Yaofo Chen

Shuaicheng Niu, Jiaxiang Wu, Yifan Zhang et al.

Test-time adaptation (TTA) seeks to tackle potential distribution shifts between training and testing data by adapting a given model w.r.t. any testing sample. This task is particularly important for deep models when the test environment changes frequently. Although some recent attempts have been made to handle this task, we still face two practical challenges: 1) existing methods have to perform backward computation for each test sample, resulting in unbearable prediction cost to many applications; 2) while existing TTA solutions can significantly improve the test performance on out-of-distribution data, they often suffer from severe performance degradation on in-distribution data after TTA (known as catastrophic forgetting). In this paper, we point out that not all the test samples contribute equally to model adaptation, and high-entropy ones may lead to noisy gradients that could disrupt the model. Motivated by this, we propose an active sample selection criterion to identify reliable and non-redundant samples, on which the model is updated to minimize the entropy loss for test-time adaptation. Furthermore, to alleviate the forgetting issue, we introduce a Fisher regularizer to constrain important model parameters from drastic changes, where the Fisher importance is estimated from test samples with generated pseudo labels. Extensive experiments on CIFAR-10-C, ImageNet-C, and ImageNet-R verify the effectiveness of our proposed method.

Shuaicheng Niu, Jiaxiang Wu, Yifan Zhang et al.

Test-time adaptation (TTA) has shown to be effective at tackling distribution shifts between training and testing data by adapting a given model on test samples. However, the online model updating of TTA may be unstable and this is often a key obstacle preventing existing TTA methods from being deployed in the real world. Specifically, TTA may fail to improve or even harm the model performance when test data have: 1) mixed distribution shifts, 2) small batch sizes, and 3) online imbalanced label distribution shifts, which are quite common in practice. In this paper, we investigate the unstable reasons and find that the batch norm layer is a crucial factor hindering TTA stability. Conversely, TTA can perform more stably with batch-agnostic norm layers, \ie, group or layer norm. However, we observe that TTA with group and layer norms does not always succeed and still suffers many failure cases. By digging into the failure cases, we find that certain noisy test samples with large gradients may disturb the model adaption and result in collapsed trivial solutions, \ie, assigning the same class label for all samples. To address the above collapse issue, we propose a sharpness-aware and reliable entropy minimization method, called SAR, for further stabilizing TTA from two aspects: 1) remove partial noisy samples with large gradients, 2) encourage model weights to go to a flat minimum so that the model is robust to the remaining noisy samples. Promising results demonstrate that SAR performs more stably over prior methods and is computationally efficient under the above wild test scenarios.

Yong Guo, Yaofo Chen, Yin Zheng et al.

Designing feasible and effective architectures under diverse computational budgets, incurred by different applications/devices, is essential for deploying deep models in real-world applications. To achieve this goal, existing methods often perform an independent architecture search process for each target budget, which is very inefficient yet unnecessary. More critically, these independent search processes cannot share their learned knowledge (i.e., the distribution of good architectures) with each other and thus often result in limited search results. To address these issues, we propose a Pareto-aware Neural Architecture Generator (PNAG) which only needs to be trained once and dynamically produces the Pareto optimal architecture for any given budget via inference. To train our PNAG, we learn the whole Pareto frontier by jointly finding multiple Pareto optimal architectures under diverse budgets. Such a joint search algorithm not only greatly reduces the overall search cost but also improves the search results. Extensive experiments on three hardware platforms (i.e., mobile device, CPU, and GPU) show the superiority of our method over existing methods.

Yaofo Chen, Yong Guo, Daihai Liao et al.

Neural Architecture Search (NAS) aims to automatically find effective architectures within a predefined search space. However, the search space is often extremely large. As a result, directly searching in such a large search space is non-trivial and also very time-consuming. To address the above issues, in each search step, we seek to limit the search space to a small but effective subspace to boost both the search performance and search efficiency. To this end, we propose a novel Neural Architecture Search method via Dominative Subspace Mining (DSM-NAS) that finds promising architectures in automatically mined subspaces. Specifically, we first perform a global search, i.e ., dominative subspace mining, to find a good subspace from a set of candidates. Then, we perform a local search within the mined subspace to find effective architectures. More critically, we further boost search performance by taking well-designed/ searched architectures to initialize candidate subspaces. Experimental results demonstrate that DSM-NAS not only reduces the search cost but also discovers better architectures than state-of-the-art methods in various benchmark search spaces.

Zeng You, Yaofo Chen, Qiuwu Chen et al.

Large language models (LLMs) face significant challenges in processing long contexts due to the linear growth of the key-value (KV) cache and quadratic complexity of self-attention. Existing approaches address these bottlenecks separately: Multi-head Latent Attention (MLA) reduces the KV cache by projecting tokens into a low-dimensional latent space, while sparse attention reduces computation. However, sparse methods cannot operate natively on MLA's compressed latent structure, missing opportunities for joint optimization. In this paper, we propose Latent-Condensed Attention (LCA), which directly condenses context within MLA's latent space, where the representation is disentangled into semantic latent vectors and positional keys. LCA separately aggregates semantic vectors via query-aware pooling and preserves positional keys via anchor selection. This approach jointly reduces both computational cost and KV cache without adding parameters. Beyond MLA, LCA's design is architecture-agnostic and readily extends to other attention mechanisms such as GQA. Theoretically, we prove a length-independent error bound. Experiments show LCA achieves up to 2.5$\times$ prefilling speedup and 90% KV cache reduction at 128K context while maintaining competitive performance.

Qianyue Wang, Jinwu Hu, Yufeng Wang et al.

Large Reasoning Models (LRMs) excel at multi-step reasoning but often suffer from inefficient reasoning processes like overthinking and overshoot, where excessive or misdirected reasoning increases computational cost and degrades performance. Existing efficient reasoning methods operate in a closed-loop manner, lacking mechanisms for external intervention to guide the reasoning process. To address this, we propose Think-with-Me, a novel test-time interactive reasoning paradigm that introduces external feedback intervention into the reasoning process. Our key insights are that transitional conjunctions serve as natural points for intervention, signaling phases of self-validation or exploration and using transitional words appropriately to prolong the reasoning enhances performance, while excessive use affects performance. Building on these insights, Think-with-Me pauses reasoning at these points for external feedback, adaptively extending or terminating reasoning to reduce redundancy while preserving accuracy. The feedback is generated via a multi-criteria evaluation (rationality and completeness) and comes from either human or LLM proxies. We train the target model using Group Relative Policy Optimization (GRPO) to adapt to this interactive mode. Experiments show that Think-with-Me achieves a superior balance between accuracy and reasoning length under limited context windows. On AIME24, Think-with-Me outperforms QwQ-32B by 7.19% in accuracy while reducing average reasoning length by 81% under an 8K window. The paradigm also benefits security and creative tasks.

Fanpu Cao, Yaofo Chen, Zeng You et al.

Diffusion Transformers (DiTs) have achieved state-of-the-art performance in generative modeling, yet their high computational cost hinders real-time deployment. While feature caching offers a promising training-free acceleration solution by exploiting temporal redundancy, existing methods suffer from two key limitations: (1) uniform caching intervals fail to align with the non-uniform temporal dynamics of DiT, and (2) naive feature reuse with excessively large caching intervals can lead to severe error accumulation. In this work, we analyze the evolution of DiT features during denoising and reveal that both feature changes and error propagation are highly time- and depth-varying. Motivated by this, we propose ProCache, a training-free dynamic feature caching framework that addresses these issues via two core components: (i) a constraint-aware caching pattern search module that generates non-uniform activation schedules through offline constrained sampling, tailored to the model's temporal characteristics; and (ii) a selective computation module that selectively computes within deep blocks and high-importance tokens for cached segments to mitigate error accumulation with minimal overhead. Extensive experiments on PixArt-alpha and DiT demonstrate that ProCache achieves up to 1.96x and 2.90x acceleration with negligible quality degradation, significantly outperforming prior caching-based methods.

Qianyue Wang, Jinwu Hu, Huanxiang Lin et al.

Reasoning in Large Language Models (LLMs) often suffers from inefficient long chain-of-thought traces with redundant self-exploration and validation, which inflate computational costs and even degrade performance. Inspired by human reasoning patterns where people solve new problems by leveraging past related cases to constrain search spaces and reduce trial-and-error, we propose Precedent Informed Reasoning (PIR) transforming LRMs'reasoning paradigm from exhaustive self-exploration to guided learning from precedents. PIR addresses two key challenges: what precedents to adopt and how to utilize them. First, Adaptive Precedent Selection (APS) constructs, for each question and LRM, a compact set of precedents that are both semantically related and informative for the model. It ranks examples by a joint score with semantic similarity and model perplexity, then adapts the amount of precedents to maximize perplexity reduction. Second, Test-time Experience Internalization (TEI) is treated as the test-time learning on precedent-informed instruction, updating lightweight adapters to internalize solution patterns and use them as a prior during subsequent reasoning. Experiments across mathematical reasoning, scientific QA, and code generation demonstrate that PIR consistently shortens reasoning traces while maintaining or improving final accuracy across LLMs, yielding outstanding accuracy-efficiency trade-offs.

Zeng You, Yaofo Chen, Shuhai Zhang et al.

Large Language Models (LLMs) have demonstrated remarkable capabilities across a wide range of natural language processing tasks. These capabilities stem primarily from the self-attention mechanism, which enables modeling of long-range dependencies. However, the quadratic complexity of self-attention with respect to sequence length poses significant computational and memory challenges, especially as sequence length extends to extremes. While various sparse attention and KV cache compression methods have been proposed to improve efficiency, they often suffer from limitations such as reliance on fixed patterns, inability to handle both prefilling and decoding stages, or the requirement for additional training. In this paper, we propose Training-free Context-adaptive Attention (TCA-Attention), a training-free sparse attention mechanism that selectively attends to only the informative tokens for efficient long-context inference. Our method consists of two lightweight phases: i) an offline calibration phase that determines head-specific sparsity budgets via a single forward pass, and ii) an online token selection phase that adaptively retains core context tokens using a lightweight redundancy metric. TCA-Attention provides a unified solution that accelerates both prefilling and decoding while reducing KV cache memory footprint, without requiring parameter updates or architectural changes. Theoretical analysis shows that our approach maintains bounded approximation error. Extensive experiments demonstrate that TCA-Attention achieves a 2.8$\times$ speedup and reduces KV cache by 61% at 128K context length while maintaining performance comparable to full attention across various benchmarks, offering a practical plug-and-play solution for efficient long-context inference.

Shuhai Zhang, Zeng You, Yaofo Chen et al.

Transformer-based large language models (LLMs) excel in natural language processing tasks by capturing long-range dependencies through self-attention mechanisms. However, long-context modeling faces significant computational inefficiencies due to \textit{redundant} attention computations: while attention weights are often \textit{sparse}, all tokens consume \textit{equal} computational resources. In this paper, we reformulate traditional probabilistic sequence modeling as a \textit{supervised learning task}, enabling the separation of relevant and irrelevant tokens and providing a clearer understanding of redundancy. Based on this reformulation, we theoretically analyze attention sparsity, revealing that only a few tokens significantly contribute to predictions. Building on this, we formulate attention optimization as a linear coding problem and propose a \textit{group coding strategy}, theoretically showing its ability to improve robustness against random noise and enhance learning efficiency. Motivated by this, we propose \textit{Dynamic Group Attention} (DGA), which leverages the group coding to explicitly reduce redundancy by aggregating less important tokens during attention computation. Empirical results show that our DGA significantly reduces computational costs while maintaining competitive performance.Code is available at https://github.com/bolixinyu/DynamicGroupAttention.

Yaofo Chen, Shuaicheng Niu, Yaowei Wang et al.

The conventional deep learning paradigm often involves training a deep model on a server and then deploying the model or its distilled ones to resource-limited edge devices. Usually, the models shall remain fixed once deployed (at least for some period) due to the potential high cost of model adaptation for both the server and edge sides. However, in many real-world scenarios, the test environments may change dynamically (known as distribution shifts), which often results in degraded performance. Thus, one has to adapt the edge models promptly to attain promising performance. Moreover, with the increasing data collected at the edge, this paradigm also fails to further adapt the cloud model for better performance. To address these, we encounter two primary challenges: 1) the edge model has limited computation power and may only support forward propagation; 2) the data transmission budget between cloud and edge devices is limited in latency-sensitive scenarios. In this paper, we establish a Cloud-Edge Elastic Model Adaptation (CEMA) paradigm in which the edge models only need to perform forward propagation and the edge models can be adapted online. In our CEMA, to reduce the communication burden, we devise two criteria to exclude unnecessary samples from uploading to the cloud, i.e., dynamic unreliable and low-informative sample exclusion. Based on the uploaded samples, we update and distribute the affine parameters of normalization layers by distilling from the stronger foundation model to the edge model with a sample replay strategy. Extensive experimental results on ImageNet-C and ImageNet-R verify the effectiveness of our CEMA.

Zeng You, Zhiquan Wen, Yaofo Chen et al.

Long video understanding has become a critical task in computer vision, driving advancements across numerous applications from surveillance to content retrieval. Existing video understanding methods suffer from two challenges when dealing with long video understanding: intricate long-context relationship modeling and interference from redundancy. To tackle these challenges, we introduce Fine-Detailed Video Story generation (FDVS), which interprets long videos into detailed textual representations. Specifically, to achieve fine-grained modeling of long-temporal content, we propose a Bottom-up Video Interpretation Mechanism that progressively interprets video content from clips to video. To avoid interference from redundant information in videos, we introduce a Semantic Redundancy Reduction mechanism that removes redundancy at both the visual and textual levels. Our method transforms long videos into hierarchical textual representations that contain multi-granularity information of the video. With these representations, FDVS is applicable to various tasks without any fine-tuning. We evaluate the proposed method across eight datasets spanning three tasks. The performance demonstrates the effectiveness and versatility of our method.

Mingkui Tan, Guohao Chen, Jiaxiang Wu et al.

Test-time adaptation (TTA) seeks to tackle potential distribution shifts between training and test data by adapting a given model w.r.t. any test sample. Although recent TTA has shown promising performance, we still face two key challenges: 1) prior methods perform backpropagation for each test sample, resulting in unbearable optimization costs to many applications; 2) while existing TTA can significantly improve the test performance on out-of-distribution data, they often suffer from severe performance degradation on in-distribution data after TTA (known as forgetting). To this end, we have proposed an Efficient Anti-Forgetting Test-Time Adaptation (EATA) method which develops an active sample selection criterion to identify reliable and non-redundant samples for test-time entropy minimization. To alleviate forgetting, EATA introduces a Fisher regularizer estimated from test samples to constrain important model parameters from drastic changes. However, in EATA, the adopted entropy loss consistently assigns higher confidence to predictions even for samples that are underlying uncertain, leading to overconfident predictions. To tackle this, we further propose EATA with Calibration (EATA-C) to separately exploit the reducible model uncertainty and the inherent data uncertainty for calibrated TTA. Specifically, we measure the model uncertainty by the divergence between predictions from the full network and its sub-networks, on which we propose a divergence loss to encourage consistent predictions instead of overconfident ones. To further recalibrate prediction confidence, we utilize the disagreement among predicted labels as an indicator of the data uncertainty, and then devise a min-max entropy regularizer to selectively increase and decrease prediction confidence for different samples. Experiments on image classification and semantic segmentation verify the effectiveness of our methods.

Qi Deng, Shuaicheng Niu, Ronghao Zhang et al.

Test-time adaptation (TTA) aims to fine-tune a trained model online using unlabeled testing data to adapt to new environments or out-of-distribution data, demonstrating broad application potential in real-world scenarios. However, in this optimization process, unsupervised learning objectives like entropy minimization frequently encounter noisy learning signals. These signals produce unreliable gradients, which hinder the model ability to converge to an optimal solution quickly and introduce significant instability into the optimization process. In this paper, we seek to resolve these issues from the perspective of optimizer design. Unlike prior TTA using manually designed optimizers like SGD, we employ a learning-to-optimize approach to automatically learn an optimizer, called Meta Gradient Generator (MGG). Specifically, we aim for MGG to effectively utilize historical gradient information during the online optimization process to optimize the current model. To this end, in MGG, we design a lightweight and efficient sequence modeling layer -- gradient memory layer. It exploits a self-supervised reconstruction loss to compress historical gradient information into network parameters, thereby enabling better memorization ability over a long-term adaptation process. We only need a small number of unlabeled samples to pre-train MGG, and then the trained MGG can be deployed to process unseen samples. Promising results on ImageNet-C, R, Sketch, and A indicate that our method surpasses current state-of-the-art methods with fewer updates, less data, and significantly shorter adaptation iterations. Compared with a previous SOTA method SAR, we achieve 7.4% accuracy improvement and 4.2 times faster adaptation speed on ImageNet-C.

Yaofo Chen, Zeng You, Shuhai Zhang et al.

Transformer-based Large Language Models (LLMs) have exhibited remarkable success in extensive tasks primarily attributed to self-attention mechanism, which requires a token to consider all preceding tokens as its context to compute attention. However, when the context length L becomes very large (e.g., 128K), the amount of potentially redundant information in the context tends to increase. The redundant context not only hampers the modeling representation performance but also incurs unnecessary computational and storage overhead. In this paper, we propose a plug-and-play Core Context Aware (CCA) Attention for efficient long-context modeling, comprising two complementary modules: 1) Globality-aware pooling module groups input tokens and dynamically compresses each group into one core token based on their significance. In this way, our method automatically focuses and strengthens core context while diminishing redundancy during the learning process, leading to effective long-term dependency modeling. 2) Locality-preserving module incorporates neighboring tokens to preserve local context for detailed representation. Notably, our CCA-Attention is able to replace the self-attention module in existing LLMs with minimal fine-tuning cost. Extensive experimental results show the superiority of our method in both long-context modeling and computational efficiency over state-of-the-art methods.

Zekang Zheng, Haokun Li, Yaofo Chen et al.

Model quantization reduces neural network parameter precision to achieve compression, but often compromises accuracy. Existing post-training quantization (PTQ) methods employ iterative parameter updates to preserve accuracy under high compression ratios, incurring significant computational complexity and resource overhead, which limits applicability in resource-constrained edge computing and real-time inference scenarios. This paper proposes an efficient PTQ method guided by parameter sensitivity analysis. The approach prioritizes quantization of high-sensitivity parameters, leveraging unquantized low-sensitivity parameters to compensate for quantization errors, thereby mitigating accuracy degradation. Furthermore, by exploiting column-wise clustering of parameter sensitivity, the method introduces a row-parallel quantization framework with a globally shared inverse Hessian matrix update mechanism, reducing computational complexity by an order of magnitude. Experimental results on ResNet-50 and YOLOv5s demonstrate a 20-200-fold quantization speedup over the Optimal Brain Quantization baseline, with mean accuracy loss below 0.3%, confirming the method's efficacy in balancing efficiency and accuracy.

Shuaicheng Niu, Guohao Chen, Deyu Chen et al.

Test-time adaptation (TTA) may fail to improve or even harm the model performance when test data have: 1) mixed distribution shifts, 2) small batch sizes, 3) online imbalanced label distribution shifts. This is often a key obstacle preventing existing TTA methods from being deployed in the real world. In this paper, we investigate the unstable reasons and find that the batch norm layer is a crucial factor hindering TTA stability. Conversely, TTA can perform more stably with batch-agnostic norm layers, i.e., group or layer norm. However, we observe that TTA with group and layer norms does not always succeed and still suffers many failure cases, i.e., the model collapses into trivial solutions by assigning the same class label for all samples. By digging into this, we find that, during the collapse process: 1) the model gradients often undergo an initial explosion followed by rapid degradation, suggesting that certain noisy test samples with large gradients may disrupt adaptation; and 2) the model representations tend to exhibit high correlations and classification bias. To address this, we first propose a sharpness-aware and reliable entropy minimization method, called SAR, for stabilizing TTA from two aspects: 1) remove partial noisy samples with large gradients, 2) encourage model weights to go to a flat minimum so that the model is robust to the remaining noisy samples. Based on SAR, we further introduce SAR^2 to prevent representation collapse with two regularizers: 1) a redundancy regularizer to reduce inter-dimensional correlations among centroid-invariant features; and 2) an inequity regularizer to maximize the prediction entropy of a prototype centroid, thereby penalizing biased representations toward any specific class. Promising results demonstrate that our methods perform more stably over prior methods and are computationally efficient under the above wild test scenarios.

Jinwu Hu, Dongjin Yang, Langyu Bian et al.

Large language models (LLMs) have demonstrated impressive performance across various language tasks. However, existing LLM reasoning strategies mainly rely on the LLM itself with fast or slow mode (like o1 thinking) and thus struggle to balance reasoning efficiency and accuracy across queries of varying difficulties. In this paper, we propose Cognitive-Inspired Elastic Reasoning (CogER), a framework inspired by human hierarchical reasoning that dynamically selects the most suitable reasoning strategy for each query. Specifically, CogER first assesses the complexity of incoming queries and assigns them to one of several predefined levels, each corresponding to a tailored processing strategy, thereby addressing the challenge of unobservable query difficulty. To achieve automatic strategy selection, we model the process as a Markov Decision Process and train a CogER-Agent using reinforcement learning. The agent is guided by a reward function that balances solution quality and computational cost, ensuring resource-efficient reasoning. Moreover, for queries requiring external tools, we introduce Cognitive Tool-Assisted Reasoning, which enables the LLM to autonomously invoke external tools within its chain-of-thought. Extensive experiments demonstrate that CogER outperforms state-of-the-art Test-Time scaling methods, achieving at least a 13% relative improvement in average exact match on In-Domain tasks and an 8% relative gain on Out-of-Domain tasks.

Yong Guo, Yaofo Chen, Mingkui Tan et al.

Convolutional Neural Networks (CNNs) have achieved great success due to the powerful feature learning ability of convolution layers. Specifically, the standard convolution traverses the input images/features using a sliding window scheme to extract features. However, not all the windows contribute equally to the prediction results of CNNs. In practice, the convolutional operation on some of the windows (e.g., smooth windows that contain very similar pixels) can be very redundant and may introduce noises into the computation. Such redundancy may not only deteriorate the performance but also incur the unnecessary computational cost. Thus, it is important to reduce the computational redundancy of convolution to improve the performance. To this end, we propose a Content-aware Convolution (CAC) that automatically detects the smooth windows and applies a 1x1 convolutional kernel to replace the original large kernel. In this sense, we are able to effectively avoid the redundant computation on similar pixels. By replacing the standard convolution in CNNs with our CAC, the resultant models yield significantly better performance and lower computational cost than the baseline models with the standard convolution. More critically, we are able to dynamically allocate suitable computation resources according to the data smoothness of different images, making it possible for content-aware computation. Extensive experiments on various computer vision tasks demonstrate the superiority of our method over existing methods.

Yaofo Chen, Yong Guo, Qi Chen et al.

One of the key steps in Neural Architecture Search (NAS) is to estimate the performance of candidate architectures. Existing methods either directly use the validation performance or learn a predictor to estimate the performance. However, these methods can be either computationally expensive or very inaccurate, which may severely affect the search efficiency and performance. Moreover, as it is very difficult to annotate architectures with accurate performance on specific tasks, learning a promising performance predictor is often non-trivial due to the lack of labeled data. In this paper, we argue that it may not be necessary to estimate the absolute performance for NAS. On the contrary, we may need only to understand whether an architecture is better than a baseline one. However, how to exploit this comparison information as the reward and how to well use the limited labeled data remains two great challenges. In this paper, we propose a novel Contrastive Neural Architecture Search (CTNAS) method which performs architecture search by taking the comparison results between architectures as the reward. Specifically, we design and learn a Neural Architecture Comparator (NAC) to compute the probability of candidate architectures being better than a baseline one. Moreover, we present a baseline updating scheme to improve the baseline iteratively in a curriculum learning manner. More critically, we theoretically show that learning NAC is equivalent to optimizing the ranking over architectures. Extensive experiments in three search spaces demonstrate the superiority of our CTNAS over existing methods.

Yong Guo, Yaofo Chen, Yin Zheng et al.

Designing feasible and effective architectures under diverse computation budgets incurred by different applications/devices is essential for deploying deep models in practice. Existing methods often perform an independent architecture search for each target budget, which is very inefficient yet unnecessary. Moreover, the repeated independent search manner would inevitably ignore the common knowledge among different search processes and hamper the search performance. To address these issues, we seek to train a general architecture generator that automatically produces effective architectures for an arbitrary budget merely via model inference. To this end, we propose a Pareto-Frontier-aware Neural Architecture Generator (NAG) which takes an arbitrary budget as input and produces the Pareto optimal architecture for the target budget. We train NAG by learning the Pareto frontier (i.e., the set of Pareto optimal architectures) over model performance and computational cost (e.g., latency). Extensive experiments on three platforms (i.e., mobile, CPU, and GPU) show the superiority of the proposed method over existing NAS methods.

Yong Guo, Yaofo Chen, Yin Zheng et al.

Neural architecture search (NAS) has become an important approach to automatically find effective architectures. To cover all possible good architectures, we need to search in an extremely large search space with billions of candidate architectures. More critically, given a large search space, we may face a very challenging issue of space explosion. However, due to the limitation of computational resources, we can only sample a very small proportion of the architectures, which provides insufficient information for the training. As a result, existing methods may often produce suboptimal architectures. To alleviate this issue, we propose a curriculum search method that starts from a small search space and gradually incorporates the learned knowledge to guide the search in a large space. With the proposed search strategy, our Curriculum Neural Architecture Search (CNAS) method significantly improves the search efficiency and finds better architectures than existing NAS methods. Extensive experiments on CIFAR-10 and ImageNet demonstrate the effectiveness of the proposed method.