arXiv subjects

Jiangsu Du

Publications and source records attributed to Jiangsu Du.

18 recordsLinked to original sources

arXiv2026

Homebot: A Personal AI Agent for Conversational Home Assistance and Automation

\texttt{Homebot} is a locally deployable AI agent for conversational household assistance and automation. It accepts voice and instant-messaging requests through a shared runtime that combines language-model responses with registered tools and task-specific skills. The design separates common request processing from session ownership: messaging history remains scoped to a channel and chat, whereas voice interaction is bounded by wake-word activation. For hands-free use, \texttt{Homebot} combines local wake-word detection, streaming speech recognition and synthesis, and an explicit dialogue-state protocol for ending, following up, or continuing a conversation. Clear channel, tool, and skill contracts support practical customization for household use.

cs.AI↗

arXiv2026

FVAttn: Adaptive Sparse Attention with Runtime Load Balancing for Video Generation

Video Diffusion Transformers process long spatio-temporal sequences, making self-attention the main bottleneck in high-resolution video generation. Training-free sparse attention reduces this cost, but adaptive Top-$p$ routing creates uneven per-head workloads under multi-GPU sequence parallelism. The resulting workload heterogeneity turns sparse attention into a rank-level straggler problem. We present \method{}, a training-free sparse-attention system that improves the distributed execution efficiency of adaptive sparse attention under multi-GPU sequence parallelism. \method{} uses Top-$p$ routing, a Top-$k$ safety floor, and video-aware block organization as the sparse-routing frontend, then repairs the materialized mask at runtime. Runtime Load Balancing migrates a small number of heavy heads via P2P communication to shorten the current critical path. Slack-Aware Sparse Augmentation fills residual non-critical-rank slack with additional high-value blocks, while overlap hides scheduling and migration overhead behind existing computation. On step-distilled Wan2.2 I2V, \method{} reduces average load imbalance from 1.34 to 1.08 and delivers a $4.41\times$ attention speedup over FlashAttention, while achieving a $2.02$--$2.11\times$ DiT inference speedup with competitive video quality.

cs.CV↗

arXiv2026

Communication-Aware Placement and Pruning for Efficient Mixture-of-Experts Inference

As MoE models scale to hundreds of experts, placement and pruning decisions increasingly dictate communication volume, affecting the performance of distributed inference across GPUs and nodes. We propose CAP (Communication-Aware Assignment and Pruning), a framework that considers computation, communication and accuracy together for efficient MoE inference through expert placement and pruning. It consists of three components: (1) Co-activation driven expert placement, which groups frequently co-activated experts to reduce inter-device and inter-node communication; (2) Communicationcomputation trade-off adjustment, which generates placements with different computational load and communication volume; and (3) Communication-aware expert pruning, which selectively removes routing destinations to reduce communication with limited accuracy degradation. By combining these components, CAP selects an efficient operating strategy for different hardware configurations. Across our single-node and multi-node experiments, it achieves 1.23x - 1.86 x throughput improvement over DeepSeek EPLB and sequential placement in vLLM, and preserves better model accuracy at the same target speedup under lossy acceleration.

cs.DC↗

arXiv2026

Chorus II: Cross-Request Sparsity Reuse for Efficient Image-to-Video Generation

Serving diffusion models for image-to-video generation is computationally expensive, posing significant challenges for large-scale deployment. Real I2V workloads often contain similar requests, such as repeated effect templates, related subjects, and recurring shot layouts. Existing cross-request acceleration methods mainly exploit this redundancy through feature reuse. We observe that similar I2V requests also share highly consistent sparse attention patterns, enabling historical sparse masks to serve as request-conditioned priors with almost no online mask-prediction overhead. We propose a cross-request reuse framework centered on \textbf{sparsity reuse}, with \textbf{feature reuse} as an optional extension safeguarded by a lightweight \textbf{guidance enhancement}. Our sparsity reuse is implemented as shared sparse mask reuse, which reuses high-quality sparse masks from similar historical requests to avoid per-request online mask prediction. Optional feature reuse applies downsampled computation to highly redundant spatiotemporal regions, mitigating boundary artifacts while preserving efficiency gains. Guidance enhancement reinforces image/text conditioning after reuse, mitigating semantic drift and condition-adherence issues. Experiments show that default sparsity reuse configuration preserves generation quality with a \textbf{2.16$\times$} speedup.

cs.CV↗

arXiv2026

AlignedServe: Orchestrating Prefix-aware Batching to Build a High-throughput and Computing-efficient LLM Serving System

High-throughput inference serving is essential for applications built on large language models (LLMs). Existing serving frameworks reduce request-level and batch-level bubbles through batching and scheduling, but often overlook bubbles within each decode iteration. Tokens generated in the same iteration may incur different costs because they depend on KV caches of different lengths; tokens with long KV caches can become bottlenecks and delay the next iteration. We propose AlignedServe, an LLM serving framework built around prefix-aware batching. It groups requests with similar KV-cache lengths into the same batch to reduce iteration-level bubbles. To support this policy efficiently, AlignedServe uses large CPU memory to maintain sufficient in-flight requests for batching and applies a batch-level scheduling policy to reduce batch-level bubbles. It also introduces a GPU-Prefetch-For-GPU architecture, where one GPU prefetches KV cache for another to reduce CPU-to-GPU transfer latency. Experiments on synthetic and application workloads show that AlignedServe improves decoding throughput by up to 1.98 times and reduces latency by up to 7.4 times over state-of-the-art systems.

cs.DC↗

arXiv2026

PipeMax: Enhancing Offline LLM Inference on Commodity GPU Servers

Offline LLM inference seeks to maximize request processing under fixed budgets, making commodity GPU servers a promising choice. However, prior work typically considers offloading and parallelism in isolation, resulting in suboptimal performance. In this paper, we propose PipeMax, a high-throughput LLM inference system that integrates pipeline parallelism with offloading to overcome interconnect and memory constraints on GPU servers. Particularly, pipeline parallelism naturally incurs low communication overhead and keeps only one batch active on each GPU at a time, which enables offloading the KV cache of inactive batches. By coordinating computation with offloading data movement, PipeMax effectively expands GPU memory capacity and sustains large-batch execution. Experiments show that PipeMax achieves up to 2.51x higher throughput than vLLM, and up to 1.42x and 1.38x higher throughput than state-of-the-art high-throughput LLM systems, respectively, on an 8-GPU node.

cs.DC↗

arXiv2026

Beyond Few-Step Inference: Accelerating Video Diffusion Transformer Model Serving with Inter-Request Caching Reuse

Video Diffusion Transformer (DiT) models are a dominant approach for high-quality video generation but suffer from high inference cost due to iterative denoising. Existing caching approaches primarily exploit similarity within the diffusion process of a single request to skip redundant denoising steps. In this paper, we introduce Chorus, a caching approach that leverages similarity across requests to accelerate video diffusion model serving. Chorus achieves up to 45\% speedup on industrial 4-step distilled models, where prior intra-request caching approaches are ineffective. Particularly, Chorus employs a three-stage caching strategy along the denoising process. Stage 1 performs full reuse of latent features from similar requests. Stage 2 exploits inter-request caching in specific latent regions during intermediate denoising steps. This stage is combined with Token-Guided Attention Amplification to improve semantic alignment between the generated video and the conditional prompts, thereby extending the applicability of full reuse to later denoising steps.

cs.CV↗

arXiv2026

Adaptive Hybrid Caching for Efficient Text-to-Video Diffusion Model Acceleration

Efficient video generation models are increasingly vital for multimedia synthetic content generation. Leveraging the Transformer architecture and the diffusion process, video DiT models have emerged as a dominant approach for high-quality video generation. However, their multi-step iterative denoising process incurs high computational cost and inference latency. Caching, a widely adopted optimization method in DiT models, leverages the redundancy in the diffusion process to skip computations in different granularities (e.g., step, cfg, block). Nevertheless, existing caching methods are limited to single-granularity strategies, struggling to balance generation quality and inference speed in a flexible manner. In this work, we propose MixCache, a training-free caching-based framework for efficient video DiT inference. It first distinguishes the interference and boundary between different caching strategies, and then introduces a context-aware cache triggering strategy to determine when caching should be enabled, along with an adaptive hybrid cache decision strategy for dynamically selecting the optimal caching granularity. Extensive experiments on diverse models demonstrate that, MixCache can significantly accelerate video generation (e.g., 1.94$\times$ speedup on Wan 14B, 1.97$\times$ speedup on HunyuanVideo) while delivering both superior generation quality and inference efficiency compared to baseline methods.

cs.GR↗

arXiv2026

Resource-Efficient Personal Large Language Models Fine-Tuning with Collaborative Edge Computing

Large language models (LLMs) have unlocked a plethora of powerful applications at the network edge, such as intelligent personal assistants. Data privacy and security concerns have prompted a shift towards edge-based fine-tuning of personal LLMs, away from cloud reliance. However, this raises issues of computational intensity and resource scarcity, hindering training efficiency and feasibility. While current studies investigate parameter-efficient fine-tuning (PEFT) techniques to mitigate resource constraints, our analysis indicates that these techniques are not sufficiently resource-efficient for edge devices. To tackle these challenges, we propose Pluto and Charon (PAC), a time and memory efficient collaborative edge AI framework for personal LLMs fine-tuning. PAC breaks the resource wall of personal LLMs fine-tuning with a sophisticated algorithm-system co-design. (1) Algorithmically, PAC implements a personal LLMs fine-tuning technique that is efficient in terms of parameters, time, and memory. It utilizes Parallel Adapters to circumvent the need for a full backward pass through the LLM backbone. Additionally, an activation cache mechanism further streamlining the process by negating the necessity for repeated forward passes across multiple epochs. (2) Systematically, PAC leverages edge devices in close proximity, pooling them as a collective resource for in-situ personal LLMs fine-tuning, utilizing a hybrid data and pipeline parallelism to orchestrate distributed training. The use of the activation cache eliminates the need for forward pass through the LLM backbone,enabling exclusive fine-tuning of the Parallel Adapters using data parallelism. Extensive evaluation based on prototype implementation demonstrates that PAC remarkably outperforms state-of-the-art approaches, achieving up to 8.64x end-to-end speedup and up to 88.16% reduction in memory footprint.

cs.DC↗

arXiv2025

TD-Pipe: Temporally-Disaggregated Pipeline Parallelism Architecture for High-Throughput LLM Inference

As the model size continuously increases, pipeline parallelism shows great promise in throughput-oriented LLM inference due to its low demand on communications. However, imbalanced pipeline workloads and complex data dependencies in the prefill and decode phases result in massive pipeline bubbles and further severe performance reduction. To better exploit the pipeline parallelism for high-throughput LLM inference, we propose TD-Pipe, with the key idea lies in the temporally-disaggregated pipeline parallelism architecture. Specifically, this architecture disaggregates the prefill and decode phases in the temporal dimension, so as to eliminate pipeline bubbles caused by the phase switching. TD-Pipe identifies potential issues of exploiting the novel architecture and provides solutions. First, a hierarchy-controller structure is used to better coordinate devices in pipeline parallelism by decoupling the scheduling from execution. Second, the AI-based greedy prefill approach aggressively performs more prefills by predicting the output length and simulating the memory usage. Third, the inter-batch work stealing approach dynamically balances decode phase workloads between different batches to reduce bubbles. Forth, the spatial-temporal intensity comparison approach determines the optimal switch from decode to prefill by comparing the performance drop from reduced computational intensity with that from phase switching bubbles. Extensive experiments show that TD-Pipe effectively increases the throughput of LLM inference by up to 1.91x over the existing tensor parallel approach and 2.73x over the existing pipeline parallel approach on GPU nodes with only PCIe interconnection.

cs.DC↗

arXiv2025

Ghidorah: Fast LLM Inference on Edge with Speculative Decoding and Hetero-Core Parallelism

In-situ LLM inference on end-user devices has gained significant interest due to its privacy benefits and reduced dependency on external infrastructure. However, as the decoding process is memory-bandwidth-bound, the diverse processing units in modern end-user devices cannot be fully exploited, resulting in slow LLM inference. This paper presents Ghidorah, a LLM inference system for end-user devices with the unified memory architecture. The key idea of Ghidorah can be summarized in two steps: 1) leveraging speculative decoding approaches to enhance parallelism, and 2) ingeniously distributing workloads across multiple heterogeneous processing units to maximize computing power utilization. Ghidorah includes the hetero-core model parallelism (HCMP) architecture and the architecture-aware profiling (ARCA) approach. The HCMP architecture guides partitioning by leveraging the unified memory design of end-user devices and adapting to the hybrid computational demands of speculative decoding. The ARCA approach is used to determine the optimal speculative strategy and partitioning strategy, balancing acceptance rate with parallel capability to maximize the speedup. Additionally, we optimize sparse computation on ARM CPUs. Experimental results show that Ghidorah can achieve up to 7.6x speedup in the dominant LLM decoding phase compared to the sequential decoding approach in NVIDIA Jetson NX.

cs.DC↗

arXiv2025

gLLM: Global Balanced Pipeline Parallelism System for Distributed LLM Serving with Token Throttling

Pipeline parallelism has emerged as a predominant approach for deploying large language models (LLMs) across distributed nodes, owing to its lower communication overhead compared to tensor parallelism. While demonstrating high throughput in request serving, pipeline parallelism often suffers from performance limitations caused by pipeline bubbles, which are primarily resulted from imbalanced computation delays across batches. Existing methods like Sarathi-Serve attempt to address this through hybrid scheduling of chunked prefill and decode tokens using a fixed token budget. However, such methods may experience significant fluctuations due to either insufficient prefill tokens or uneven distribution of decode tokens, ultimately leading to computational imbalance. To overcome these inefficiencies, we present gLLM, a globally balanced pipeline parallelism system incorporating Token Throttling to effectively mitigate the pipeline bubbles. Our Token Throttling mechanism is a fine-grained scheduling policy that independently regulates the quantities of prefill and decode tokens, thus enabling balanced computation by leveraging global information from the inference system. Specifically, for decode tokens, gLLM maintains near-consistent token count across processing batches. For prefill tokens, it dynamically adjusts batch sizes based on both total pending tokens and the memory utilization rates of key-value cache (KV cache). Furthermore, gLLM runtime adopts an asynchronous execution and message passing architecture specifically optimized for pipeline parallelism characteristics. Experimental evaluations with representative LLMs show that gLLM achieves significant performance improvements, delivering 11% to 398% higher maximum throughput compared to state-of-the-art pipeline or tensor parallelism systems, while simultaneously maintaining lower latency.

cs.DC↗

arXiv2025

SRDiffusion: Accelerate Video Diffusion Inference via Sketching-Rendering Cooperation

Leveraging the diffusion transformer (DiT) architecture, models like Sora, CogVideoX and Wan have achieved remarkable progress in text-to-video, image-to-video, and video editing tasks. Despite these advances, diffusion-based video generation remains computationally intensive, especially for high-resolution, long-duration videos. Prior work accelerates its inference by skipping computation, usually at the cost of severe quality degradation. In this paper, we propose SRDiffusion, a novel framework that leverages collaboration between large and small models to reduce inference cost. The large model handles high-noise steps to ensure semantic and motion fidelity (Sketching), while the smaller model refines visual details in low-noise steps (Rendering). Experimental results demonstrate that our method outperforms existing approaches, over 3$\times$ speedup for Wan with nearly no quality loss for VBench, and 2$\times$ speedup for CogVideoX. Our method is introduced as a new direction orthogonal to existing acceleration strategies, offering a practical solution for scalable video generation.

cs.GR↗

arXiv2025

EcoServe: Enabling Cost-effective LLM Serving with Proactive Intra- and Inter-Instance Orchestration

Existing LLM serving strategies can be categorized based on whether prefill and decode phases are disaggregated: non-disaggregated (NoDG) or fully disaggregated (FuDG). However, the NoDG strategy leads to strong prefill-decode interference and the FuDG strategy highly relies on high-performance interconnects, making them less cost-effective. We introduce EcoServe, a system that enables cost-effective LLM serving on clusters with commodity interconnects. EcoServe is built on the partially disaggregated (PaDG) strategy, applying temporal disaggregation and rolling activation for proactive intra- and inter-instance scheduling. It first disaggregates the prefill and decode phases along the time dimension within a single instance to mitigate inter-phase interference and enhance throughput. Next, it coordinates multiple instances and cyclically activates them to ensure the continuous availability of prefill processing, thereby improving latency. Thus, EcoServe's basic serving unit is the macro instance, within which multiple instances collaborate. It further integrates an adaptive scheduling algorithm to route requests in a macro instance and a mitosis scaling approach to enable fine-grained capacity scaling. Beyond delivering high goodput, EcoServe excels in load balancing, hardware cost, parallelism compatibility, and even engineering simplicity compared to existing solutions. When serving 30B- and 70B-scale models on a production-level cluster with 32 NVIDIA L20 GPUs using commodity Ethernet, EcoServe averagely improves goodput by 82.49%, 86.17%, 122.76%, and 126.96% over four representative NoDG and FuDG systems.

cs.DC↗

arXiv2024

Galaxy: A Resource-Efficient Collaborative Edge AI System for In-situ Transformer Inference

Transformer-based models have unlocked a plethora of powerful intelligent applications at the edge, such as voice assistant in smart home. Traditional deployment approaches offload the inference workloads to the remote cloud server, which would induce substantial pressure on the backbone network as well as raise users' privacy concerns. To address that, in-situ inference has been recently recognized for edge intelligence, but it still confronts significant challenges stemming from the conflict between intensive workloads and limited on-device computing resources. In this paper, we leverage our observation that many edge environments usually comprise a rich set of accompanying trusted edge devices with idle resources and propose Galaxy, a collaborative edge AI system that breaks the resource walls across heterogeneous edge devices for efficient Transformer inference acceleration. Galaxy introduces a novel hybrid model parallelism to orchestrate collaborative inference, along with a heterogeneity-aware parallelism planning for fully exploiting the resource potential. Furthermore, Galaxy devises a tile-based fine-grained overlapping of communication and computation to mitigate the impact of tensor synchronizations on inference latency under bandwidth-constrained edge environments. Extensive evaluation based on prototype implementation demonstrates that Galaxy remarkably outperforms state-of-the-art approaches under various edge environment setups, achieving up to 2.5x end-to-end latency reduction.

cs.DC↗

arXiv2023

ATP: Adaptive Tensor Parallelism for Foundation Models

Foundation models have impressive performance and generalization capabilities across a wide range of applications. The increasing size of the models introduces great challenges for the training. Tensor parallelism is a critical technique that is currently used in almost all foundation model training and has a significant impact on overall training performance. However, current tensor parallelism in machine learning frameworks misses optimization opportunities in fitting various interconnection topologies. In this work, we present ATP, an adaptive tensor parallelism framework for foundation models, which can automatically select the optimal parallel strategy on different interconnections. We propose column- and row-first tensor parallelism based on 2D device meshes and construct a search space. Combined with the hierarchical communication matrix, ATP can identify the optimal strategy in the search space. We also propose chunk-based overlapping to reduce communication overhead. Our evaluations show ATP consistently outperforms the state-of-the-art approaches for various model sizes and interconnects, achieving end-to-end training performance improvements of up to 37-64% on specific interconnects. Based on our theoretical model, the communication overhead of ATP decreases with scaling, indicating a qualitative leap forward.

cs.DC↗

arXiv2022

SAIH: A Scalable Evaluation Methodology for Understanding AI Performance Trend on HPC Systems

Novel artificial intelligence (AI) technology has expedited various scientific research, e.g., cosmology, physics and bioinformatics, inevitably becoming a significant category of workload on high performance computing (HPC) systems. Existing AI benchmarks tend to customize well-recognized AI applications, so as to evaluate the AI performance of HPC systems under predefined problem size, in terms of datasets and AI models. Due to lack of scalability on the problem size, static AI benchmarks might be under competent to help understand the performance trend of evolving AI applications on HPC systems, in particular, the scientific AI applications on large-scale systems. In this paper, we propose a scalable evaluation methodology (SAIH) for analyzing the AI performance trend of HPC systems with scaling the problem sizes of customized AI applications. To enable scalability, SAIH builds a set of novel mechanisms for augmenting problem sizes. As the data and model constantly scale, we can investigate the trend and range of AI performance on HPC systems, and further diagnose system bottlenecks. To verify our methodology, we augment a cosmological AI application to evaluate a real HPC system equipped with GPUs as a case study of SAIH.

cs.DC↗

arXiv2022

EnergonAI: An Inference System for 10-100 Billion Parameter Transformer Models

Large transformer models display promising performance on a wide range of natural language processing (NLP) tasks. Although the AI community has expanded the model scale to the trillion parameter level, the practical deployment of 10-100 billion parameter models is still uncertain due to the latency, throughput, and memory constraints. In this paper, we proposed EnergonAI to solve the challenges of the efficient deployment of 10-100 billion parameter transformer models on single- or multi-GPU systems. EnergonAI adopts a hierarchy-controller system architecture to coordinate multiple devices and efficiently support different parallel patterns. It delegates the execution of sub-models to multiple workers in the single-controller style and applies tensor parallelism and pipeline parallelism among the workers in a multi-controller style. Upon the novel architecture, we propose three techniques, i.e. non-blocking pipeline parallelism, distributed redundant computation elimination, and peer memory pooling. EnergonAI enables the users to program complex parallel code the same as a serial one. Compared with the FasterTransformer, we have proven that EnergonAI has superior performance on latency and throughput. In our experiments, EnergonAI can achieve 37% latency reduction in tensor parallelism, 10% scalability improvement in pipeline parallelism, and it improves the model scale inferred on a single GPU by using a larger heterogeneous memory space at cost of limited performance reduction.

cs.LG↗