EDBT 2026 Demo / reviewers in the wild / expert
Xinkai Song
dblp:221/6104
· DBLP profile ↗
28ranked-venue papers
2as first author
28since 2021 · last 2026
0000-0003-3458-1555ORCID · corroborated
Domains — the database's venue-derived domains; a paper can count in several
Systems, architecture and hardware · 18 · 2 first-author · 18 since 2021Artificial intelligence and machine learning · 7 · 7 since 2021Software engineering, systems software and programming languages · 6 · 6 since 2021Graphics, computer vision, multimedia, augmented reality and games · 3 · 3 since 2021Applied, interdisciplinary, general and emerging computing · 1 · 1 since 2021
| Year | Publication | Venue | Position |
|---|---|---|---|
| 2026 | Cambricon-GS: An Accelerator for 3D Gaussian Splatting Training With Gaussian-Pixel Hybrid Parallelismabstract3D Gaussian Splatting (3DGS) is a breakthrough in 3D reconstruction using 3D Gaussians. However, even on highend GPUs like the NVIDIA A100, reconstructing complex scenes remains time-consuming, taking over 15 minutes. The main bottleneck is α-computation, which accounts for 71.25 % of training workload, yet 93.03 % of it is invalid due to the localized influence of Gaussians. To address this issue, we propose Cambricon-GS, an accelerator for 3DGS training with Gaussian-Pixel hybrid parallelism. At the software level, we introduce a hybrid parallel workflow that breaks the limitation of conventional pixel-only parallelism through two key techniques: Center-Pixel Gaussian Culling (CPGC), which eliminates invalid Gaussians early, and SeedDriven Gaussian Region Exploration (SDGRE), which reduces invalid computation for partially valid Gaussians by selectively exploring valid regions. Overall, the workflow significantly reduces α computations, lowering the workload to 17.99 %. At the hardware level, Cambricon-GS decouples α-computation and α blending into GUnits and PUnits, organized in a 2D mesh-based NoC that supports asynchronous execution and efficient data routing. We further boost performance via Gaussian/Pixel load balancing and tiled SSIM based pipelining. The evaluation results show that Cambricon-GS achieves 19.63×, 14.86×, 15.42×, 2.98× and 2.63× speedup, and 78.62×, 63.00×, 61.72×, 3.89× and 3.22× energy saving, compared to A100, GSCore, GBU, GSArch, and GauSPU, respectively, with negligible image quality loss. Zhifei Yue, Tianbo Liu 0006, Xinkai Song, Jiaming Guo, Xing Hu 0001, Zidong Du, Qi Guo 0001, Tianshi Chen 0002 |
HPCA | 4 |
| 2026 | Cambricon-QM: A Hybrid Architecture for Microscaling Format Training
Yongwei Zhao 0001, Chang Liu 0021, Zidong Du, Xing Hu 0001, Yimin Zhuang, Yifan Hao 0001, Xinkai Song, Wei Li 0008, Xishan Zhang, Ling Li 0001, Zhiwei Xu 0002, Tianshi Chen 0002, Qi Guo 0001 |
IEEE Trans. Computers | 9 |
| 2026 | TensorFHE+: Fully Homomorphic Encryption Acceleration Based on Linear AlgebraabstractFully Homomorphic Encryption (FHE) enables encrypted data processing on untrusted cloud servers, crucial for privacy-sensitive applications. Despite its potential, performance overheads (about 10, 000× slower) limit adoption. ASIC accelerators outperform GPUs/FPGAs by optimizing specific operations but rely on costly 7nm processes and large on-chip memory, hindering cost-effective deployment. Balancing efficiency with manufacturing constraints remains critical. This paper presents TensorFHE+, a GPU-optimized FHE acceleration framework leveraging Tensor Cores to accelerate Number Theoretic Transform (NTT) operations. Key innovations include: 1) Decomposing CKKS kernels into vector/matrix operations for hardware utilization; 2) Vectorized modulo arithmetic; 3) Data layout optimization for memory efficiency. Evaluated on NVIDIA A100, TensorFHE+ outperforms TensorFHE [1] by 1.44× in average (up to 1.69× on ResNet-20) and surpasses prior GPU implementations [2], [3]. The design also demonstrates compatibility with commercial linear algebra accelerators, enabling efficient FHE deployment. Yintai Sun, Shengyu Fan, Zhenhua Yin, Xinkai Song, Xing Hu 0001, Zidong Du, Qi Guo 0001, Weizhi Xu 0001, Rui Hou 0001, Dan Meng 0002, Song Bian 0001, Mingzhe Zhang 0005 |
IEEE Trans. Computers | 4 |
| 2026 | DASA: Distribution-Aware Sparse Attention for Accelerating Diffusion TransformerabstractDiffusion Transformers (DiTs) have demonstrated remarkable success in text-to-video generation. However, the self-attention mechanism in DiTs imposes significant computational and memory burdens, particularly when handling long patch sequences like high-resolution or long-time videos. While sparse attention shows promise in reducing self-attention costs, existing approaches struggle to deliver performance gains due to the unique challenges in DiTs,i.e., varied sparse patterns across layers and timesteps, and the cumulative nature of inference errors over timesteps. In this paper, we propose DASA, an algorithm-hardware co-design that effectively addresses these challenges of attention sparsification in DiTs. Specifically, leveraging the insight that the generation quality is primarily influenced by overall distribution drift rather than changes in specific values, we introduce a novel Distribution-Aware Filtering (DAF) mechanism for sparsification. To further accelerate the process, we design a specialized Filtering Unit that enables fast candidate selection based on the proposed DAF mechanism. Experimental results show that DASA achieves 2.52× speed up compared to A100 GPU, and up to 1.22× speedup over state-of-the-art accelerators for self-attention computation. Tianyun Ma, Jiaming Guo, Xinkai Song, Husheng Han, Pengwei Jin, Xiangtao Guan, Yifan Hao 0001, Yuanbo Wen 0001, Shuyao Cheng, Zidong Du, Rui Zhang 0040, Xing Hu 0001, Qi Guo 0001 |
IEEE Trans. Comput. Aided Des. Integr. Circuits Syst. | 4 |
| 2025 | Mosaic: Exploiting Instruction-Level Parallelism on Deep Learning Accelerators with iTex TessellationabstractDeep learning has achieved great success in numerous application areas at the cost of high computational complexity. To meet the ever-increasing computational demand, commodity hardware platforms (e.g., CPUs and GPUs) offer abundant computing resources including scalar, vector, and tensor units for deep learning that could execute in parallel. However, existing top-down tiling-based deep learning compilers often generate a homogeneous mapping from the given tensor computation task to hardware arithmetic instructions, failing to utilize different computing units simultaneously to achieve higher performance. Jianxing Xu, Yuanbo Wen 0001, Ruibai Xu, Tingfeng Ruan, Jun Bi, Rui Zhang 0040, Xinkai Song, Yifan Hao 0001, Xing Hu 0001, Zidong Du, Chongqing Zhao, Jiang Jie, Qi Guo 0001 |
ASPLOS (2) | 9 |
| 2025 | SEEN-DA: SEmantic ENtropy guided Domain-aware Attention for Domain Adaptive Object DetectionabstractDomain adaptive object detection (DAOD) aims to generalize detectors trained on an annotated source domain to an unlabelled target domain. Traditional works focus on aligning visual features between domains to extract domain-invariant knowledge, and recent VLM-based DAOD methods leverage semantic information provided by the textual encoder to supplement domain-specific features for each domain. However, they overlook the role of semantic information in guiding the learning of visual features that are beneficial for adaptation. To solve the problem, we propose semantic entropy to quantify the semantic information contained in visual features, and design SEmantic ENtropy guided Domain-aware Attention (SEEN-DA) to adaptively refine visual features with the semantic information of two domains. Semantic entropy reflects the importance of features based on semantic information, which can serve as attention to select discriminative visual features and suppress semantically irrelevant redundant information. Guided by semantic entropy, we introduce domain-aware attention modules into the visual encoder in SEEN-DA. It utilizes an inter-domain attention branch to extract domain-invariant features and eliminate redundant information, and an intra-domain attention branch to supplement the domain-specific semantic information discriminative on each domain. Comprehensive experiments validate the effectiveness of SEEN-DA, demonstrating significant improvements in cross-domain object detection performance. Haochen Li 0002, Rui Zhang 0040, Hantao Yao, Xin Zhang 0062, Yifan Hao 0001, Xinkai Song, Shaohui Peng, Yongwei Zhao 0001, Ling Li 0001 |
CVPR | 6 |
| 2025 | Cambricon-DG: An Accelerator for Redundant-Free Dynamic Graph Neural Networks Based on Nonlinear Isolation
Zhifei Yue, Xinkai Song, Tianbo Liu 0006, Xing Hu 0001, Rui Zhang 0040, Zidong Du, Wei Li 0008, Qi Guo 0001, Tianshi Chen 0002 |
HPCA | 2 |
| 2025 | Cambricon-SR: An Accelerator for Neural Scene Representation with Sparse Encoding TableabstractNeural Scene Representation (NSR) is a promising technique for representing real scenes.By learning from dozens of 2D photos captured from different viewpoints, NSR computes the 3D representation of real scenes.However, the performance of NSR processing running on GPU is insufficient for applications.Cambricon-R achieves high performance of more than 60 scenes per second, but at the cost of modeling quality. Tianbo Liu 0006, Xinkai Song, Zhifei Yue, Xing Hu 0001, Zhuoran Song, Yuanbo Wen 0001, Yifan Hao 0001, Wei Li 0008, Zidong Du, Rui Zhang 0040, Jiaming Guo, Shaohui Peng, Guangzhong Sun, Qi Guo 0001, Tianshi Chen 0002 |
ISCA | 2 |
| 2025 | QiMeng-Xpiler: Transcompiling Tensor Programs for Deep Learning Systems with a Neural-Symbolic Approach
Shouyang Dong, Jun Bi, Jiaming Guo, Jianxing Xu, Ruibai Xu, Xinkai Song, Yifan Hao 0001, Ling Li 0001, Xuehai Zhou, Tianshi Chen 0002, Qi Guo 0001, Yunji Chen |
OSDI | 7 |
| 2025 | Harmonia: A Unified Architecture for Efficient Deep Symbolic RegressionabstractSymbolic regression (SR), the process of formulating a mathematical expression based on observed data points, is a fundamental task in artificial intelligence but is often hindered by its intense computational demands. Deep-learning-based SR methods (DSR) aim to alleviate these demands by breaking down the SR process into two stages: 1) neural network (NN) inference and 2) Broyden-Fletcher–Goldfarb-Shanno (BFGS) optimization. Although NN accelerators can expedite the NN stage, the performance of the BFGS optimization is compromised due to its poor performance for the variety of transcendental functions. Moreover, the distinct computational characteristics of NN inference and BFGS cause not only low hardware utilization but also significant area waste. To address these issues, we propose Harmonia, a unified architecture with the neural transcendental function unit (NTFU) and the Unified Array for efficient DSR. The NTFU utilizes the radial basis function network (RBFN) as a universal approximator for various transcendental functions, which significantly reduces the heavy transcendental function computation cost. We further propose an efficient training algorithm called random nonlinear optimization (RNO) to obtain a lightweight RBFN without accuracy loss. Moreover, Harmonia supports configurable dataflow which integrates the two computing stages into the Unified Array. Experimental results show that Harmonia achieves hardware utilization of 83.83%, on average. Compared to the GPU baseline, Harmonia achieves$4.8\times $speedup and$47.6\times $energy saving, alongside considerable low area cost. Tianyun Ma, Yuanbo Wen 0001, Xinkai Song, Pengwei Jin, Husheng Han, Ziyuan Nan, Zhongkai Yu, Shaohui Peng, Yongwei Zhao 0001, Huaping Chen 0001, Zidong Du, Xing Hu 0001, Qi Guo 0001 |
IEEE Trans. Comput. Aided Des. Integr. Circuits Syst. | 3 |
| 2024 | Emergent Communication for Numerical Concepts GeneralizationabstractResearch on emergent communication has recently gained significant traction as a promising avenue for the linguistic community to unravel human language's origins and explore artificial intelligence's generalization capabilities. Current research has predominantly concentrated on recognizing qualitative patterns of object attributes(e.g., shape and color) and paid little attention to the quantitative relationship among object quantities which is known as the part of numerical concepts. The ability to generalize numerical concepts, i.e., counting and calculations with unseen quantities, is essential, as it mirrors humans' foundational abstract reasoning abilities. In this work, we introduce the NumGame, leveraging the referential game framework, forcing agents to communicate and generalize the numerical concepts effectively. Inspired by the human learning process of numbers, we present a two-stage training approach that sequentially fosters a rudimentary numerical sense followed by the ability of arithmetic calculation, ultimately aiding agents in generating semantically stable and unambiguous language for numerical concepts. The experimental results indicate the impressive generalization capabilities to unseen quantities and regularity of the language emergence from communication. Enshuai Zhou, Yifan Hao 0001, Rui Zhang 0040, Zidong Du, Xishan Zhang, Xinkai Song, Chao Wang 0003, Xuehai Zhou, Jiaming Guo, Qi Yi, Shaohui Peng, Ruizhi Chen, Qi Guo 0001, Yunji Chen |
AAAI | 7 |
| 2024 | TensorTEE: Unifying Heterogeneous TEE Granularity for Efficient Secure Collaborative Tensor ComputingabstractHeterogeneous collaborative computing with NPU and CPU has received widespread attention due to its substantial performance benefits. To ensure data confidentiality and integrity during computing, Trusted Execution Environments (TEE) is considered a promising solution because of its comparatively lower overhead. However, existing heterogeneous TEE designs are inefficient for collaborative computing due to fine and different memory granularities between CPU and NPU. 1) The cacheline granularity of CPU TEE intensifies memory pressure due to its extra memory access, and 2) the cacheline granularity MAC of NPU escalates the pressure on the limited memory storage. 3) Data transfer across heterogeneous enclaves relies on the transit of non-secure regions, resulting in cumbersome re-encryption and scheduling. Husheng Han, Xinyao Zheng, Yuanbo Wen 0001, Yifan Hao 0001, Erhu Feng, Ling Liang 0003, Jianan Mu, Xiaqing Li, Tianyun Ma, Pengwei Jin, Xinkai Song, Zidong Du, Qi Guo 0001, Xing Hu 0001 |
ASPLOS (4) | 11 |
| 2024 | AutoOS: Make Your OS More Powerful by Exploiting Large Language ModelsabstractWith the rapid development of Artificial Intelligence of Things (AIoT), customizing and optimizing operating system (OS) kernel configurations for various AIoT application scenarios is crucial for maximizing system performance. However, existing approaches falter due to the overwhelming problem complexity (i.e., over 15,000 configuration options in the Linux kernel), together with the huge evaluation costs and error-prone options that may result in OS boot-up failure, which all make it an unresolved problem to optimize the Linux kernel automatically. In this paper, we introduce AutoOS, a novel framework exploiting Large Language Models for customizing and optimizing OS kernel configurations automatically for various AIoT application scenarios.Inspired by the inherently directory-structured kernel configuration process, we first formulate our research problem as optimizing on a dynamic tree. We then propose a novel framework integrating a state machine-based traversal algorithm as the observe-prune-propose-act-correct loop, which can effectively refine the optimization space and ensure a successful OS boot-up.Experimental results show that AutoOS can automatically customize and optimize the OS kernel configurations without human effort. More importantly, AutoOS even achieves better performance by up to 25% than vendor-provided configuration. Huilai Chen, Yuanbo Wen 0001, Limin Cheng, Shouxu Kuang, Ling Li 0001, Rui Zhang 0040, Xinkai Song, Wei Li 0008, Qi Guo 0001, Yunji Chen |
ICML | 9 |
| 2024 | Cambricon-D: Full-Network Differential Acceleration for Diffusion ModelsabstractDiffusion models have made significant progress in current image generation tasks, thus becoming a prominent area of research. Diffusion models necessitate repetitive iterations on minimally altered input data across timesteps, each timestep requiring the recalculation of the entire model, resulting in a remarkable computational redundancy and substantial hardware expenditures.Performing differential computing on input data seems to be a feasible approach for addressing such computational redundancy and improving hardware efficacy. However, non-linear operations (particularly activation functions) necessitate the merging of deltas (i.e., differential values) with raw inputs repeatedly to ensure computational correctness, leading to significant memory access for loading raw inputs, which fragmentedly blocks the forwarding of deltas throughout the network and undermines performance.To solve this problem, we propose Cambricon-D, a fullnetwork differential computing architecture with concise memory access. While maintaining the computational efficiency brought by differential computing, Cambricon-D employs a sign-mask dataflow, which requires only the loading of 1-bit signs (instead of large bitwidth raw inputs), thereby facilitating the seamless forwarding of deltas and effectively mitigating memory access overheads. Experimental results show that, compared to Diffy, Cambricon-D’s dataflow reduces 66% ~ 82% off-chip memory access. In total, Cambricon-D achieves 1.46× ~ 2.38× speedup over A100 on various diffusion models with different resolutions. Weihao Kong, Yifan Hao 0001, Qi Guo 0001, Yongwei Zhao 0001, Xinkai Song, Xiaqing Li, Mo Zou, Zidong Du, Rui Zhang 0040, Chang Liu 0021, Yuanbo Wen 0001, Pengwei Jin, Xing Hu 0001, Wei Li 0008, Zhiwei Xu 0002, Tianshi Chen 0002 |
ISCA | 5 |
| 2024 | Cambricon-C: Efficient 4-Bit Matrix Unit via PrimitivizationabstractDeep learning trends to use low precision numeral formats to cope with the ever-growing model sizes. For example, the large language model LLaMA2 has been widely deployed in 4-bit precision. With larger models and fewer unique values caused by low precision, an increasing proportion of arithmetic in matrix multiplication is repeating. Although discussed in prior works, such value redundancy has not been fully exploited, and the cost to leverage the value redundancy often offsets any advantages. In this paper, we propose to primitivize the matrix multiplication, that is decomposing it down to the 1-ary successor function (a.k.a. counting) to merge repeating arithmetic. We revisited various techniques to propose Cambricon-C SA, a 4-bit primitive matrix multiplication unit that doubles the energy efficiency over conventional systolic arrays. Experimental results show that Cambricon-C SA can achieve$\mathbf{1}.\mathbf{95}\times$energy efficiency improvement compared with MAC-based systolic array. Yongwei Zhao 0001, Yifan Hao 0001, Yuanbo Wen 0001, Yuntao Dai, Xiaqing Li, Yang Liu 0466, Rui Zhang 0040, Mo Zou, Xinkai Song, Xing Hu 0001, Zidong Du, Huaping Chen 0001, Qi Guo 0001, Tianshi Chen 0002 |
MICRO | 10 |
| 2024 | SRender: Boosting Neural Radiance Field Efficiency via Sensitivity-Aware Dynamic Precision RenderingabstractNeural Radiance Field (NeRF) holds immense promise for generating photo-realistic images and videos. How-ever, the computation and memory demands significantly impede its applicability in real-time tasks such as virtual reality and augmented reality. While accelerators dedicatedly designed for NeRF have emerged, they may not be the optimal solution as they only consider the optimization of full-precision NeRF models, leading to unsatisfactory memory and computation saving. This paper proposes Sensitivity-Aware Dynamic Precision Rendering, abbreviated as SRender, an algorithm-hardware co-design framework that borrows the idea of Adaptive Rendering commonly used in conventional graphics rendering to accelerate the rendering process of NeRF exploiting the fine-grained per-ray and per-point levels sensitivity. The fundamental idea behind SRender is that rays and points with high sensitivity require high-precision data types, while the insensitive rays and points do not, allowing them to be sacrificed for greater acceleration. Accordingly, the algorithm part of SRender dynamically configures the precision of the hash table and the multilayer perceptron (MLP), and the hardware part features a specialized accelerator that exploits such mixed-precision memory access and computation patterns. Specifically, we predict the sensitive rays and points based on low-precision hash encoding and feature computation. Then, we trigger a high-precision rendering again on the sensitive ones to guarantee the rendering quality. The hardware design of SRender includes a sensitivity prediction engine, coarse-then-fine-grained encoding engines, and a recovery engine, which translates the computational saving in the algorithm to real speedup. Using the coarse-then-fine-grained encoding engine, we maximize the data reuse on-chip, thereby reducing the stress of off-chip memory bandwidth. Experiments show that the edge-level SRender can achieve 3.94x and 34.06x speedup over state-of-the-art NeRF accelerator NeuRex and Jetson AGX Xavier GPU with negligible rendering quality loss. Zhuoran Song, Houshu He, Fangxin Liu, Yifan Hao 0001, Xinkai Song, Li Jiang 0002, Xiaoyao Liang |
MICRO | 5 |
| 2024 | Cambricon-LLM: A Chiplet-Based Hybrid Architecture for On-Device Inference of 70B LLMabstractDeploying advanced large language models on edge devices, such as smartphones and robotics, is a growing trend that enhances user data privacy and network connectivity resilience while preserving intelligent capabilities. However, such a task exhibits single-batch computing with incredibly low arithmetic intensity, which poses the significant challenges of huge memory footprint and bandwidth demands on limited edge resources. To address these issues, we introduce Cambricon-LLM, a chiplet-based hybrid architecture with NPU and a dedicated NAND flash chip to enable efficient on-device inference of 70B LLMs. Such a hybrid architecture utilizes both the high computing capability of NPU and the data capacity of the NAND flash chip, with the proposed hardware-tiling strategy that minimizes the data movement overhead between NPU and NAND flash chip. Specifically, the NAND flash chip, enhanced by our innovative in-flash computing and on-die ECC techniques, excels at performing precise lightweight on-die processing. Simultaneously, the NPU collaborates with the flash chip for matrix operations and handles special function computations beyond the flash's on-die processing capabilities. Overall, Cambricon-LLM enables the on-device inference of 70B LLMs at a speed of 3.44 token/s, and 7B LLMs at a speed of 36.34 token/s, which is over 22× to 45× faster than existing flash-offloading technologies, showing the potentiality of deploying powerful LLMs in edge devices. Zhongkai Yu, Shengwen Liang, Tianyun Ma, Yunke Cai, Ziyuan Nan, Xinkai Song, Yifan Hao 0001, Jie Zhang 0048, Tian Zhi, Yongwei Zhao 0001, Zidong Du, Xing Hu 0001, Qi Guo 0001, Tianshi Chen 0002 |
MICRO | 7 |
| 2024 | DA-Ada: Learning Domain-Aware Adapter for Domain Adaptive Object DetectionabstractDomain adaptive object detection (DAOD) aims to generalize detectors trained on an annotated source domain to an unlabelled target domain.
As the visual-language models (VLMs) can provide essential general knowledge on unseen images, freezing the visual encoder and inserting a domain-agnostic adapter can learn domain-invariant knowledge for DAOD.
However, the domain-agnostic adapter is inevitably biased to the source domain.
It discards some beneficial knowledge discriminative on the unlabelled domain, \ie domain-specific knowledge of the target domain.
To solve the issue, we propose a novel Domain-Aware Adapter (DA-Ada) tailored for the DAOD task.
The key point is exploiting domain-specific knowledge between the essential general knowledge and domain-invariant knowledge.
DA-Ada consists of the Domain-Invariant Adapter (DIA) for learning domain-invariant knowledge and the Domain-Specific Adapter (DSA) for injecting the domain-specific knowledge from the information discarded by the visual encoder.
Comprehensive experiments over multiple DAOD tasks show that DA-Ada can efficiently infer a domain-aware visual encoder for boosting domain adaptive object detection.
Our code is available at https://github.com/Therock90421/DA-Ada. Haochen Li 0002, Rui Zhang 0040, Hantao Yao, Xin Zhang 0062, Yifan Hao 0001, Xinkai Song, Xiaqing Li, Yongwei Zhao 0001, Yunji Chen, Ling Li 0001 |
NeurIPS | 6 |
| 2024 | Environmental Condition Aware Super-Resolution Acceleration Framework in Server-Client HierarchiesabstractIn the current landscape, high-resolution (HR) videos have gained immense popularity, promising an elevated viewing experience. Recent research has demonstrated that the video super-resolution (SR) algorithm, empowered by deep neural networks (DNNs), can substantially enhance the quality of HR videos by processing low-resolution (LR) frames. However, the existing DNN models demand significant computational resources, posing challenges for the deployment of SR algorithms on client devices. While numerous accelerators have proposed solutions, their primary focus remains on client-side optimization. In contrast, our research recognizes that the HR video is originally stored in the cloud server and presents an untapped opportunity for achieving both high accuracy and performance improvements. Building on this insight, this article introduces an end-to-end video CODEC-assisted super-resolution (E 2 SR+) algorithm, which tightly integrates the cloud server with the client device to deliver a seamless and real-time video viewing experience. We propose the motion vector search algorithm executed in a cloud server, which can search the motion vectors and residuals for a part of the HR video frames and then pack them as add-ons. We also design an auto-encoder algorithm to down-sample the residuals to save the bitstream cost while guaranteeing the quality of the residuals. Lastly, we propose a reconstruction algorithm performed in the client to quickly reconstruct the corresponding HR frames using the add-ons to skip part of the DNN computations. To implement the E 2 SR+ algorithm, we design corresponding E 2 SR+ architecture in the client, which achieves significant speedup with minimal hardware overhead. Given that the environmental condition varies in the server–client hierarchies, we believe that simply applying E 2 SR+ to all frames is irrational. Accordingly, we offer an environmental condition–aware system to chase the best performance while adapting to the diverse environment. In the system, we design a linear programming (LP) model to simulate the environment and allocate frames to three existing mechanisms. Our experimental results demonstrate that the E 2 SR+ algorithm enhances the peak signal-to-noise ratio by 1.2, 2.5, and 2.3 compared with the state-of-the-art (SOTA) methods EDVR, BasicVSR, and BasicVSR++, respectively. In terms of performance, the E 2 SR+ architecture offers significant improvements over existing SOTA methods. For instance, while BasicVSR++ requires 98 ms on an NVIDIA V100 graphics processing unit (GPU) to generate a 1,280 × 720 HR frame, the E 2 SR+ architecture reduces the execution time to just 39 ms, highlighting the efficiency and effectiveness of our proposed method. Overall, the E 2 SR+ architecture respectively achieves 1.4×, 2.2×, 4.6×, and 442.0× performance improvement compared with ADAS, ISRAcc, the NVIDIA V100 GPU, and a central processing unit. Lastly, the proposed system showcases its superiority and surpasses all the existing mechanisms in terms of execution time when varying environmental conditions. Zhuoran Song, Zhongkai Yu, Xinkai Song, Yifan Hao 0001, Li Jiang 0002, Naifeng Jing, Xiaoyao Liang |
ACM Trans. Archit. Code Optim. | 3 |
| 2024 | REACT: Remainder Adaptive Compensation for Domain Adaptive Object DetectionabstractDomain adaptive object detection (DAOD) aims to infer a robust detector on the target domain with the labelled source datasets. Recent studies utilize a feature extractor shared on the source and target domains to capture the domain-invariant features and the task-relevant information with both feature-alignment constraint and source annotations. However, the feature extractor shared across domains discards partial task-relevant information of the target domain due to the domain gap and lack of target annotations, leading to compromised discrimination capabilities within target domain. To this end, we propose a novel REmainder Adaptive CompensaTion network (REACT) to adaptively compensate the extracted features with the remainder features for generating task-relevant features. The key insight is that the remainder features contain the discarded task-relevant information, so they can be adapted to compensate for the inadequate target features. Especially, REACT introduces an additional remainder branch to regain the remainder features, and then adaptively utilizes them to compensate for the discarded task-relevant information, improving discrimination on the target domain. Extensive experiments over multiple cross-domain adaptation tasks with three baselines demonstrate that our approach gains significant improvements and achieves superior performance compared with highly-optimized state-of-the-art methods. Haochen Li 0002, Rui Zhang 0040, Hantao Yao, Xin Zhang 0062, Yifan Hao 0001, Xinkai Song, Ling Li 0001 |
IEEE Trans. Image Process. | 6 |
| 2023 | BALTO: fast tensor program optimization with diversity-based active learning
Jun Bi, Xiaqing Li, Qi Guo 0001, Rui Zhang 0040, Yuanbo Wen 0001, Xing Hu 0001, Zidong Du, Xinkai Song, Yifan Hao 0001, Yunji Chen |
ICLR | 8 |
| 2023 | Cambricon-U: A Systolic Random Increment Memory Architecture for Unary ComputingabstractUnary computing, whose arithmetics require only one logic gate, has enabled efficient DNN processing, especially on strictly power-constrained devices. However, unary computing still confronts the power efficiency bottleneck for buffering unary bitstreams. The buffering of unary bitstreams requires accumulating bits into large bitwidth binary numbers. The large bitwidth binary number needs to activate all bits per cycle in case of carry propagation. As a result, the accumulation process accounts for 32%-70% of the power budget. Hongrui Guo, Yongwei Zhao 0001, Zhangmai Li, Yifan Hao 0001, Chang Liu 0021, Xinkai Song, Xiaqing Li, Zidong Du, Rui Zhang 0040, Qi Guo 0001, Tianshi Chen 0002, Zhiwei Xu 0002 |
MICRO | 6 |
| 2023 | Cambricon-R: A Fully Fused Accelerator for Real-Time Learning of Neural Scene RepresentationabstractNeural scene representation (NSR) initiates a new methodology of encoding a 3D scene with neural networks by learning from dozens of photos taken from different camera positions. NSR not only achieves significant improvement in the quality of novel view synthesis and 3D reconstruction but also reduces the camera cost from the expensive laser cameras to the cheap color cameras on the shelf. However, performing 3D scene encoding using NSR is far from real-time due to the extremely low hardware utilization (only utilization of hardware peak performance), which greatly limits its applications in real-time AR/VR interactions Xinkai Song, Yuanbo Wen 0001, Xing Hu 0001, Tianbo Liu 0006, Haoxuan Zhou, Husheng Han, Tian Zhi, Zidong Du, Wei Li 0008, Rui Zhang 0040, Chen Zhang 0001, Lin Gao 0004, Qi Guo 0001, Tianshi Chen 0002 |
MICRO | 1 |
| 2023 | Emergent Communication for Rules ReasoningabstractResearch on emergent communication between deep-learning-based agents has received extensive attention due to its inspiration for linguistics and artificial intelligence.
However, previous attempts have hovered around emerging communication under perception-oriented environmental settings,
that forces agents to describe low-level perceptual features intra image or symbol contexts.
In this work, inspired by the classic human reasoning test (namely Raven's Progressive Matrix), we propose the Reasoning Game, a cognition-oriented environment that encourages agents to reason and communicate high-level rules, rather than perceived low-level contexts.
Moreover, we propose 1) an unbiased dataset (namely rule-RAVEN) as a benchmark to avoid overfitting, 2) and a two-stage curriculum agent training method as a baseline for more stable convergence in the Reasoning Game,
where contexts and semantics are bilaterally drifting.
Experimental results show that, in the Reasoning Game, a semantically stable and compositional language emerges to solve reasoning problems.
The emerged language helps agents apply the extracted rules to the generalization of unseen context attributes, and to the transfer between different context attributes or even tasks. Yifan Hao 0001, Rui Zhang 0040, Enshuai Zhou, Zidong Du, Xishan Zhang, Xinkai Song, Yuanbo Wen 0001, Yongwei Zhao 0001, Xuehai Zhou, Jiaming Guo, Qi Yi, Shaohui Peng, Ruizhi Chen, Qi Guo 0001, Yunji Chen |
NeurIPS | 7 |
| 2023 | Learning Domain-Aware Detection Head with Prompt TuningabstractDomain adaptive object detection (DAOD) aims to generalize detectors trained on an annotated source domain to an unlabelled target domain.
However, existing methods focus on reducing the domain bias of the detection backbone by inferring a discriminative visual encoder, while ignoring the domain bias in the detection head.
Inspired by the high generalization of vision-language models (VLMs), applying a VLM as the robust detection backbone following a domain-aware detection head is a reasonable way to learn the discriminative detector for each domain, rather than reducing the domain bias in traditional methods.
To achieve the above issue, we thus propose a novel DAOD framework named Domain-Aware detection head with Prompt tuning (DA-Pro), which applies the learnable domain-adaptive prompt to generate the dynamic detection head for each domain.
Formally, the domain-adaptive prompt consists of the domain-invariant tokens, domain-specific tokens, and the domain-related textual description along with the class label.
Furthermore, two constraints between the source and target domains are applied to ensure that the domain-adaptive prompt can capture the domains-shared and domain-specific knowledge.
A prompt ensemble strategy is also proposed to reduce the effect of prompt disturbance.
Comprehensive experiments over multiple cross-domain adaptation tasks demonstrate that using the domain-adaptive prompt can produce an effectively domain-related detection head for boosting domain-adaptive object detection.
Our code is available at https://github.com/Therock90421/DA-Pro. Haochen Li 0002, Rui Zhang 0040, Hantao Yao, Xinkai Song, Yifan Hao 0001, Yongwei Zhao 0001, Ling Li 0001, Yunji Chen |
NeurIPS | 4 |
| 2023 | Chip design with machine learning: a survey from algorithm perspective
Wenkai He, Xiaqing Li, Xinkai Song, Yifan Hao 0001, Rui Zhang 0040, Zidong Du, Yunji Chen |
Sci. China Inf. Sci. | 3 |
| 2022 | Cambricon-G: A Polyvalent Energy-Efficient Accelerator for Dynamic Graph Neural NetworksabstractGraph neural networks (GNNs), which extend traditional neural networks for processing graph-structured data, have been widely used in many fields. The GNN computation mainly consists of theedge processingto generate messages by combining the edge/vertex features and thevertex processingto update the vertex features with aggregated messages. In addition to nontrivial vector operations in the edge processing, huge random accesses and neural network operations in the vertex processing, the graph topology of GNNs may also vary during the computation (i.e., dynamic GNNs). The above characteristics pose significant challenges on existing architectures. In this article, we propose a novel accelerator named CAMBRICON-G for efficient processing of both dynamic and static GNNs. The key of CAMBRICON-G is to abstract the irregular computation of a broad range of GNN variants to the process of regularly tiledadjacent cuboid(which extends the traditional adjacent matrix of graph by adding the dimension of vertex features). The intuition is that the adjacent cuboid facilitates exploitation of both data locality and parallelism by offeringmultidimensional multilevel tiling(including spatial and temporal tiling) opportunities. To perform themultidimensional spatial tiling, the CAMBRICON-G architecture mainly consists of the cuboid engine (CE) and hybrid on-chip memory. The CE has multiple vertex processing units (VPUs) working in a coordinated manner to efficiently process the sparse data and dynamically update the graph topology with dedicated instructions. The hybrid on-chip memory contains the topology-aware cache and multiple scratchpad memory to reduce off-chip memory access. To perform themultidimensional temporal tiling, an easy-to-use programming model is provided to flexibly explore different tiling options for large graphs. Experimental results show that compared against Nvidia P100 GPU, the performance and energy efficiency can be improved by$7.14\times $and$20.18\times $, respectively, on various GNNs, which validates both the versatility and energy efficiency of CAMBRICON-G. Xinkai Song, Tian Zhi, Zhe Fan, Wei Li 0008, Xing Hu 0001, Zidong Du, Qi Guo 0001, Yunji Chen |
IEEE Trans. Comput. Aided Des. Integr. Circuits Syst. | 1 |
| 2021 | Cambricon-Q: A Hybrid Architecture for Efficient TrainingabstractDeep neural network (DNN) training is notoriously time-consuming, and quantization is promising to improve the training efficiency with reduced bandwidth/storage requirements and computation costs. However, state-of-the-art quantized algorithms with negligible training accuracy loss, which require on-the-fly statistic-based quantization over a great amount of data (e.g., neurons and weights) and high-precision weight update, cannot be effectively deployed on existing DNN accelerators. To address this problem, we propose the first customized architecture for efficient quantized training with negligible accuracy loss, which is named as Cambricon-Q. Cambricon-Q features a hybrid architecture consisting of an ASIC acceleration core and a near-data-processing (NDP) engine. The acceleration core mainly targets at improving the efficiency of statistic-based quantization with specialized computing units for both statistical analysis (e.g., determining maximum) and data reformating, while the NDP engine avoids transferring the high-precision weights from the off-chip memory to the acceleration core. Experimental results show that on the evaluated benchmarks, Cambricon-Q improves the energy efficiency of DNN training by 6.41× and 1.62×, performance by 4.20× and 1.70× compared to GPU and TPU, respectively, with only ⩽ 0.4% accuracy degradation compared with full precision training. Yongwei Zhao 0001, Chang Liu 0021, Zidong Du, Qi Guo 0001, Xing Hu 0001, Yimin Zhuang, Xinkai Song, Wei Li 0008, Xishan Zhang, Ling Li 0001, Zhiwei Xu 0002, Tianshi Chen 0002 |
ISCA | 8 |