A-Li Luo

Ng Shyh-Chang, A-Li Luo, Bo Qiu

We prove the converse of the universal approximation theorem, i.e. a neural network (NN) encoding theorem which shows that for every stably converged NN of continuous activation functions, its weight matrix actually encodes a continuous function that approximates its training dataset to within a finite margin of error over a bounded domain. We further show that using the Eckart-Young theorem for truncated singular value decomposition of the weight matrix for every NN layer, we can illuminate the nature of the latent space manifold of the training dataset encoded and represented by every NN layer, and the geometric nature of the mathematical operations performed by each NN layer. Our results have implications for understanding how NNs break the curse of dimensionality by harnessing memory capacity for expressivity, and that the two are complementary. This Layer Matrix Decomposition (LMD) further suggests a close relationship between eigen-decomposition of NN layers and the latest advances in conceptualizations of Hopfield networks and Transformer NN models.

Xiaohua Ma, Xiangru Li, Ali Luo et al.

With the development of a series of Galaxy sky surveys in recent years, the observations increased rapidly, which makes the research of machine learning methods for galaxy image recognition a hot topic. Available automatic galaxy image recognition researches are plagued by the large differences in similarity between categories, the imbalance of data between different classes, and the discrepancy between the discrete representation of Galaxy classes and the essentially gradual changes from one morphological class to the adjacent class (DDRGC). These limitations have motivated several astronomers and machine learning experts to design projects with improved galaxy image recognition capabilities. Therefore, this paper proposes a novel learning method, ``Hierarchical Imbalanced data learning with Weighted sampling and Label smoothing" (HIWL). The HIWL consists of three key techniques respectively dealing with the above-mentioned three problems: (1) Designed a hierarchical galaxy classification model based on an efficient backbone network; (2) Utilized a weighted sampling scheme to deal with the imbalance problem; (3) Adopted a label smoothing technique to alleviate the DDRGC problem. We applied this method to galaxy photometric images from the Galaxy Zoo-The Galaxy Challenge, exploring the recognition of completely round smooth, in between smooth, cigar-shaped, edge-on and spiral. The overall classification accuracy is 96.32\%, and some superiorities of the HIWL are shown based on recall, precision, and F1-Score in comparing with some related works. In addition, we also explored the visualization of the galaxy image features and model attention to understand the foundations of the proposed scheme.

Mohsen Hariri, Alan Luo, Weicong Chen et al.

Large Language Models (LLMs) suffer inference-time memory bottlenecks dominated by the attention Key-Value (KV) cache, which scales with model size and context length. While KV-cache quantization alleviates this cost, bit allocation between keys and values is often tuned heuristically, lacking theoretical grounding and generalizability. This paper proposes two theorems that anchor mixed-precision KV quantization in the intrinsic geometry of Transformer models. First, key projections systematically have larger spectral and Frobenius norms than value matrices, implying higher information density along the key path. Second, for any given memory budget, prioritizing precision for keys over values strictly reduces quantization error and better preserves accuracy. Empirical evaluations across various prominent LLMs and benchmarks show that key-favored allocations (e.g., 4-bit keys, 2-bit values) retain up to 98.3\% accuracy compared to uniform allocations (e.g., 4-bit for both), while conserving memory. These results transform bit allocation from ad hoc tuning into a theoretically grounded, geometry-driven design principle for efficient LLM inference. Source code is available at https://github.com/mohsenhariri/spectral-kv.

Junyao Shi, Zhuolun Zhao, Tianyou Wang et al.

Many recent advances in robotic manipulation have come through imitation learning, yet these rely largely on mimicking a particularly hard-to-acquire form of demonstrations: those collected on the same robot in the same room with the same objects as the trained policy must handle at test time. In contrast, large pre-recorded human video datasets demonstrating manipulation skills in-the-wild already exist, which contain valuable information for robots. Is it possible to distill a repository of useful robotic skill policies out of such data without any additional requirements on robot-specific demonstrations or exploration? We present the first such system ZeroMimic, that generates immediately deployable image goal-conditioned skill policies for several common categories of manipulation tasks (opening, closing, pouring, pick&place, cutting, and stirring) each capable of acting upon diverse objects and across diverse unseen task setups. ZeroMimic is carefully designed to exploit recent advances in semantic and geometric visual understanding of human videos, together with modern grasp affordance detectors and imitation policy classes. After training ZeroMimic on the popular EpicKitchens dataset of ego-centric human videos, we evaluate its out-of-the-box performance in varied real-world and simulated kitchen settings with two different robot embodiments, demonstrating its impressive abilities to handle these varied tasks. To enable plug-and-play reuse of ZeroMimic policies on other task setups and robots, we release software and policy checkpoints of our skill policies.

Mingxiang Fu, Yu Song, Jiameng Lv et al.

The exponential growth of astronomical datasets provides an unprecedented opportunity for humans to gain insight into the Universe. However, effectively analyzing this vast amount of data poses a significant challenge. Astronomers are turning to deep learning techniques to address this, but the methods are limited by their specific training sets, leading to considerable duplicate workloads too. Hence, as an example to present how to overcome the issue, we built a framework for general analysis of galaxy images, based on a large vision model (LVM) plus downstream tasks (DST), including galaxy morphological classification, image restoration, object detection, parameter extraction, and more. Considering the low signal-to-noise ratio of galaxy images and the imbalanced distribution of galaxy categories, we have incorporated a Human-in-the-loop (HITL) module into our large vision model, which leverages human knowledge to enhance the reliability and interpretability of processing galaxy images interactively. The proposed framework exhibits notable few-shot learning capabilities and versatile adaptability to all the abovementioned tasks on galaxy images in the DESI legacy imaging surveys. Expressly, for object detection, trained by 1000 data points, our DST upon the LVM achieves an accuracy of 96.7%, while ResNet50 plus Mask R-CNN gives an accuracy of 93.1%; for morphology classification, to obtain AUC ~0.9, LVM plus DST and HITL only requests 1/50 training sets compared to ResNet18. Expectedly, multimodal data can be integrated similarly, which opens up possibilities for conducting joint analyses with datasets spanning diverse domains in the era of multi-message astronomy.

Jiahang Cao, Yize Huang, Hanzhong Guo et al.

Diffusion-based models for robotic control, including vision-language-action (VLA) and vision-action (VA) policies, have demonstrated significant capabilities. Yet their advancement is constrained by the high cost of acquiring large-scale interaction datasets. This work introduces an alternative paradigm for enhancing policy performance without additional model training. Perhaps surprisingly, we demonstrate that the composed policies can exceed the performance of either parent policy. Our contribution is threefold. First, we establish a theoretical foundation showing that the convex composition of distributional scores from multiple diffusion models can yield a superior one-step functional objective compared to any individual score. A Grönwall-type bound is then used to show that this single-step improvement propagates through entire generation trajectories, leading to systemic performance gains. Second, motivated by these results, we propose General Policy Composition (GPC), a training-free method that enhances performance by combining the distributional scores of multiple pre-trained policies via a convex combination and test-time search. GPC is versatile, allowing for the plug-and-play composition of heterogeneous policies, including VA and VLA models, as well as those based on diffusion or flow-matching, irrespective of their input visual modalities. Third, we provide extensive empirical validation. Experiments on Robomimic, PushT, and RoboTwin benchmarks, alongside real-world robotic evaluations, confirm that GPC consistently improves performance and adaptability across a diverse set of tasks. Further analysis of alternative composition operators and weighting strategies offers insights into the mechanisms underlying the success of GPC. These results establish GPC as a simple yet effective method for improving control performance by leveraging existing policies.

Tianyu Su, Zhiqiang Zou, Ali Luo et al.

Stellar flare forecasting, a critical research frontier in astronomy, offers profound insights into stellar activity. However, the field is constrained by both the sparsity of recorded flare events and the absence of domain-specific large-scale predictive models. To address these challenges, this study introduces StellarF (Stellar Flare Forecasting), a novel large model that leverages Low-Rank (LoRA) and Adapter techniques to parameter-efficient learning for stellar flare forecasting. At its core, StellarF integrates an flare statistical information module with a historical flare record module, enabling multi-scale pattern recognition from observational data. Extensive experiments on our self-constructed datasets (derived from Kepler and TESS light curves) demonstrate that StellarF achieves state-of-the-art performance compared to existing methods. The proposed prediction paradigm establishes a novel methodological framework for advancing astrophysical research and cross-disciplinary applications.

Bingke Zhu, Xiaoxiao Wang, Minghui Jia et al.

Stellar flare events are critical observational samples for astronomical research; however, recorded flare events remain limited. Stellar flare forecasting can provide additional flare event samples to support research efforts. Despite this potential, no specialized models for stellar flare forecasting have been proposed to date. In this paper, we present extensive experimental evidence demonstrating that both stellar physical properties and historical flare records are valuable inputs for flare forecasting tasks. We then introduce FLARE (Forecasting Light-curve-based Astronomical Records via features Ensemble), the first-of-its-kind large model specifically designed for stellar flare forecasting. FLARE integrates stellar physical properties and historical flare records through a novel Soft Prompt Module and Residual Record Fusion Module. Our experiments on the publicly available Kepler light curve dataset demonstrate that FLARE achieves superior performance compared to other methods across all evaluation metrics. Finally, we validate the forecast capability of our model through a comprehensive case study.

Guiyu Zhao, Bo Qiu, A-Li Luo et al.

The Wide-field Infrared Survey Explorer (WISE) has detected hundreds of millions of sources over the entire sky. However, classifying them reliably is a great challenge due to degeneracies in WISE multicolor space and low detection levels in its two longest-wavelength bandpasses. In this paper, the deep learning classification network, IICnet (Infrared Image Classification network), is designed to classify sources from WISE images to achieve a more accurate classification goal. IICnet shows good ability on the feature extraction of the WISE sources. Experiments demonstrates that the classification results of IICnet are superior to some other methods; it has obtained 96.2% accuracy for galaxies, 97.9% accuracy for quasars, and 96.4% accuracy for stars, and the Area Under Curve (AUC) of the IICnet classifier can reach more than 99%. In addition, the superiority of IICnet in processing infrared images has been demonstrated in the comparisons with VGG16, GoogleNet, ResNet34, MobileNet, EfficientNetV2, and RepVGG-fewer parameters and faster inference. The above proves that IICnet is an effective method to classify infrared sources.