We adopt four bioinformatics datasets in the experiment. Given the input graph, it will randomly add or cut a certain portion ofconnections between nodes withtheprobability of0.2. It will set the feature of 20% nodes in the graph to Gaussian noises with mean and standard deviation is 0.5. We adopt the Adam [5] optimizer, which is a variant of Stochastic Gradient Descent (SGD) with adaptivemoment estimation.

Estimating counterfactual outcome of different treatments from observational data is an important problem to assist decision making in a variety of fields. Among the various forms of treatment specification, bundle treatment has been widely adopted inmanyscenarios, such asrecommendation systems andonline marketing.

We take 1000 training images from CIFAR-10 as a fixed batch, randomly sample the neural architecture for inference, and computevar(µ) of the last BN layer of a NSA and a NSA-i trained givenS = 5000architectures. Inthissection, wecalculate thetestaccuracyof200randomly sampled architectures based onthe vanilla NSA models trained under various spaces. A half of these architectures are seen during trainingwhiletheotherhalfnot.

We randomly selected action sequences from test episodes collected with action noise alongside the training episodes. Next, we analyze the IDM framework based on Janner's work [1]. Denote pθ(z |z,a) as the state transition probability predicted by model.

This work was done during the author's internship at Alibaba Group

An operation w Model weight α The architecture parameter N The number of chunks θ The trainable parameter in the soft task-collaborative module p The parameter generated by Eq.(9) p The parameter generated by Eq.(11), replacing p during curriculum training δ The parameter to control graph structure diversity γ The parameter to control task-wise curriculum training BNRist is the abbreviation of Beijing National Research Center for Information Science and Technology. Here we provide the detailed derivation process of Eq.(10). For the other datasets, we use the task-separate head. The experiment results on OGBG datasets are shown in Table 5. From the table, our method can outperform all the multi-task NAS baselines in the three datasets.

Proof A.1 W e firstly give Lemma 1: Lemma 1 The operation weights are caculated by a softmax function. The number of target node's intra-group neighbors is "S" indicates the setting of searching phase. "E" indicates the setting of evaluation phase. The hyper-parameter λ which controls the hidden feature smoothness is set to be 0 .125 . We show the variance of synthetic graph experiment in Table 1 to endorse our analysis in Section 3. The table shows that the variance of accuracy is relatively big in the experiment setting. However, all the results are average of 100 runs.

Evaluating the quality of slide-based multimedia instruction is challenging. Existing methods like manual assessment, reference-based metrics, and large language model evaluators face limitations in scalability, context capture, or bias. In this paper, we introduce LecEval, an automated metric grounded in Mayer's Cognitive Theory of Multimedia Learning, to evaluate multimodal knowledge acquisition in slide-based learning. LecEval assesses effectiveness using four rubrics: Content Relevance (CR), Expressive Clarity (EC), Logical Structure (LS), and Audience Engagement (AE). We curate a large-scale dataset of over 2,000 slides from more than 50 online course videos, annotated with fine-grained human ratings across these rubrics. A model trained on this dataset demonstrates superior accuracy and adaptability compared to existing metrics, bridging the gap between automated and human assessments. We release our dataset and toolkits at https://github.com/JoylimJY/LecEval.

Cross-view geo-localization in GNSS-denied environments aims to determine an unknown location by matching drone-view images with the correct geo-tagged satellite-view images from a large gallery. Recent research shows that learning discriminative image representations under specific weather conditions can significantly enhance performance. However, the frequent occurrence of unseen extreme weather conditions hinders progress. This paper introduces MCGF, a Multi-weather Cross-view Geo-localization Framework designed to dynamically adapt to unseen weather conditions. MCGF establishes a joint optimization between image restoration and geo-localization using denoising diffusion models. For image restoration, MCGF incorporates a shared encoder and a lightweight restoration module to help the backbone eliminate weather-specific information. For geo-localization, MCGF uses EVA-02 as a backbone for feature extraction, with cross-entropy loss for training and cosine distance for testing. Extensive experiments on University160k-WX demonstrate that MCGF achieves competitive results for geo-localization in varying weather conditions.