Zhenheng Tang

Xin He, Jiangchao Yao, Yuxin Wang et al.

One-shot neural architecture search (NAS) substantially improves the search efficiency by training one supernet to estimate the performance of every possible child architecture (i.e., subnet). However, the inconsistency of characteristics among subnets incurs serious interference in the optimization, resulting in poor performance ranking correlation of subnets. Subsequent explorations decompose supernet weights via a particular criterion, e.g., gradient matching, to reduce the interference; yet they suffer from huge computational cost and low space separability. In this work, we propose a lightweight and effective local intrinsic dimension (LID)-based method NAS-LID. NAS-LID evaluates the geometrical properties of architectures by calculating the low-cost LID features layer-by-layer, and the similarity characterized by LID enjoys better separability compared with gradients, which thus effectively reduces the interference among subnets. Extensive experiments on NASBench-201 indicate that NAS-LID achieves superior performance with better efficiency. Specifically, compared to the gradient-driven method, NAS-LID can save up to 86% of GPU memory overhead when searching on NASBench-201. We also demonstrate the effectiveness of NAS-LID on ProxylessNAS and OFA spaces. Source code: https://github.com/marsggbo/NAS-LID.

Zhenheng Tang, Yonggang Zhang, Shaohuai Shi et al.

In federated learning (FL), model performance typically suffers from client drift induced by data heterogeneity, and mainstream works focus on correcting client drift. We propose a different approach named virtual homogeneity learning (VHL) to directly "rectify" the data heterogeneity. In particular, VHL conducts FL with a virtual homogeneous dataset crafted to satisfy two conditions: containing no private information and being separable. The virtual dataset can be generated from pure noise shared across clients, aiming to calibrate the features from the heterogeneous clients. Theoretically, we prove that VHL can achieve provable generalization performance on the natural distribution. Empirically, we demonstrate that VHL endows FL with drastically improved convergence speed and generalization performance. VHL is the first attempt towards using a virtual dataset to address data heterogeneity, offering new and effective means to FL.

Zhenheng Tang, Xiaowen Chu, Ryan Yide Ran et al.

Federated Learning (FL) enables collaborations among clients for train machine learning models while protecting their data privacy. Existing FL simulation platforms that are designed from the perspectives of traditional distributed training, suffer from laborious code migration between simulation and production, low efficiency, low GPU utility, low scalability with high hardware requirements and difficulty of simulating stateful clients. In this work, we firstly demystify the challenges and bottlenecks of simulating FL, and design a new FL system named as FedML \texttt{Parrot}. It improves the training efficiency, remarkably relaxes the requirements on the hardware, and supports efficient large-scale FL experiments with stateful clients by: (1) sequential training clients on devices; (2) decomposing original aggregation into local and global aggregation on devices and server respectively; (3) scheduling tasks to mitigate straggler problems and enhance computing utility; (4) distributed client state manager to support various FL algorithms. Besides, built upon our generic APIs and communication interfaces, users can seamlessly transform the simulation into the real-world deployment without modifying codes. We evaluate \texttt{Parrot} through extensive experiments for training diverse models on various FL datasets to demonstrate that \texttt{Parrot} can achieve simulating over 1000 clients (stateful or stateless) with flexible GPU devices setting ($4 \sim 32$) and high GPU utility, 1.2 $\sim$ 4 times faster than FedScale, and 10 $\sim$ 100 times memory saving than FedML. And we verify that \texttt{Parrot} works well with homogeneous and heterogeneous devices in three different clusters. Two FL algorithms with stateful clients and four algorithms with stateless clients are simulated to verify the wide adaptability of \texttt{Parrot} to different algorithms.

Peijie Dong, Lujun Li, Yuedong Zhong et al.

In this paper, we present the first structural binarization method for LLM compression to less than 1-bit precision. Although LLMs have achieved remarkable performance, their memory-bound nature during the inference stage hinders the adoption of resource-constrained devices. Reducing weights to 1-bit precision through binarization substantially enhances computational efficiency. We observe that some weights in binarized LLMs can be randomly flipped without significant performance degradation, suggesting the potential for further compression. To exploit this, our STBLLM employs an N:M sparsity technique to achieve structural binarization of the weights. Specifically, we introduce a novel Standardized Importance (SI) metric, which considers weight magnitude and input feature norm to more accurately assess weight significance. Then, we propose a layer-wise approach, allowing different layers of the LLM to be sparsified with varying N:M ratios, thereby balancing compression and accuracy. Furthermore, we implement a fine-grained grouping strategy for less important weights, applying distinct quantization schemes to sparse, intermediate, and dense regions. Finally, we design a specialized CUDA kernel to support structural binarization. We conduct extensive experiments on LLaMA-1/2/3, OPT family, and Mistral to evaluate the effectiveness of STBLLM. The results demonstrate that our approach performs better than other compressed binarization LLM methods while significantly reducing memory requirements.

Zichen Tang, Junlin Huang, Rudan Yan et al.

Current data compression methods, such as sparsification in Federated Averaging (FedAvg), effectively enhance the communication efficiency of Federated Learning (FL). However, these methods encounter challenges such as the straggler problem and diminished model performance due to heterogeneous bandwidth and non-IID (Independently and Identically Distributed) data. To address these issues, we introduce a bandwidth-aware compression framework for FL, aimed at improving communication efficiency while mitigating the problems associated with non-IID data. First, our strategy dynamically adjusts compression ratios according to bandwidth, enabling clients to upload their models at a close pace, thus exploiting the otherwise wasted time to transmit more data. Second, we identify the non-overlapped pattern of retained parameters after compression, which results in diminished client update signals due to uniformly averaged weights. Based on this finding, we propose a parameter mask to adjust the client-averaging coefficients at the parameter level, thereby more closely approximating the original updates, and improving the training convergence under heterogeneous environments. Our evaluations reveal that our method significantly boosts model accuracy, with a maximum improvement of 13% over the uncompressed FedAvg. Moreover, it achieves a $3.37\times$ speedup in reaching the target accuracy compared to FedAvg with a Top-K compressor, demonstrating its effectiveness in accelerating convergence with compression. The integration of common compression techniques into our framework further establishes its potential as a versatile foundation for future cross-device, communication-efficient FL research, addressing critical challenges in FL and advancing the field of distributed machine learning.

Zichen Tang, Zirui Zhang, Qian Wang et al.

Current Large Language Models (LLMs) are gradually exploited in practically valuable agentic workflows such as Deep Research, E-commerce recommendation, and job recruitment. In these applications, LLMs need to select some optimal solutions from massive candidates, which we term as \textit{LLM-as-a-Recommender} paradigm. However, the reliability of using LLM agents for recommendations is underexplored. In this work, we introduce a \textbf{Bias} \textbf{Rec}ommendation \textbf{Bench}mark (\textbf{BiasRecBench}) to highlight the critical vulnerability of such agents to biases in high-value real-world tasks. The benchmark includes three practical domains: paper review, e-commerce, and job recruitment. We construct a \textsc{Bias Synthesis Pipeline with Calibrated Quality Margins} that 1) synthesizes evaluation data by controlling the quality gap between optimal and sub-optimal options to provide a calibrated testbed to elicit the vulnerability to biases; 2) injects contextual biases that are logical and suitable for option contexts. Extensive experiments on both SOTA (Gemini-{2.5,3}-pro, GPT-4o, DeepSeek-R1) and small-scale LLMs reveal that agents frequently succumb to injected biases despite having sufficient reasoning capabilities to identify the ground truth. These findings expose a significant reliability bottleneck in current agentic workflows, calling for specialized alignment strategies for LLM-as-a-Recommender. The complete code and evaluation datasets will be made publicly available shortly.

Yujin Tang, Peijie Dong, Zhenheng Tang et al.

Combining CNNs or ViTs, with RNNs for spatiotemporal forecasting, has yielded unparalleled results in predicting temporal and spatial dynamics. However, modeling extensive global information remains a formidable challenge; CNNs are limited by their narrow receptive fields, and ViTs struggle with the intensive computational demands of their attention mechanisms. The emergence of recent Mamba-based architectures has been met with enthusiasm for their exceptional long-sequence modeling capabilities, surpassing established vision models in efficiency and accuracy, which motivates us to develop an innovative architecture tailored for spatiotemporal forecasting. In this paper, we propose the VMRNN cell, a new recurrent unit that integrates the strengths of Vision Mamba blocks with LSTM. We construct a network centered on VMRNN cells to tackle spatiotemporal prediction tasks effectively. Our extensive evaluations show that our proposed approach secures competitive results on a variety of tasks while maintaining a smaller model size. Our code is available at https://github.com/yyyujintang/VMRNN-PyTorch.

Xinglin Pan, Shaohuai Shi, Wenxiang Lin et al.

The mixture-of-experts (MoE) architecture scales model size with sublinear computational increase but suffers from memory-intensive inference due to KV caches and sparse expert activation. Recent disaggregated expert parallelism (DEP) distributes attention and experts to dedicated GPU groups but lacks support for shared experts and efficient task scheduling, limiting performance. We propose FinDEP, a fine-grained task scheduling algorithm for DEP that maximizes task overlap to improve MoE inference throughput. FinDEP introduces three innovations: 1) partitioning computation/communication into smaller tasks for fine-grained pipelining, 2) formulating a scheduling optimization supporting variable granularity and ordering, and 3) developing an efficient solver for this large search space. Experiments on four GPU systems with DeepSeek-V2 and Qwen3-MoE show FinDEP improves throughput by up to 1.61x over prior methods, achieving up to 1.24x speedup on a 32-GPU system.

Peijie Dong, Ruibo Fan, Yuechen Tao et al.

Training large language models using 4-bit arithmetic enhances throughput and memory efficiency. Yet, the limited dynamic range of FP4 increases sensitivity to outliers. While NVFP4 mitigates quantization error via hierarchical microscaling, a persistent loss gap remains compared to BF16. This study conducts a longitudinal analysis of outlier dynamics across architecture during NVFP4 pretraining, focusing on where they localize, why they occur, and how they evolve temporally. We find that, compared with Softmax Attention (SA), Linear Attention (LA) reduces per-tensor heavy tails but still exhibits persistent block-level spikes under block quantization. Our analysis attributes outliers to specific architectural components: Softmax in SA, gating in LA, and SwiGLU in FFN, with "post-QK" operations exhibiting higher sensitivity to quantization. Notably, outliers evolve from transient spikes early in training to a small set of persistent hot channels (i.e., channels with persistently large magnitudes) in later stages. Based on these findings, we introduce Hot-Channel Patch (HCP), an online compensation mechanism that identifies hot channels and reinjects residuals using hardware-efficient kernels. We then develop CHON, an NVFP4 training recipe integrating HCP with post-QK operation protection. On GLA-1.3B model trained for 60B tokens, CHON reduces the loss gap to BF16 from 0.94% to 0.58% while maintaining downstream accuracy.

Qian Wang, Jiaying Wu, Zhenheng Tang et al.

We argue that advancing LLM-based human simulation requires addressing both LLM's inherent limitations and simulation framework design challenges. Recent studies have revealed significant gaps between LLM-based human simulations and real-world observations, highlighting these dual challenges. To address these gaps, we present a comprehensive analysis of LLM limitations and our design issues, proposing targeted solutions for both aspects. Furthermore, we explore future directions that address both challenges simultaneously, particularly in data collection, LLM generation, and evaluation. To support further research in this field, we provide a curated collection of LLM-based human simulation resources.\footnote{https://github.com/Persdre/llm-human-simulation}

Huacan Wang, Chaofa Yuan, Xialie Zhuang et al.

General-purpose technologies reshape economies less by improving individual tools than by enabling new ways to organize production and coordination. We believe AI agents are approaching a similar inflection point: as foundation models make broad task execution and tool use increasingly accessible, the binding constraint shifts from raw capability to how work is delegated, verified, and rewarded at scale. We introduce EpochX, a credits-native marketplace infrastructure for human-agent production networks. EpochX treats humans and agents as peer participants who can post tasks or claim them. Claimed tasks can be decomposed into subtasks and executed through an explicit delivery workflow with verification and acceptance. Crucially, EpochX is designed so that each completed transaction can produce reusable ecosystem assets, including skills, workflows, execution traces, and distilled experience. These assets are stored with explicit dependency structure, enabling retrieval, composition, and cumulative improvement over time. EpochX also introduces a native credit mechanism to make participation economically viable under real compute costs. Credits lock task bounties, budget delegation, settle rewards upon acceptance, and compensate creators when verified assets are reused. By formalizing the end-to-end transaction model together with its asset and incentive layers, EpochX reframes agentic AI as an organizational design problem: building infrastructures where verifiable work leaves persistent, reusable artifacts, and where value flows support durable human-agent collaboration.

Qi Li, Xiang Liu, Zhenheng Tang et al.

Model editing has become an increasingly popular alternative for efficiently updating knowledge within language models. Current methods mainly focus on reliability, generalization, and locality, with many methods excelling across these criteria. Some recent works disclose the pitfalls of these editing methods such as knowledge distortion or conflict. However, the general abilities of post-edited language models remain unexplored. In this paper, we perform a comprehensive evaluation on various editing methods and different language models, and have following findings. (1) Existing editing methods lead to inevitable performance deterioration on general benchmarks, indicating that existing editing methods maintain the general abilities of the model within only a few dozen edits. When the number of edits is slightly large, the intrinsic knowledge structure of the model is disrupted or even completely damaged. (2) Instruction-tuned models are more robust to editing, showing less performance drop on general knowledge after editing. (3) Language model with large scale is more resistant to editing compared to small model. (4) The safety of the edited model, is significantly weakened, even for those safety-aligned models. Our findings indicate that current editing methods are only suitable for small-scale knowledge updates within language models, which motivates further research on more practical and reliable editing methods. The details of code and reproduction can be found in https://github.com/lqinfdim/EditingEvaluation.

Zhenheng Tang, Xiang Liu, Qian Wang et al.

As Large Language Models (LLMs) become more powerful and autonomous, they increasingly face conflicts and dilemmas in many scenarios. We first summarize and taxonomize these diverse conflicts. Then, we model the LLM's preferences to make different choices as a priority graph, where instructions and values are nodes, and the edges represent context-specific priorities determined by the model's output distribution. This graph reveals that a unified stable LLM alignment is very challenging, because the graph is neither static nor necessarily consistent in different contexts. Besides, it also reveals a potential vulnerability: priority hacking, where adversaries can craft deceptive contexts to manipulate the graph and bypass safety alignments. To counter this, we propose a runtime verification mechanism, enabling LLMs to query external sources to ground their context and resist manipulation. While this approach enhances robustness, we also acknowledge that many ethical and value dilemmas are philosophically irreducible, posing a long-term, open challenge for the future of AI alignment.

Yehua Huang, Penglei Sun, Zebin Chen et al.

Academic peer review remains the cornerstone of scholarly validation, yet the field faces some challenges in data and methods. From the data perspective, existing research is hindered by the scarcity of large-scale, verified benchmarks and oversimplified evaluation metrics that fail to reflect real-world editorial workflows. To bridge this gap, we present OmniReview, a comprehensive dataset constructed by integrating multi-source academic platforms encompassing comprehensive scholarly profiles through the disambiguation pipeline, yielding 202, 756 verified review records. Based on this data, we introduce a three-tier hierarchical evaluaion framework to assess recommendations from recall to precise expert identification. From the method perspective, existing embedding-based approaches suffer from the information bottleneck of semantic compression and limited interpretability. To resolve these method limitations, we propose Profiling Scholars with Multi-gate Mixture-of-Experts (Pro-MMoE), a novel framework that synergizes Large Language Models (LLMs) with Multi-task Learning. Specifically, it utilizes LLM-generated semantic profiles to preserve fine-grained expertise nuances and interpretability, while employing a Task-Adaptive MMoE architecture to dynamically balance conflicting evaluation goals. Comprehensive experiments demonstrate that Pro-MMoE achieves state-of-the-art performance across six of seven metrics, establishing a new benchmark for realistic reviewer recommendation.

Xiang Liu, Zhenheng Tang, Peijie Dong et al.

Large Language Models (LLMs) require significant GPU memory when processing long texts, with the key value (KV) cache consuming up to 70\% of total memory during inference. Although existing compression methods reduce memory by evaluating the importance of individual tokens, they overlook critical semantic relationships between tokens, resulting in fragmented context and degraded performance. We introduce ChunkKV, which fundamentally reimagines KV cache compression by treating semantic chunks - rather than isolated tokens - as basic compression units. This approach preserves complete linguistic structures and contextual integrity, ensuring that essential meaning is retained even under aggressive compression. Our innovation includes a novel layer-wise index reuse technique that exploits the higher cross-layer similarity of preserved indices in ChunkKV, reducing computational overhead and improving throughput by 26.5\%. Comprehensive evaluations on challenging benchmarks: LongBench, Needle-In-A-HayStack, GSM8K, and JailbreakV demonstrate that ChunkKV outperforms state-of-the-art methods by up to 8.7\% in precision while maintaining the same compression ratio. These results confirm that semantic-aware compression significantly enhances both efficiency and performance for long-context LLM inference, providing a simple yet effective solution to the memory bottleneck problem. The code is available at \href{https://github.com/NVIDIA/kvpress}{link}.

Ziyi Ni, Huacan Wang, Shuo Zhang et al.

Beyond scratch coding, exploiting large-scale code repositories (e.g., GitHub) for practical tasks is vital in real-world software development, yet current benchmarks rarely evaluate code agents in such authentic, workflow-driven scenarios. To bridge this gap, we introduce GitTaskBench, a benchmark designed to systematically assess this capability via 54 realistic tasks across 7 modalities and 7 domains. Each task pairs a relevant repository with an automated, human-curated evaluation harness specifying practical success criteria. Beyond measuring execution and task success, we also propose the alpha-value metric to quantify the economic benefit of agent performance, which integrates task success rates, token cost, and average developer salaries. Experiments across three state-of-the-art agent frameworks with multiple advanced LLMs show that leveraging code repositories for complex task solving remains challenging: even the best-performing system, OpenHands+Claude 3.7, solves only 48.15% of tasks (recent progress has pushed the frontier further, with RepoMaster+Claude 3.5 achieving a new record of 62.96%). Error analysis attributes over half of failures to seemingly mundane yet critical steps like environment setup and dependency resolution, highlighting the need for more robust workflow management and increased timeout preparedness. By releasing GitTaskBench, we aim to drive progress and attention toward repository-aware code reasoning, execution, and deployment -- moving agents closer to solving complex, end-to-end real-world tasks. The benchmark and code are open-sourced at https://github.com/QuantaAlpha/GitTaskBench.

Xiaoyan Su, Peijie Dong, Zhenheng Tang et al.

Despite the rapid advancements in Vision-Language Models (VLMs), a critical gap remains in their ability to handle structured, controllable diagrammatic tasks essential for professional workflows. Existing methods predominantly rely on pixel-based synthesis, which operates in probabilistic pixel spaces and is inherently limited in editability and fidelity. Instead, we propose a new Diagram-as-Code paradigm with symbolic logic that leverages mxGraph Extensible Markup Language (XML) for precise diagram generation and editing. We present VCG-Bench, a unified benchmark for visual-centric \texttt{mxGraph} tasks. VCG-Bench comprises: (1) a taxonomized dataset of 1,449 diverse diagrams spanning 6 domains and 15 sub-domains, (2) a paradigm definition that integrates Generation (Vision-to-Code) and Editability (Code-to-Code), (3) a Tailored Evaluation Protocol employing multi-dimensional metrics such as \texttt{mxGraph} Execution Success Rate, Style Consistency Score (SCS), etc. Experimental results highlight the challenges faced by current State-of-the-Art (SOTA) VLMs in structured fidelity and instruction compliance, reflecting their vision and reasoning capabilities.

Xiang Liu, Shimiao Yuan, Zhenheng Tang et al.

LLM inference is still evaluated mainly as a model or software problem: accuracy, latency, throughput, and hardware utilization. This is incomplete. At deployment scale, the relevant output is a quality-conditioned token produced under joint constraints from effective compute, delivered data-center power, cooling capacity, PUE, and utilization. We argue that the ML community should treat inference as \emph{energy-to-token production}. We formalize this view with a dimensionally consistent Token Production Function in which token rate is bounded by both compute-per-token and energy-per-token ceilings. Listed API prices vary by over an order of magnitude across providers, but we use price dispersion only as directional motivation, not as causal evidence of marginal cost. The core physical question is instead: under fixed quality and service targets, when does the binding constraint move from theoretical peak compute toward delivered power, cooling, and operational efficiency? Under this framing, system optimizations -- latent KV-cache compression, sparse or heavily compressed attention, quantization, routing, and difficulty-adaptive reasoning -- are not merely local engineering tricks. They are energy-to-token levers because they reduce FLOPs/token, joules/token, memory traffic, or utilization losses under fixed $(q^{*},s^{*})$. We therefore call for inference papers and benchmarks to report Joules/token, active binding constraint, PUE-adjusted delivered power, and utilization-adjusted token output alongside accuracy and latency.

Peijie Dong, Zhenheng Tang, Xiang Liu et al.

Post-training compression reduces the computational and memory costs of large language models (LLMs), enabling resource-efficient deployment. However, existing compression benchmarks only focus on language modeling (e.g., perplexity) and natural language understanding tasks (e.g., GLUE accuracy), ignoring the agentic capabilities - workflow, tool use/function call, long-context understanding and real-world application. We introduce the Agent Compression Benchmark (ACBench), the first comprehensive benchmark for evaluating how compression impacts LLMs' agentic abilities. ACBench spans (1) 12 tasks across 4 capabilities (e.g., WorfBench for workflow generation, Needle-in-Haystack for long-context retrieval), (2) quantization (GPTQ, AWQ) and pruning (Wanda, SparseGPT), and (3) 15 models, including small (Gemma-2B), standard (Qwen2.5 7B-32B), and distilled reasoning LLMs (DeepSeek-R1-Distill). Our experiments reveal compression tradeoffs: 4-bit quantization preserves workflow generation and tool use (1%-3% drop) but degrades real-world application accuracy by 10%-15%. We introduce ERank, Top-k Ranking Correlation and Energy to systematize analysis. ACBench provides actionable insights for optimizing LLM compression in agentic scenarios. The code can be found in https://github.com/pprp/ACBench.

Peijie Dong, Lujun Li, Zhenheng Tang et al.

Despite the remarkable capabilities, Large Language Models (LLMs) face deployment challenges due to their extensive size. Pruning methods drop a subset of weights to accelerate, but many of them require retraining, which is prohibitively expensive and computationally demanding. Recently, post-training pruning approaches introduced novel metrics, enabling the pruning of LLMs without retraining. However, these metrics require the involvement of human experts and tedious trial and error. To efficiently identify superior pruning metrics, we develop an automatic framework for searching symbolic pruning metrics using genetic programming. In particular, we devise an elaborate search space encompassing the existing pruning metrics to discover the potential symbolic pruning metric. We propose an opposing operation simplification strategy to increase the diversity of the population. In this way, Pruner-Zero allows auto-generation of symbolic pruning metrics. Based on the searched results, we explore the correlation between pruning metrics and performance after pruning and summarize some principles. Extensive experiments on LLaMA and LLaMA-2 on language modeling and zero-shot tasks demonstrate that our Pruner-Zero obtains superior performance than SOTA post-training pruning methods. Code at: \url{https://github.com/pprp/Pruner-Zero}.

Chaoyang He, Alay Dilipbhai Shah, Zhenheng Tang et al.

Federated Learning (FL) is a distributed learning paradigm that can learn a global or personalized model from decentralized datasets on edge devices. However, in the computer vision domain, model performance in FL is far behind centralized training due to the lack of exploration in diverse tasks with a unified FL framework. FL has rarely been demonstrated effectively in advanced computer vision tasks such as object detection and image segmentation. To bridge the gap and facilitate the development of FL for computer vision tasks, in this work, we propose a federated learning library and benchmarking framework, named FedCV, to evaluate FL on the three most representative computer vision tasks: image classification, image segmentation, and object detection. We provide non-I.I.D. benchmarking datasets, models, and various reference FL algorithms. Our benchmark study suggests that there are multiple challenges that deserve future exploration: centralized training tricks may not be directly applied to FL; the non-I.I.D. dataset actually downgrades the model accuracy to some degree in different tasks; improving the system efficiency of federated training is challenging given the huge number of parameters and the per-client memory cost. We believe that such a library and benchmark, along with comparable evaluation settings, is necessary to make meaningful progress in FL on computer vision tasks. FedCV is publicly available: https://github.com/FedML-AI/FedCV.

Shuo Zhang, Chaofa Yuan, Ryan Guo et al.

While LLM-based agents have shown promise for deep research, most existing approaches rely on fixed workflows that struggle to adapt to real-world, open-ended queries. Recent work therefore explores self-evolution by allowing agents to rewrite their own code or prompts to improve problem-solving ability, but unconstrained optimization often triggers instability, hallucinations, and instruction drift. We propose EvoFSM, a structured self-evolving framework that achieves both adaptability and control by evolving an explicit Finite State Machine (FSM) instead of relying on free-form rewriting. EvoFSM decouples the optimization space into macroscopic Flow (state-transition logic) and microscopic Skill (state-specific behaviors), enabling targeted improvements under clear behavioral boundaries. Guided by a critic mechanism, EvoFSM refines the FSM through a small set of constrained operations, and further incorporates a self-evolving memory that distills successful trajectories as reusable priors and failure patterns as constraints for future queries. Extensive evaluations on five multi-hop QA benchmarks demonstrate the effectiveness of EvoFSM. In particular, EvoFSM reaches 58.0% accuracy on the DeepSearch benchmark. Additional results on interactive decision-making tasks further validate its generalization.

Zhenheng Tang, Yonggang Zhang, Shaohuai Shi et al.

Federated Learning (FL) models often experience client drift caused by heterogeneous data, where the distribution of data differs across clients. To address this issue, advanced research primarily focuses on manipulating the existing gradients to achieve more consistent client models. In this paper, we present an alternative perspective on client drift and aim to mitigate it by generating improved local models. First, we analyze the generalization contribution of local training and conclude that this generalization contribution is bounded by the conditional Wasserstein distance between the data distribution of different clients. Then, we propose FedImpro, to construct similar conditional distributions for local training. Specifically, FedImpro decouples the model into high-level and low-level components, and trains the high-level portion on reconstructed feature distributions. This approach enhances the generalization contribution and reduces the dissimilarity of gradients in FL. Experimental results show that FedImpro can help FL defend against data heterogeneity and enhance the generalization performance of the model.

Xin He, Shunkang Zhang, Yuxin Wang et al.

Sparse Mixture of Experts (MoE) models, while outperforming dense Large Language Models (LLMs) in terms of performance, face significant deployment challenges during inference due to their high memory demands. Existing offloading techniques, which involve swapping activated and idle experts between the GPU and CPU, often suffer from rigid expert caching mechanisms. These mechanisms fail to adapt to dynamic routing, leading to inefficient cache utilization, or incur prohibitive costs for prediction training. To tackle these inference-specific challenges, we introduce ExpertFlow, a comprehensive system specifically designed to enhance inference efficiency by accommodating flexible routing and enabling efficient expert scheduling between CPU and GPU. This reduces overhead and boosts system performance. Central to our approach is a predictive routing path-based offloading mechanism that utilizes a lightweight predictor to accurately forecast routing paths before computation begins. This proactive strategy allows for real-time error correction in expert caching, significantly increasing cache hit ratios and reducing the frequency of expert transfers, thereby minimizing I/O overhead. Additionally, we implement a dynamic token scheduling strategy that optimizes MoE inference by rearranging input tokens across different batches. This method not only reduces the number of activated experts per batch but also improves computational efficiency. Our extensive experiments demonstrate that ExpertFlow achieves up to 93.72\% GPU memory savings and enhances inference speed by 2 to 10 times compared to baseline methods, highlighting its effectiveness and utility as a robust solution for resource-constrained inference scenarios.

Zhenheng Tang, Yonggang Zhang, Peijie Dong et al.

One-shot Federated Learning (OFL) significantly reduces communication costs in FL by aggregating trained models only once. However, the performance of advanced OFL methods is far behind the normal FL. In this work, we provide a causal view to find that this performance drop of OFL methods comes from the isolation problem, which means that local isolatedly trained models in OFL may easily fit to spurious correlations due to the data heterogeneity. From the causal perspective, we observe that the spurious fitting can be alleviated by augmenting intermediate features from other clients. Built upon our observation, we propose a novel learning approach to endow OFL with superb performance and low communication and storage costs, termed as FuseFL. Specifically, FuseFL decomposes neural networks into several blocks, and progressively trains and fuses each block following a bottom-up manner for feature augmentation, introducing no additional communication costs. Comprehensive experiments demonstrate that FuseFL outperforms existing OFL and ensemble FL by a significant margin. We conduct comprehensive experiments to show that FuseFL supports high scalability of clients, heterogeneous model training, and low memory costs. Our work is the first attempt using causality to analyze and alleviate data heterogeneity of OFL.

Xiang Liu, Zhenheng Tang, Hong Chen et al.

This paper investigates an underexplored challenge in large language models (LLMs): the impact of KV cache compression methods on LLMs' fundamental capabilities. Although existing methods achieve impressive compression ratios on long-context benchmarks, their effects on core model capabilities remain understudied. We present a comprehensive benchmark KVFundaBench to systematically evaluate the effects of KV cache compression across diverse fundamental LLM capabilities, spanning world knowledge, commonsense reasoning, arithmetic reasoning, code generation, safety, and long-context understanding and generation.Our analysis reveals serval key findings: (1) \textit{Task-Dependent Degradation}; (2) \textit{Model-Type Robustness} (3) \textit{Prompt Length Vulnerability}; (4) \textit{Chunk-Level Superiority}; (5) \textit{Prompt-Gain Sensitivity}; (6) \textit{Long-Context Generation Sensitivity}. Based on our analysis of attention patterns and cross-task compression performance, we propose ShotKV, a novel compression approach that distinctly handles prefill and decoding phases while maintaining shot-level semantic coherence. Empirical results show that ShotKV achieves $9\%$-$18\%$ performance improvements on long-context generation tasks under aggressive compression ratios.

Xinglin Pan, Wenxiang Lin, Lin Zhang et al.

Recent large language models (LLMs) have tended to leverage sparsity to reduce computations, employing the sparsely activated mixture-of-experts (MoE) technique. MoE introduces four modules, including token routing, token communication, expert computation, and expert parallelism, that impact model quality and training efficiency. To enable versatile usage of MoE models, we introduce FSMoE, a flexible training system optimizing task scheduling with three novel techniques: 1) Unified abstraction and online profiling of MoE modules for task scheduling across various MoE implementations. 2) Co-scheduling intra-node and inter-node communications with computations to minimize communication overheads. 3) To support near-optimal task scheduling, we design an adaptive gradient partitioning method for gradient aggregation and a schedule to adaptively pipeline communications and computations. We conduct extensive experiments with configured MoE layers and real-world MoE models on two GPU clusters. Experimental results show that 1) our FSMoE supports four popular types of MoE routing functions and is more efficient than existing implementations (with up to a 1.42$\times$ speedup), and 2) FSMoE outperforms the state-of-the-art MoE training systems (DeepSpeed-MoE and Tutel) by 1.18$\times$-1.22$\times$ on 1458 MoE layers and 1.19$\times$-3.01$\times$ on real-world MoE models based on GPT-2 and Mixtral using a popular routing function.

Qian Wang, Zhanzhi Lou, Zhenheng Tang et al. · berkeley

Large Reasoning Models (LRMs) like DeepSeek-R1 and OpenAI-o1 have demonstrated remarkable reasoning capabilities, raising important questions about their biases in LLM-as-a-judge settings. We present a comprehensive benchmark comparing judging biases between LLMs and LRMs across both subjective preference-alignment datasets and objective fact-based datasets. Through investigation of bandwagon, authority, position, and distraction biases, we uncover four key findings: (1) despite their advanced reasoning capabilities, LRMs remain susceptible to the above biases; (2) LRMs demonstrate better robustness than LLMs specifically on fact-related datasets; (3) LRMs exhibit notable position bias, preferring options in later positions; and (4) we identify a novel "superficial reflection bias" where phrases mimicking reasoning (e.g., "wait, let me think...") significantly influence model judgments. To address these biases, we design and evaluate three mitigation strategies: specialized system prompts that reduce judging biases by up to 19\% in preference alignment datasets and 14\% in fact-related datasets, in-context learning that provides up to 27\% improvement on preference tasks but shows inconsistent results on factual tasks, and a self-reflection mechanism that reduces biases by up to 10\% in preference datasets and 16\% in fact-related datasets, with self-reflection proving particularly effective for LRMs. Our work provides crucial insights for developing more reliable LLM-as-a-Judge frameworks, especially as LRMs become increasingly deployed as automated judges.

Zhenheng Tang, Xueze Kang, Yiming Yin et al.

To alleviate hardware scarcity in training large deep neural networks (DNNs), particularly large language models (LLMs), we present FusionLLM, a decentralized training system designed and implemented for training DNNs using geo-distributed GPUs across different computing clusters or individual devices. Decentralized training faces significant challenges regarding system design and efficiency, including: 1) the need for remote automatic differentiation (RAD), 2) support for flexible model definitions and heterogeneous software, 3) heterogeneous hardware leading to low resource utilization or the straggler problem, and 4) slow network communication. To address these challenges, in the system design, we represent the model as a directed acyclic graph of operators (OP-DAG). Each node in the DAG represents the operator in the DNNs, while the edge represents the data dependency between operators. Based on this design, 1) users are allowed to customize any DNN without caring low-level operator implementation; 2) we enable the task scheduling with the more fine-grained sub-tasks, offering more optimization space; 3) a DAG runtime executor can implement RAD withour requiring the consistent low-level ML framework versions. To enhance system efficiency, we implement a workload estimator and design an OP-Fence scheduler to cluster devices with similar bandwidths together and partition the DAG to increase throughput. Additionally, we propose an AdaTopK compressor to adaptively compress intermediate activations and gradients at the slowest communication links. To evaluate the convergence and efficiency of our system and algorithms, we train ResNet-101 and GPT-2 on three real-world testbeds using 48 GPUs connected with 8 Mbps~10 Gbps networks. Experimental results demonstrate that our system and method can achieve 1.45 - 9.39x speedup compared to baseline methods while ensuring convergence.

Zhenheng Tang, Xiang Liu, Qian Wang et al.

Motivated by reducing the computational and storage costs of LLMs, model compression and KV cache compression have attracted much attention from researchers. However, current methods predominantly emphasize maintaining the performance of compressed LLMs, as measured by perplexity or simple accuracy on tasks of common sense knowledge QA and basic arithmetic reasoning. In this blog, we present a brief review of recent advancements in LLMs related to retrieval-augmented generation, multi-step reasoning, external tools, and computational expressivity, all of which substantially enhance LLM performance. Then, we propose a lottery LLM hypothesis suggesting that for a given LLM and task, there exists a smaller lottery LLM capable of producing the same performance as the original LLM with the assistance of multi-step reasoning and external tools. Based on the review of current progress in LLMs, we discuss and summarize the essential capabilities that the lottery LLM and KV cache compression must possess, which are currently overlooked in existing methods.

Kunfeng Lai, Zhenheng Tang, Xinglin Pan et al.

Model merging aggregates Large Language Models (LLMs) finetuned on different tasks into a stronger one. However, parameter conflicts between models leads to performance degradation in averaging. While model routing addresses this issue by selecting individual models during inference, it imposes excessive storage and compute costs, and fails to leverage the common knowledge from different models. In this work, we observe that different layers exhibit varying levels of parameter conflicts. Building on this insight, we average layers with minimal parameter conflicts and use a novel task-level expert routing for layers with significant conflicts. To further reduce storage costs, inspired by task arithmetic sparsity, we decouple multiple fine-tuned experts into a dense expert and several sparse experts. Considering the out-of-distribution samples, we select and merge appropriate experts based on the task uncertainty of the input data. We conduct extensive experiments on both LLaMA and Qwen with varying parameter scales, and evaluate on real-world reasoning tasks. Results demonstrate that our method consistently achieves significant performance improvements while requiring less system cost compared to existing methods.

Xiang Liu, Zhenheng Tang, Xia Li et al.

One-shot Federated Learning (OFL) is a distributed machine learning paradigm that constrains client-server communication to a single round, addressing privacy and communication overhead issues associated with multiple rounds of data exchange in traditional Federated Learning (FL). OFL demonstrates the practical potential for integration with future approaches that require collaborative training models, such as large language models (LLMs). However, current OFL methods face two major challenges: data heterogeneity and model heterogeneity, which result in subpar performance compared to conventional FL methods. Worse still, despite numerous studies addressing these limitations, a comprehensive summary is still lacking. To address these gaps, this paper presents a systematic analysis of the challenges faced by OFL and thoroughly reviews the current methods. We also offer an innovative categorization method and analyze the trade-offs of various techniques. Additionally, we discuss the most promising future directions and the technologies that should be integrated into the OFL field. This work aims to provide guidance and insights for future research.

Zixuan Yu, Zhenheng Tang, Tongliang Liu et al.

Despite the integration of search tools, Deep Search Agents often suffer from a misalignment between reasoning-driven queries and the underlying web indexing structures. Existing frameworks treat the search engine as a static utility, leading to queries that are either too coarse or too granular to retrieve precise evidence. We propose WeDas, a Web Content Distribution Aware framework that incorporates search-space structural characteristics into the agent's observation space. Central to our method is the Query-Result Alignment Score, a metric quantifying the compatibility between agent intent and retrieval outcomes. To overcome the intractability of indexing the dynamic web, we introduce a few-shot probing mechanism that iteratively estimates this score via limited query accesses, allowing the agent to dynamically recalibrate sub-goals based on the local content landscape. As a plug-and-play module, WeDas consistently improves sub-goal completion and accuracy across four benchmarks, effectively bridging the gap between high-level reasoning and low-level retrieval.

Qi Li, Junpan Wu, Xiang Liu et al.

The reasoning large language model (RLLM) has been proven competitive in solving complex reasoning tasks such as mathematics, coding, compared to general LLM. However, the serving performance and behavior of RLLM remains unexplored, which may undermine the deployment and utilization of RLLM in real-world scenario. To close this gap, in this paper, we conduct a comprehensive study of RLLM service. We first perform a pilot study on comparing the serving performance between RLLM and traditional LLM and reveal that there are several distinct differences regarding serving behavior: (1) significant memory usage and fluctuations; (2) straggler requests; (3) adaptive running time; (4) domain preference. Then we further investigate whether existing inference optimization techniques are valid for RLLM. Our main takeaways are that model quantization methods and speculative decoding can improve service system efficiency with small compromise to RLLM accuracy, while prefix caching, KV cache quantization may even degrade accuracy or serving performance for small RLLM. Lastly, we conduct evaluation under real world workload modeled by Gamma distribution to verify our findings. Empirical results of real world workload evaluation across different dataset are aligned with our main findings regarding RLLM serving. We hope our work can provide the research community and industry with insights to advance RLLM inference serving.

Xin He, Junxi Shen, Zhenheng Tang et al.

Model merging via Mixture-of-Experts (MoE) has emerged as a scalable solution for consolidating multiple task-specific models into a unified sparse architecture, where each expert is derived from a model fine-tuned on a distinct task. While effective for multi-task integration, this paradigm introduces a critical yet underexplored challenge: how to attribute and protect the intellectual property (IP) of individual experts after merging. We propose RouteMark, a framework for IP protection in merged MoE models through the design of expert routing fingerprints. Our key insight is that task-specific experts exhibit stable and distinctive routing behaviors under probing inputs. To capture these patterns, we construct expert-level fingerprints using two complementary statistics: the Routing Score Fingerprint (RSF), quantifying the intensity of expert activation, and the Routing Preference Fingerprint (RPF), characterizing the input distribution that preferentially activates each expert. These fingerprints are reproducible, task-discriminative, and lightweight to construct. For attribution and tampering detection, we introduce a similarity-based matching algorithm that compares expert fingerprints between a suspect and a reference (victim) model. Extensive experiments across diverse tasks and CLIP-based MoE architectures show that RouteMark consistently yields high similarity for reused experts and clear separation from unrelated ones. Moreover, it remains robust against both structural tampering (expert replacement, addition, deletion) and parametric tampering (fine-tuning, pruning, permutation), outperforming weight- and activation-based baseliness. Our work lays the foundation for RouteMark as a practical and broadly applicable framework for IP verification in MoE-based model merging.

Zeyu Li, Chuanfu Xiao, Yang Wang et al.

Quantization has emerged as an effective and lightweight solution to reduce the memory footprint of the KV cache in Large Language Models. Nevertheless, minimizing the accuracy degradation caused by ultra-low-bit KV cache quantization remains a significant challenge. While scalar quantization is constrained by 1-bit bound, vector quantization exploits intra-vector correlations and enables sub-bit regimes, making it more suitable for ultra-low-bit quantization. To further mitigate quantization-induced degradation, we reveal that the degradation is highly uneven across tokens in attention quality. To investigate this unevenness, we introduce anchor score to measure each token's sensitivity to quantization. Our analysis and experiments show that preserving a small subset (1\%) of tokens with the highest Anchor Score significantly mitigates accuracy loss under aggressive quantization. We propose AnTKV, a dual-stage framework that leverages anchor token-aware vector quantization to compress the KV cache. It combines offline token-aware centroids learning and online anchor token selection to balance compression and accuracy. To enable efficient deployment, we design an online anchor token selection kernel compatible with FlashAttention. It allows LLaMA3-8B to scale to 840K tokens on a single 80GB A100, while delivering up to $3.5\times$ higher decoding throughput over the FP16 baseline. Experiments demonstrate that AnTKV matches or surpasses prior methods at 4-bit, and significantly reduce perplexity under ultra-low-bit quantization, achieving 6.32 at 1-bit on Mistral-7B, compared to 7.25 for CQ and 15.36 for KVQuant.

Junlin Huang, Wenyi Fang, Zhenheng Tang et al.

Training Large Language Models (LLMs) at ultra-low precision is critically impeded by instability rooted in the conflict between discrete quantization constraints and the intrinsic heavy-tailed spectral nature of linguistic data. By formalizing the connection between Zipfian statistics and random matrix theory, we prove that the power-law decay in the singular value spectra of embeddings is a fundamental requisite for semantic encoding. We derive theoretical bounds showing that uniform quantization introduces a noise floor that disproportionately truncates this spectral tail, which induces spectral flattening and a strictly provable increase in the stable rank of representations. Empirical validation across diverse architectures including GPT-2 and TinyLlama corroborates that this geometric degradation precipitates representational collapse. This work not only quantifies the spectral sensitivity of LLMs but also establishes spectral fidelity as a necessary condition for stable low-bit optimization.

Yifu Guo, Zishan Xu, Zhiyuan Yao et al.

Existing multimodal reasoning models and frameworks suffer from fundamental architectural limitations: most lack the human-like ability to autonomously explore diverse reasoning pathways-whether in direct inference, tool-driven visual exploration, programmatic visual manipulation, or intrinsic visual imagination. Consequently, they struggle to adapt to dynamically changing capability requirements in real-world tasks. Meanwhile, humans exhibit a complementary set of thinking abilities when addressing such tasks, whereas existing methods typically cover only a subset of these dimensions. Inspired by this, we propose Octopus: Agentic Multimodal Reasoning with Six-Capability Orchestration, a new paradigm for multimodal agentic reasoning. We define six core capabilities essential for multimodal reasoning and organize a comprehensive evaluation benchmark, Octopus-Bench, accordingly. Octopus is capable of autonomously exploring during reasoning and dynamically selecting the most appropriate capability based on the current state. Experimental results show that Octopus achieves the best performance on the vast majority of tasks in Octopus-Bench, highlighting the crucial role of capability coordination in agentic multimodal reasoning.

Zhenheng Tang, Yuxin Wang, Xin He et al.

The rapid growth of memory and computation requirements of large language models (LLMs) has outpaced the development of hardware, hindering people who lack large-scale high-end GPUs from training or deploying LLMs. However, consumer-level GPUs, which constitute a larger market share, are typically overlooked in LLM due to their weaker computing performance, smaller storage capacity, and lower communication bandwidth. Additionally, users may have privacy concerns when interacting with remote LLMs. In this paper, we envision a decentralized system unlocking the potential vast untapped consumer-level GPUs in pre-training, inference and fine-tuning of LLMs with privacy protection. However, this system faces critical challenges, including limited CPU and GPU memory, low network bandwidth, the variability of peer and device heterogeneity. To address these challenges, our system design incorporates: 1) a broker with backup pool to implement dynamic join and quit of computing providers; 2) task scheduling with hardware performance to improve system efficiency; 3) abstracting ML procedures into directed acyclic graphs (DAGs) to achieve model and task universality; 4) abstracting intermediate represention and execution planes to ensure compatibility of various devices and deep learning (DL) frameworks. Our performance analysis demonstrates that 50 RTX 3080 GPUs can achieve throughputs comparable to those of 4 H100 GPUs, which are significantly more expensive.

Shaohuai Shi, Zhenheng Tang, Xiaowen Chu et al.

Nowadays, large and complex deep learning (DL) models are increasingly trained in a distributed manner across multiple worker machines, in which extensive communications between workers pose serious scaling problems. In this article, we present a quantitative survey of communication optimization techniques for data parallel distributed DL. We first identify the major communication challenges and classify the existing solutions into three levels, namely the learning algorithm, the system architecture, and the network infrastructure. We present the state-of-the-art communication optimization techniques and conduct a comparative study of seven common lossless distributed DL methods on a 32-GPU cluster with 100Gbps InfiniBand (IB). We show that (1) the DL models with low model intensity (such as BERT and BERT-Large) are difficult to scale out even with the best available lossless algorithm over 100Gbps IB; (2) the system architecture and scheduling algorithms have a critical impact on the scaling property. We conclude the article with discussions on the open issues for further investigations.

Zhenheng Tang, Shaohuai Shi, Wei Wang et al.

Distributed deep learning (DL) has become prevalent in recent years to reduce training time by leveraging multiple computing devices (e.g., GPUs/TPUs) due to larger models and datasets. However, system scalability is limited by communication becoming the performance bottleneck. Addressing this communication issue has become a prominent research topic. In this paper, we provide a comprehensive survey of the communication-efficient distributed training algorithms, focusing on both system-level and algorithmic-level optimizations. We first propose a taxonomy of data-parallel distributed training algorithms that incorporates four primary dimensions: communication synchronization, system architectures, compression techniques, and parallelism of communication and computing tasks. We then investigate state-of-the-art studies that address problems in these four dimensions. We also compare the convergence rates of different algorithms to understand their convergence speed. Additionally, we conduct extensive experiments to empirically compare the convergence performance of various mainstream distributed training algorithms. Based on our system-level communication cost analysis, theoretical and experimental convergence speed comparison, we provide readers with an understanding of which algorithms are more efficient under specific distributed environments. Our research also extrapolates potential directions for further optimizations.

Zhenheng Tang, Shaohuai Shi, Xiaowen Chu

Distributed learning techniques such as federated learning have enabled multiple workers to train machine learning models together to reduce the overall training time. However, current distributed training algorithms (centralized or decentralized) suffer from the communication bottleneck on multiple low-bandwidth workers (also on the server under the centralized architecture). Although decentralized algorithms generally have lower communication complexity than the centralized counterpart, they still suffer from the communication bottleneck for workers with low network bandwidth. To deal with the communication problem while being able to preserve the convergence performance, we introduce a novel decentralized training algorithm with the following key features: 1) It does not require a parameter server to maintain the model during training, which avoids the communication pressure on any single peer. 2) Each worker only needs to communicate with a single peer at each communication round with a highly compressed model, which can significantly reduce the communication traffic on the worker. We theoretically prove that our sparsification algorithm still preserves convergence properties. 3) Each worker dynamically selects its peer at different communication rounds to better utilize the bandwidth resources. We conduct experiments with convolutional neural networks on 32 workers to verify the effectiveness of our proposed algorithm compared to seven existing methods. Experimental results show that our algorithm significantly reduces the communication traffic and generally select relatively high bandwidth peers.

Shaohuai Shi, Zhenheng Tang, Qiang Wang et al.

To reduce the long training time of large deep neural network (DNN) models, distributed synchronous stochastic gradient descent (S-SGD) is commonly used on a cluster of workers. However, the speedup brought by multiple workers is limited by the communication overhead. Two approaches, namely pipelining and gradient sparsification, have been separately proposed to alleviate the impact of communication overheads. Yet, the gradient sparsification methods can only initiate the communication after the backpropagation, and hence miss the pipelining opportunity. In this paper, we propose a new distributed optimization method named LAGS-SGD, which combines S-SGD with a novel layer-wise adaptive gradient sparsification (LAGS) scheme. In LAGS-SGD, every worker selects a small set of "significant" gradients from each layer independently whose size can be adaptive to the communication-to-computation ratio of that layer. The layer-wise nature of LAGS-SGD opens the opportunity of overlapping communications with computations, while the adaptive nature of LAGS-SGD makes it flexible to control the communication time. We prove that LAGS-SGD has convergence guarantees and it has the same order of convergence rate as vanilla S-SGD under a weak analytical assumption. Extensive experiments are conducted to verify the analytical assumption and the convergence performance of LAGS-SGD. Experimental results on a 16-GPU cluster show that LAGS-SGD outperforms the original S-SGD and existing sparsified S-SGD without losing obvious model accuracy.

Xin He, Shihao Wang, Shaohuai Shi et al.

Skin disease is one of the most common types of human diseases, which may happen to everyone regardless of age, gender or race. Due to the high visual diversity, human diagnosis highly relies on personal experience; and there is a serious shortage of experienced dermatologists in many countries. To alleviate this problem, computer-aided diagnosis with state-of-the-art (SOTA) machine learning techniques would be a promising solution. In this paper, we aim at understanding the performance of convolutional neural network (CNN) based approaches. We first build two versions of skin disease datasets from Internet images: (a) Skin-10, which contains 10 common classes of skin disease with a total of 10,218 images; (b) Skin-100, which is a larger dataset that consists of 19,807 images of 100 skin disease classes. Based on these datasets, we benchmark several SOTA CNN models and show that the accuracy of skin-100 is much lower than the accuracy of skin-10. We then implement an ensemble method based on several CNN models and achieve the best accuracy of 79.01\% for Skin-10 and 53.54\% for Skin-100. We also present an object detection based approach by introducing bounding boxes into the Skin-10 dataset. Our results show that object detection can help improve the accuracy of some skin disease classes.

Yuxin Wang, Qiang Wang, Shaohuai Shi et al.

Deep learning has become widely used in complex AI applications. Yet, training a deep neural network (DNNs) model requires a considerable amount of calculations, long running time, and much energy. Nowadays, many-core AI accelerators (e.g., GPUs and TPUs) are designed to improve the performance of AI training. However, processors from different vendors perform dissimilarly in terms of performance and energy consumption. To investigate the differences among several popular off-the-shelf processors (i.e., Intel CPU, NVIDIA GPU, AMD GPU, and Google TPU) in training DNNs, we carry out a comprehensive empirical study on the performance and energy efficiency of these processors by benchmarking a representative set of deep learning workloads, including computation-intensive operations, classical convolutional neural networks (CNNs), recurrent neural networks (LSTM), Deep Speech 2, and Transformer. Different from the existing end-to-end benchmarks which only present the training time, We try to investigate the impact of hardware, vendor's software library, and deep learning framework on the performance and energy consumption of AI training. Our evaluation methods and results not only provide an informative guide for end-users to select proper AI accelerators, but also expose some opportunities for the hardware vendors to improve their software library.

Shaohuai Shi, Qiang Wang, Kaiyong Zhao et al.

Distributed synchronous stochastic gradient descent (S-SGD) has been widely used in training large-scale deep neural networks (DNNs), but it typically requires very high communication bandwidth between computational workers (e.g., GPUs) to exchange gradients iteratively. Recently, Top-$k$ sparsification techniques have been proposed to reduce the volume of data to be exchanged among workers. Top-$k$ sparsification can zero-out a significant portion of gradients without impacting the model convergence. However, the sparse gradients should be transferred with their irregular indices, which makes the sparse gradients aggregation difficult. Current methods that use AllGather to accumulate the sparse gradients have a communication complexity of $O(kP)$, where $P$ is the number of workers, which is inefficient on low bandwidth networks with a large number of workers. We observe that not all top-$k$ gradients from $P$ workers are needed for the model update, and therefore we propose a novel global Top-$k$ (gTop-$k$) sparsification mechanism to address the problem. Specifically, we choose global top-$k$ largest absolute values of gradients from $P$ workers, instead of accumulating all local top-$k$ gradients to update the model in each iteration. The gradient aggregation method based on gTop-$k$ sparsification reduces the communication complexity from $O(kP)$ to $O(k\log P)$. Through extensive experiments on different DNNs, we verify that gTop-$k$ S-SGD has nearly consistent convergence performance with S-SGD, and it has only slight degradations on generalization performance. In terms of scaling efficiency, we evaluate gTop-$k$ on a cluster with 32 GPU machines which are interconnected with 1 Gbps Ethernet. The experimental results show that our method achieves $2.7-12\times$ higher scaling efficiency than S-SGD and $1.1-1.7\times$ improvement than the existing Top-$k$ S-SGD.