bandwidth requirements in the fronthaul (FH). In addition to network performance, energy consumption is a critical concern for mobile network operators (MNO), since RAN operations constitute a significant portion of their operational expenditure. RAN configuration optimization is essential for balancing service performance and cost-effective energy consumption. In this paper, we propose a mixed-integer linear programming (MILP) model incorporating three objective functions: (i) minimizing fronthaul latency, (ii) minimizing energy consumption, and (iii) a bi-objective optimization that jointly balances both latency and energy consumption. The model determines the optimal FS Option, RAN function placement, and routing for eMBB, URLLC, and mMTC slices. While prior studies have addressed RAN configuration from either an energy minimization or a latency reduction perspective, few have considered both aspects simultaneously in realistic scenarios. Our evaluation accounts for different topologies, accounts for variations in aggregated gNB demand, explores diverse FS combinations, and incorporates Time Sensitive Networking modeling for latency analysis, which is critical for RAN performance. Given that MILP execution can be computationally intensive, we propose a heuristic algorithm that adheres to RAN constraints while providing near-optimal solutions. The results reveal an inherent trade-off between latency and energy consumption, highlighting the need for dynamic RAN reconfiguration. These insights provide a foundation for optimizing existing and future RAN deployments.
arXiv:2510.15747v4 Announce Type: replace Abstract: A grassroots platform is a multiagent distributed system in which multiple independent instances can form and operate independently of each other and of any global resource, yet may coalesce into ever larger instances, possibly resulting in a single global instance. Grassroots platforms aim to offer an egalitarian/democratic alternative to centralised/autocratic and decentralised/plutocratic global platforms. Here, we present Grassroots Logic Programs (GLP), a multiagent concurrent logic programming language designed for the implementation of grassroots platforms: we recall the standard operational semantics of logic programs; introduce the concurrent operational semantics of GLP as its restriction; recall multiagent atomic transactions; use them to introduce a multiagent operational semantics of GLP; and prove multiagent GLP to be grassroots. The grassroots social graph -- the foundational grassroots platform on which all others are based -- serves as a GLP programming example. These mathematical foundations are being used by AI to implement GLP as well as to program in GLP: a workstation-based implementation of concurrent GLP in Dart was derived from the concurrent operational semantics of GLP; a multiagent smartphone-based implementation of GLP in Dart/Flutter is being developed based on the multiagent operational semantics of GLP; a moded type system for GLP was designed (and implemented by AI in Dart) to facilitate collaborative human-AI development of GLP programs, where AI derives working GLP programs from human-approved type definitions and declarations; GLP implementations of grassroots platforms for the social graph, social networks, currencies and bonds, and more, have been derived by AI from mathematical specifications written as volitional multiagent atomic transactions.
arXiv:2510.16923v3 Announce Type: replace Abstract: Deep learning models deployed in safety critical applications like autonomous driving use simulations to test their robustness against adversarial attacks in realistic conditions. However, these simulations are non-differentiable, forcing researchers to create attacks that do not integrate simulation environmental factors, reducing attack success. To address this limitation, we introduce UNDREAM, the first software framework that bridges the gap between photorealistic simulators and differentiable renderers to enable end-to-end optimization of adversarial perturbations on any 3D objects. UNDREAM enables manipulation of the environment by offering complete control over weather, lighting, backgrounds, camera angles, trajectories, and realistic human and object movements, thereby allowing the creation of diverse scenes. We showcase a wide array of distinct physically plausible adversarial objects that UNDREAM enables researchers to swiftly explore in different configurable environments. This combination of photorealistic simulation and differentiable optimization opens new avenues for advancing research of physical adversarial attacks.
arXiv:2510.20692v3 Announce Type: replace Abstract: Access control policies are reliability-critical configuration artifacts in cloud systems, yet administrators frequently struggle to verify that a policy permits exactly what they intend. This verification gap cannot be remedied by using LLMs to synthesize policies: we find that reasoning and non-reasoning LLMs fluently explain policy behavior but cannot reason about policy semantics with reliability-grade precision, even when the specification is the LLM's own explanation. We formulate this impasse as the Verifiable Synthesis Paradox: the verification gap persists regardless of how the policy was authored. To remedy this, we introduce PolicySummarizer, a neurosymbolic tool that pairs finite-state automata with an LLM-based simplification to generate precise human-readable characterizations of requests allowed by a policy. PolicySummarizer uses model counting to guarantee the fidelity of the generated characterization by rejecting characterizations that fall below a user-configured threshold in favor of the formally derived one. On 546 AWS, 100 Microsoft Azure, and 100 Google Cloud Platform policies, PolicySummarizer achieves a mean similarity score of 0.93 and a 2.7x improvement over an SMT-based baseline. In a user study, PolicySummarizer raised policy-change-review accuracy from 39% to 93% on the hardest sub-task while reducing self-reported mental demand. We release PolicySummarizer as an open-source tool.
arXiv:2511.05747v4 Announce Type: replace Abstract: Long Chain-of-Thought (CoT) traces can improve reasoning accuracy, but repeatedly generating them is costly for smaller or latency-constrained language models. This paper studies a practical alternative: produce a rich rationale once with a capable \emph{thinking} model, compress it, and reuse the compressed trace as context for a cheaper \emph{answering} model. We introduce CoT-X, an adaptive framework for cross-model CoT transfer. CoT-X segments reasoning traces into semantic units, scores their diagnostic and logical importance, selects budget-feasible evidence paths, and reconstructs a coherent compressed rationale for the answering model. On $7,501$ Japanese medical licensing questions spanning $10$ specialties, CoT-X improves accuracy over direct truncation by up to $40.5\%$ under the same token budget, with the largest gains at $64$--$256$ tokens. Across $64$ thinking--answering pairs from eight DeepSeek-R1 and Qwen3 models (1.5B--32B parameters), reasoning transfer is most reliable within a model family, yet remains effective across families once compression normalizes the trace. A Gaussian Process Bayesian optimization layer finds near-optimal model--budget configurations with $15$ evaluations rather than an exhaustive search over all $64$ pairs, reducing evaluation cost by $84\%$. These results show that reasoning quality, token budget, and model compatibility can be optimized jointly, making CoT-style reasoning more practical under realistic deployment constraints.
arXiv:2511.07202v3 Announce Type: replace Abstract: Failures are the norm in highly complex and heterogeneous devices spanning the distributed computing continuum (DCC), from resource-constrained IoT and edge nodes to high-performance computing systems. Ensuring reliability and global consistency across these layers remains a major challenge, especially for AI-driven workloads requiring real-time, adaptive coordination. This work-in-progress paper introduces a Probabilistic Active Inference Resilience Agent (PAIR-Agent) to achieve resilience in DCC systems. PAIR-Agent performs three core operations: (i) constructing a causal fault graph from device logs, (ii) identifying faults while managing certainties and uncertainties using Markov blankets and the free energy principle, and (iii) autonomously healing issues through active inference. Through continuous monitoring and adaptive reconfiguration, the agent maintains service continuity and stability under diverse failure conditions. Theoretical validations confirm the reliability and effectiveness of the proposed framework.
arXiv:2607.04557v1 Announce Type: new Abstract: Accurate prediction of patient-specific therapeutic response from pre-treatment transcriptomes is hindered by the scarcity of matched clinical response labels and post-treatment molecular profiles. Preclinical transfer-learning models can simulate drug-induced expression changes but are often hard to interpret and unstable, whereas knowledge-graph methods provide mechanistic context yet remain static and fail to capture drug-induced transcriptomic perturbation dynamics. We propose PREDIKTOR, a patient-centered multi-view framework that aligns a personalized network view with a transferable transcriptomic perturbation view to predict clinical drug response. For each patient, we construct an individualized gene regulatory network from tumor expression using DysRegNet and augment it with drug-target links from DrugBank; a graph neural encoder yields a drug-centric, mechanistically grounded embedding. In parallel, a frozen condition-specific gene-gene attention model pretrained on LINCS L1000 generates a simulated post-perturbation transcriptomic profile for the same patient-drug pair. We align the two views in a shared latent space via a CLIP-style contrastive objective with drug-context hard negatives, then concatenate the representations for end-to-end response classification. On TCGA, PREDIKTOR consistently outperforms state-of-the-art baselines under patient-, drug-, and tissue-split evaluations, and transfers zero-shot to the I-SPY2 trial, improving AUROC by 5.6% over competing methods. The aligned embeddings yield stable gene and pathway attributions that recover known mechanisms, supporting actionable and interpretable precision oncology.
arXiv:2511.13039v2 Announce Type: replace Abstract: Open-Vocabulary Temporal Action Localization (OV-TAL) aims to recognize and localize instances of any desired action categories in videos without explicitly curating training data for all categories. Existing methods mostly recognize action categories at a single granularity, which degrades the recognition accuracy of both base and novel action categories. To address these issues, we propose a Multi-Grained Category-Aware Network (MGCA-Net) comprising a localizer, an action presence predictor, a conventional classifier, and a coarse-to-fine classifier. Specifically, the localizer localizes category-agnostic action proposals. For these action proposals, the action presence predictor estimates the probability that they belong to an action instance. At the same time, the conventional classifier predicts the probability of each action proposal over base action categories at the snippet granularity. Novel action categories are recognized by the coarse-to-fine classifier, which first identifies action presence at the video granularity. Finally, it assigns each action proposal to one category from the coarse categories at the proposal granularity. Through coarse-to-fine category awareness for novel actions and the conventional classifier's awareness of base actions, multi-grained category awareness is achieved, effectively enhancing localization performance. Comprehensive evaluations on the THUMOS'14 and ActivityNet-1.3 benchmarks demonstrate that our method achieves state-of-the-art performance. Furthermore, our MGCA-Net achieves state-of-the-art results under the Zero-Shot Temporal Action Localization setting.
arXiv:2511.17384v2 Announce Type: replace Abstract: While Visual Large Language Models (VLLMs) show great promise as embodied agents, they continue to face substantial challenges in spatial reasoning. Existing embodied benchmarks largely focus on passive, static household environments and evaluate isolated capabilities, failing to capture holistic performance in interactive and dynamic complexity of specific domains. To fill this gap, we present IndustryNav, the first dynamic industrial navigation benchmark for active spatial reasoning. IndustryNav leverages 12 manually created, high-fidelity Unity warehouse scenarios featuring dynamic objects and human movement. We proposes a zero-shot PointGoal navigation pipeline that effectively combines egocentric vision with global odometry to assess holistic local-global planning. Furthermore, we introduce the "collision rate" and "warning rate" metrics to measure safety-oriented behaviors. A comprehensive study of fourteen state-of-the-art VLLMs (including models such as GPT-5.2, Claude-4.6, and Gemini-3) reveals that closed-source models maintain a consistent advantage; however, all agents exhibit notable deficiencies in robust path planning, collision avoidance and active exploration. This highlights a critical need for embodied research to move beyond passive perception and toward tasks that demand stable planning, active exploration, and safe behavior in vivid, dynamic environments.
arXiv:2511.17812v3 Announce Type: replace Abstract: Flow matching models effectively represent complex distributions, yet estimating expectations of functions of their outputs remains challenging under limited sampling budgets. Independent sampling often yields high-variance estimates, especially when rare but high-impact outcomes dominate the expectation. We propose a non-IID sampling framework that jointly draws multiple samples to cover diverse, salient regions of a flow matching model's generative distribution. To balance diversity and quality, we introduce a score-based regularization for the diversity mechanism (SR), which uses the score function, i.e., the gradient of the log probability, to ensure samples are pushed apart within high-density regions of the data manifold, mitigating off-manifold drift. To enable unbiased estimation when desired, we further develop an approach for importance weighting of non-IID flow samples by learning a residual velocity field that reproduces the marginal distribution of the non-IID samples and by evolving importance weights along trajectories. Empirically, our method produces diverse, high-quality samples and accurate importance-weight estimates and debiased expectation estimates, advancing the reliable characterization of flow matching model outputs.
arXiv:2511.20878v2 Announce Type: replace Abstract: The advent of Artificial Intelligence (AI), particularly large language models (LLMs), has revolutionized software development by enabling developers to specify tasks in natural language and receive corresponding code, boosting productivity. However, this shift also introduces security risks, as LLMs may generate insecure code that can be exploited by adversaries. Conventional educational approaches emphasize efficiency while overlooking these risks, leaving students unprepared to identify and mitigate security issues in AI-assisted workflows. To surface this gap, we present \texttt{Bifr\"ost}, a classroom measurement and feedback framework that pairs an adversarially configured code-generation model with a VS Code extension and automated vulnerability reports that instructors can use to guide follow-up discussions. Through classroom deployments with undergraduate students ($n=61$), we observe that students frequently accepted insecure LLM-generated code despite prior security coursework and stated skepticism. A post-feedback survey ($n=21$) provides preliminary evidence that students' stated trust shifted toward greater skepticism after receiving Bifr\"ost feedback, and that some students articulated more security-specific concerns about AI-generated code.
arXiv:2511.21176v2 Announce Type: replace Abstract: Scientific publications have long served as the cornerstone of innovation, exhibiting stable growth over the years. Recently, however, retractions have surged dramatically, driven largely by the proliferation of low-quality and fraudulent articles, posing a substantial threat to research integrity. By integrating annual publication and retraction data, this study employs the relative retraction rate (R3) to systematically examine disparities and evolving trends from a topical perspective. Our analysis reveals that the number of retractions has grown significantly faster than that of global publications, yielding an overall retraction rate of 0.12%. While retractions occur across all disciplines, substantial disparities exist, ranging from 0.035% in Physics to 0.34% in Computer Science. This gap widens at finer levels of granularity, reaching roughly 8.99% in Human-Computer Interaction. Moreover, unusually high R3 values frequently coincide with rapid publication growth in specific fields. We also developed Retraction Monitor, a web application for monitoring retraction dynamics across diverse fields, enabling stakeholders to visualize these trends and assess risks to research integrity. These findings provide valuable insights for identifying high-risk fields and developing tailored governance policies to strengthen research rigor and mitigate field-specific retraction risks.
arXiv:2607.04516v1 Announce Type: new Abstract: This study presents a near-real-time, meter-scale three-dimensional urban wind simulation framework for low-altitude flight events in complex urban meteorological environments. It reconstructs high-resolution wind fields by combining sparse observations with efficient microscale flow modeling. The framework integrates lattice Boltzmann method large-eddy simulation (LBM-LES), high-fidelity urban morphology reconstruction that explicitly resolves real building details, and observation-driven boundary assimilation into a rapid end-to-end pipeline for realistic urban domains. Multi-site Doppler lidar measurements from dense urban Guangzhou, China, are used for evaluation. The system reconstructs three-dimensional wind fields at 5 m resolution over kilometer-scale domains within minutes. Robustness and accuracy are tested through controlled observation reduction, independent validation against withheld lidar stations, and sensitivity analyses of grid resolution and precursor domain extent. Results show stable reproduction of vertical wind structures and key local flow features under complex morphology and limited observations, providing a scalable pathway for near-real-time urban wind reconstruction.
arXiv:2512.01016v2 Announce Type: replace Abstract: We present the first deterministic, finite-step algorithm for exact tensor ring (TR) decomposition. Our method leverages blockwise simultaneous diagonalization to recover TR cores from a limited number of tensor observations, under a dimension condition requiring each mode size to be at least quadratic in the TR rank and under a genericity assumption on the cores, thereby providing both algebraic insight and practical efficiency. We extend the approach to the symmetric TR setting, where parameter complexity is significantly reduced and applications arise naturally in physics-based modeling and exchangeable data analysis. To handle noisy observations, we develop a robust recovery scheme that couples our initialization with alternating least squares, achieving faster convergence and improved accuracy compared to classical methods. As applications, we obtain new algorithms for questions in other domains where tensor ring decomposition is a key primitive, namely matrix product state tomography in quantum information and provable learning of pushforward distributions in the foundations of machine learning. These contributions advance the algorithmic foundations of TR decomposition and open new opportunities for scalable tensor network computation.
arXiv:2512.06628v4 Announce Type: replace Abstract: Scalable embodied intelligence is constrained by the scarcity of diverse, long-horizon robotic manipulation data. Existing video world models in this domain are limited to synthesizing short clips of simple actions and often rely on manually defined trajectories. To this end, we introduce MIND-V, a cognitive hierarchical world model designed to synthesize physically plausible and logically coherent videos of long-horizon robotic manipulation. Inspired by cognitive science, MIND-V bridges high-level reasoning with pixel-level synthesis through three core components: a Semantic Reasoning Hub (SRH) that leverages a pre-trained vision-language model for task planning; a Behavioral Semantic Bridge (BSB) that translates abstract instructions into domain-invariant representations; and a Motor Video Generator (MVG) for conditional video rendering. MIND-V employs Staged Visual Future Rollouts, a test-time optimization strategy to enhance long-horizon robustness. To enforce adherence to physical laws, we introduce a GRPO reinforcement learning post-training phase guided by a novel Physical Foresight Coherence (PFC) reward. PFC leverages the V-JEPA2 world model as a physics referee to penalize implausible dynamics in the latent feature space. Experiments confirm MIND-V's SOTA performance in long-horizon simulation and its significant value for policy learning, introducing a scalable and fully autonomous framework for embodied data synthesis.
arXiv:2512.08029v3 Announce Type: replace Abstract: Clinical decision-making in oncology requires predicting dynamic disease evolution, a task current static AI predictors cannot perform. While world models (WMs) offer a paradigm for generative prediction, existing medical applications remain limited. Existing methods often rely on stochastic diffusion models, focusing on visual reconstruction rather than causal, physiological transitions. Furthermore, in medical domain, models like MeWM typically ignore patient-specific temporal and clinical contexts and lack a feedback mechanism to link predictions to treatment decisions. To address these gaps, we introduce CLARITY, a medical world model that forecasts disease evolution directly within a structured latent space. It explicitly integrates time intervals (temporal context) and patient-specific data (clinical context) to model treatment-conditioned progression as a smooth, interpretable trajectory, and thus generate physiologically faithful, individualized treatment plans. Finally, CLARITY introduces a novel prediction-to-decision framework, translating latent rollouts into transparent, actionable recommendations. CLARITY demonstrates state-of-the-art performance in treatment planning. On the MU-Glioma-Post dataset, our approach outperforms recent MeWM by 12\%, and significantly surpasses all other medical-specific large language models.
arXiv:2512.09423v2 Announce Type: replace Abstract: Learning natural body motion remains challenging due to the strong coupling between spatial geometry and temporal dynamics. Embedding motion in phase manifolds, latent spaces that capture local periodicity, has proven effective for motion prediction; however, existing approaches are tied to fixed skeletons and narrow motion distributions, limiting their applicability across diverse settings. We introduce FunPhase, a functional periodic autoencoder that learns a phase manifold for motion and replaces discrete temporal decoding with a function-space formulation, enabling smooth trajectories that can be sampled at arbitrary temporal resolutions. FunPhase unifies motion prediction and generation within a single interpretable phase manifold, enabling motion generation via latent diffusion, generalizes across skeletons and datasets, and supports downstream tasks such as motion super-resolution and partial-body completion. Our model achieves substantially lower reconstruction error than prior periodic autoencoder baselines, achieving uniform improvements of at least $45\%$ across all metrics, while enabling a broader range of applications and performing on par with state-of-the-art motion generation methods.
arXiv:2607.04509v1 Announce Type: new Abstract: We consider the 2D RMQ encoding problem: given an $m\times n$ array of $mn$ elements over a total order, encode it such that, for any query rectangle, the position of its maximum element can be reported without accessing the original array. For $m \le n$, it is known how to encode the array in $O(mn \min\{m, \log n\})$ bits with $O(1)$-time queries [Brodal et al., Algorithmica 2012], and also how to obtain an asymptotically optimal encoding consisting of $O(mn \log m)$ bits [Brodal et al., ESA 2013]. However, the latter approach does not prove any guarantee on the query time, and it appears to be inherently sequential: it requires scanning the whole encoding to answer a query. We design a different encoding that uses near-optimal space while allowing for efficient queries. More concretely, for every parameter $\kappa\in[1, \log\log n]$, our encoding uses $O(\kappa mn(\log m+\log\log n))$ bits and answers 2D RMQ queries in $O(\log^{1/\kappa}n)$ time.