HiStream: Efficient High-Resolution Video Generation via Redundancy-Eliminated Streaming

High-resolution video generation, while crucial for digital media and film, is computationally bottlenecked by the quadratic complexity of diffusion models, making practical inference infeasible. To address this, we introduce HiStream, an efficient autoregressive framework that systematically reduces redundancy across three axes: i) Spatial Compression: denoising at low resolution before refining at high resolution with cached features; ii) Temporal Compression: a chunk-by-chunk strategy with a fixed-size anchor cache, ensuring stable inference speed; and iii) Timestep Compression: applying fewer denoising steps to subsequent, cache-conditioned chunks. On 1080p benchmarks, our primary HiStream model (i+ii) achieves state-of-the-art visual quality while demonstrating up to 76.2x faster denoising compared to the Wan2.1 baseline and negligible quality loss. Our faster variant, HiStream+, applies all three optimizations (i+ii+iii), achieving a 107.5x acceleration over the baseline, offering a compelling trade-off between speed and quality, thereby making high-resolution video generation both practical and scalable.

Branch Learning in MRI: More Data, More Models, More Training

We investigated two complementary strategies for multicontrast cardiac MR reconstruction: physics-consistent data-space augmentation (DualSpaceCMR) and parameter-efficient capacity scaling via VQPrompt and Moero. DualSpaceCMR couples image-level transforms with kspace noise and motion simulations while preserving forwardmodel consistency. VQPrompt adds a lightweight bottleneck prompt; Moero embeds a sparse mixture of experts within a deep unrolled network with histogram-based routing. In the multivendor, multisite CMRxRecon25 benchmark, we evaluate fewshot and out-of-distribution generalization. On small datasets, k-space motion-plus-noise improves reconstruction; on the large benchmark it degrades performance, revealing sensitivity to augmentation ratio and schedule. VQPrompt produces modest and consistent gains with negligible memory overhead. Moero continues to improve after early plateaus and maintains baseline-like fewshot and out-of-distribution behavior despite mild overfitting, but sparse routing lowers PyTorch throughput and makes wall clock time the main bottleneck. These results motivate scale-aware augmentation and suggest prompt-based capacity scaling as a practical path, while efficiency improvements are crucial for sparse expert models.

OneStory: Coherent Multi-Shot Video Generation with Adaptive Memory

Storytelling in real-world videos often unfolds through multiple shots -- discontinuous yet semantically connected clips that together convey a coherent narrative. However, existing multi-shot video generation (MSV) methods struggle to effectively model long-range cross-shot context, as they rely on limited temporal windows or single keyframe conditioning, leading to degraded performance under complex narratives. In this work, we propose OneStory, enabling global yet compact cross-shot context modeling for consistent and scalable narrative generation. OneStory reformulates MSV as a next-shot generation task, enabling autoregressive shot synthesis while leveraging pretrained image-to-video (I2V) models for strong visual conditioning. We introduce two key modules: a Frame Selection module that constructs a semantically-relevant global memory based on informative frames from prior shots, and an Adaptive Conditioner that performs importance-guided patchification to generate compact context for direct conditioning. We further curate a high-quality multi-shot dataset with referential captions to mirror real-world storytelling patterns, and design effective training strategies under the next-shot paradigm. Finetuned from a pretrained I2V model on our curated 60K dataset, OneStory achieves state-of-the-art narrative coherence across diverse and complex scenes in both text- and image-conditioned settings, enabling controllable and immersive long-form video storytelling.

Mixture of States: Routing Token-Level Dynamics for Multimodal Generation

We introduce MoS (Mixture of States), a novel fusion paradigm for multimodal diffusion models that merges modalities using flexible, state-based interactions. The core of MoS is a learnable, token-wise router that creates denoising timestep- and input-dependent interactions between modalities' hidden states, precisely aligning token-level features with the diffusion trajectory. This router sparsely selects the top-$k$ hidden states and is trained with an $ε$-greedy strategy, efficiently selecting contextual features with minimal learnable parameters and negligible computational overhead. We validate our design with text-to-image generation (MoS-Image) and editing (MoS-Editing), which achieve state-of-the-art results. With only 3B to 5B parameters, our models match or surpass counterparts up to $4\times$ larger. These findings establish MoS as a flexible and compute-efficient paradigm for scaling multimodal diffusion models.

Uncovering Scaling Laws for Large Language Models via Inverse Problems

Large Language Models (LLMs) are large-scale pretrained models that have achieved remarkable success across diverse domains. These successes have been driven by unprecedented complexity and scale in both data and computations. However, due to the high costs of training such models, brute-force trial-and-error approaches to improve LLMs are not feasible. Inspired by the success of inverse problems in uncovering fundamental scientific laws, this position paper advocates that inverse problems can also efficiently uncover scaling laws that guide the building of LLMs to achieve the desirable performance with significantly better cost-effectiveness.

Energy Efficiency Optimization for Movable Antenna-Aided Communication Systems

This paper investigates the energy efficiency optimization for movable antenna (MA) systems by considering the time delay and energy consumption introduced by MA movement. We first derive the upper bound on energy efficiency for a single-user downlink communication system, where the user is equipped with a single MA. Then, the energy efficiency maximization problem is formulated to optimize the MA position, and an efficient algorithm based on successive convex approximation is proposed to solve this non-convex optimization problem. Simulation results show that, despite the overhead caused by MA movement, the MA system can still improve the energy efficiency compared to the conventional fixed-position antenna (FPA) system.

Polarforming Design with Phase Shifter Based Polarization Reconfigurable Antennas

In this paper, we propose a new form of polarization reconfigurable antennas (PRAs) that can form linear, circular, and general elliptical polarizations assisted by phase shifters (PSs). With PRAs, polarforming is achieved, which enables the antenna to shape its polarization into a desired state for aligning with that of the received electromagnetic (EM) wave or reconfiguring that of the transmit EM wave. To demonstrate the benefits of polarforming, we investigate a PRA-aided single-input single-output (SISO) communication system equipped with tunable PSs for polarization adaptation. We characterize the achievable signal-to-noise ratio (SNR) at the receiver as a function of the phase shifts of PS-based PRAs. Moreover, we develop an alternating optimization approach to maximize the SNR by optimizing the phase shifts at both the transmitter and receiver. Finally, comprehensive simulation results are presented, which not only validate the effectiveness of polarforming in mitigating the channel depolarization effects, but also demonstrate its substantial performance improvement over conventional systems.

Polarforming for Wireless Networks: Opportunities and Challenges

Polarforming emerges as a promising technique for manipulating the polarization of electromagnetic (EM) waves by shaping the polarization of an antenna into a desired state. By dynamically adjusting antenna polarization, polarforming enables real-time polarization matching or mismatching with received EM waves, thereby leveraging polarization degrees of freedom (DoFs) to enhance wireless communication performance. In this article, we first present an overview of the fundamental principles and design approaches underlying the polarforming technique. We then analyze the key advantages of polarforming, including hardware cost reduction, depolarization mitigation, channel adaptation, signal power enhancement, and interference suppression. Furthermore, we explore promising applications of polarforming for next-generation wireless networks. Numerical case studies demonstrate the substantial performance gains of polarforming over conventional fixed-polarization antenna (FPA) systems, along with a discussion of implementation challenges to motivate future research.

MPDrive: Improving Spatial Understanding with Marker-Based Prompt Learning for Autonomous Driving

Autonomous driving visual question answering (AD-VQA) aims to answer questions related to perception, prediction, and planning based on given driving scene images, heavily relying on the model's spatial understanding capabilities. Prior works typically express spatial information through textual representations of coordinates, resulting in semantic gaps between visual coordinate representations and textual descriptions. This oversight hinders the accurate transmission of spatial information and increases the expressive burden. To address this, we propose a novel Marker-based Prompt learning framework (MPDrive), which represents spatial coordinates by concise visual markers, ensuring linguistic expressive consistency and enhancing the accuracy of both visual perception and spatial expression in AD-VQA. Specifically, we create marker images by employing a detection expert to overlay object regions with numerical labels, converting complex textual coordinate generation into straightforward text-based visual marker predictions. Moreover, we fuse original and marker images as scene-level features and integrate them with detection priors to derive instance-level features. By combining these features, we construct dual-granularity visual prompts that stimulate the LLM's spatial perception capabilities. Extensive experiments on the DriveLM and CODA-LM datasets show that MPDrive achieves state-of-the-art performance, particularly in cases requiring sophisticated spatial understanding.

TETRIS: Optimal Draft Token Selection for Batch Speculative Decoding

We propose TETRIS, a novel method that optimizes the total throughput of batch speculative decoding in multi-request settings. Unlike existing methods that optimize for a single request or a group of requests as a whole, TETRIS actively selects the most promising draft tokens (for every request in a batch) to be accepted when verified in parallel, resulting in fewer rejected tokens and hence less wasted computing resources. Such an effective resource utilization to achieve fast inference in large language models (LLMs) is especially important to service providers with limited inference capacity. Compared to baseline speculative decoding, TETRIS yields a consistently higher acceptance rate and more effective utilization of the limited inference capacity. We show theoretically and empirically that TETRIS outperforms baseline speculative decoding and existing methods that dynamically select draft tokens, leading to a more efficient batch inference in LLMs.