Qiming Hu

Qingguo Hu, Zhenghao Lin, Ziyue Yang et al.

Mixture-of-Experts (MoE) has emerged as a promising paradigm for foundation models due to its efficient and powerful scalability. In this work, we present Sigma-MoE-Tiny, an MoE language model that achieves the highest sparsity compared to existing open-source models. Sigma-MoE-Tiny employs fine-grained expert segmentation with up to 96 experts per layer, while activating only one expert for each token, resulting in 20B total parameters with just 0.5B activated. The major challenge introduced by such extreme sparsity lies in expert load balancing. We find that the widely-used load balancing loss tends to become ineffective in the lower layers under this setting. To address this issue, we propose a progressive sparsification schedule aiming to balance expert utilization and training stability. Sigma-MoE-Tiny is pre-trained on a diverse and high-quality corpus, followed by post-training to further unlock its capabilities. The entire training process remains remarkably stable, with no occurrence of irrecoverable loss spikes. Comprehensive evaluations reveal that, despite activating only 0.5B parameters, Sigma-MoE-Tiny achieves top-tier performance among counterparts of comparable or significantly larger scale. In addition, we provide an in-depth discussion of load balancing in highly sparse MoE models, offering insights for advancing sparsity in future MoE architectures. Project page: https://qghuxmu.github.io/Sigma-MoE-Tiny Code: https://github.com/microsoft/ltp-megatron-lm

Qiming Hu, Linlong Fan, Yiyan Luo et al.

The introduction of generative models has significantly advanced image super-resolution (SR) in handling real-world degradations. However, they often incur fidelity-related issues, particularly distorting textual structures. In this paper, we introduce a novel diffusion-based SR framework, namely TADiSR, which integrates text-aware attention and joint segmentation decoders to recover not only natural details but also the structural fidelity of text regions in degraded real-world images. Moreover, we propose a complete pipeline for synthesizing high-quality images with fine-grained full-image text masks, combining realistic foreground text regions with detailed background content. Extensive experiments demonstrate that our approach substantially enhances text legibility in super-resolved images, achieving state-of-the-art performance across multiple evaluation metrics and exhibiting strong generalization to real-world scenarios. Our code is available at \href{https://github.com/mingcv/TADiSR}{here}.

Qiming Hu, Xiaojie Guo

Single image reflection separation (SIRS), as a representative blind source separation task, aims to recover two layers, $\textit{i.e.}$, transmission and reflection, from one mixed observation, which is challenging due to the highly ill-posed nature. Existing deep learning based solutions typically restore the target layers individually, or with some concerns at the end of the output, barely taking into account the interaction across the two streams/branches. In order to utilize information more efficiently, this work presents a general yet simple interactive strategy, namely $\textit{your trash is my treasure}$ (YTMT), for constructing dual-stream decomposition networks. To be specific, we explicitly enforce the two streams to communicate with each other block-wisely. Inspired by the additive property between the two components, the interactive path can be easily built via transferring, instead of discarding, deactivated information by the ReLU rectifier from one stream to the other. Both ablation studies and experimental results on widely-used SIRS datasets are conducted to demonstrate the efficacy of YTMT, and reveal its superiority over other state-of-the-art alternatives. The implementation is quite simple and our code is publicly available at $\href{https://github.com/mingcv/YTMT-Strategy}{\textit{https://github.com/mingcv/YTMT-Strategy}}$.

Xinhui Li, Xinyu He, Qiming Hu et al.

In this paper, we introduce, for the first time, the concept of Set Pivot Learning, a paradigm shift that redefines domain generalization (DG) based on Vision Foundation Models (VFMs). Traditional DG assumes that the target domain is inaccessible during training, but the emergence of VFMs, trained on vast and diverse data, renders this assumption unclear and obsolete. Traditional DG assumes that the target domain is inaccessible during training, but the emergence of VFMs, which are trained on vast and diverse datasets, renders this assumption unclear and obsolete. To address this challenge, we propose Set Pivot Learning (SPL), a new definition of domain migration task based on VFMs, which is more suitable for current research and application requirements. Unlike conventional DG methods, SPL prioritizes adaptive refinement over rigid domain transfer, ensuring continuous alignment with evolving real-world conditions. Specifically, SPL features two key attributes: (i) Dynamic adaptation, transitioning from static domain alignment to flexible, task-driven feature optimization, enabling models to evolve with downstream scenarios; (ii) VFM-centric tuning, leveraging pretrained knowledge as a pivot to hone task-specific representations while preserving cross-domain robustness. Building on SPL, we propose a Dynamic Prompt Fine-Tuning method, which combines a Dynamic Class-aware Prompter with a Prompt-guided Feature Focuser, to elevate VFM performance in targeted scenarios. Extensive experiments on benchmark datasets show the effectiveness of our method, highlighting its superiority over state-of-the-art methods, particularly in generalized segmentation.

Azarakhsh Jalalvand, SangKyeun Kim, Jaemin Seo et al.

A non-linear system governed by multi-spatial and multi-temporal physics scales cannot be fully understood with a single diagnostic, as each provides only a partial view, leading to information loss. Combining multiple diagnostics may also result in incomplete projections of the system's physics. By identifying hidden inter-correlations between diagnostics, we can leverage mutual support to fill in these gaps, but uncovering such correlations analytically is too complex. We introduce a machine learning methodology to address this issue. Unlike traditional methods, our multimodal approach does not rely on the target diagnostic's direct measurements to generate its super-resolution version. Instead, it uses other diagnostics to produce super-resolution data, capturing detailed structural evolution and responses to perturbations previously unobservable. This not only enhances the resolution of a diagnostic for deeper insights but also reconstructs the target diagnostic, providing a valuable tool to mitigate diagnostic failure. This methodology addresses a key challenge in fusion plasmas: the Edge Localized Mode (ELM), a plasma instability that can cause significant erosion of plasma-facing materials. A method to stabilize ELM is using resonant magnetic perturbation (RMP) to trigger magnetic islands. However, limited spatial and temporal resolution restricts analysis of these islands due to their small size, rapid dynamics, and complex plasma interactions. With super-resolution diagnostics, we can experimentally verify theoretical models of magnetic islands for the first time, providing insights into their role in ELM stabilization. This advancement supports the development of effective ELM suppression strategies for future fusion reactors like ITER and has broader applications, potentially revolutionizing diagnostics in fields such as astronomy, astrophysics, and medical imaging.