arXiv subjects

Tuowei Wang

Publications and source records attributed to Tuowei Wang.

11 recordsLinked to original sources

arXiv2026

Efficient On-Device Diffusion LLM Inference with Mobile NPU

Diffusion large language models (dLLMs) accelerate generation by denoising multiple tokens in parallel, making them attractive for latency-sensitive mobile inference. However, repeated denoising introduces substantial computation on smartphones. Mobile neural processing units (NPUs) offer high-throughput dense matrix computation, but efficiently exploiting them remains challenging: token commitment shrinks per-block effective workloads, token revision complicates KV cache reuse, and limited NPU-visible address space incurs costly remapping and data transfer overheads. In this paper, we propose llada.cpp, the first NPU-aware inference framework for accelerating dLLMs on smartphones. llada.cpp aligns block-wise dLLM inference with the execution characteristics of mobile NPUs through three techniques. (1) Multi-Block Speculative Decoding fills the shrinking workload in late-stage current-block decoding with speculative future-block tokens. (2) Dual-Path Progressive Revision keeps committed tokens revisable until stable and refreshes unstable tokens through a CPU-side path without stalling dense NPU execution. (3) Swap-Optimized Memory Runtime compacts NPU-visible address layouts and overlaps data staging with NPU computation to reduce remapping and transfer overheads. We implement llada.cpp as an end-to-end framework and evaluate it across diverse hardware platforms and dLLM workloads. llada.cpp reduces LLaDA-8B generation latency by 17x-42x over the CPU baseline with prefix KV cache reuse, while preserving generation quality.

cs.LG↗

arXiv2026

Lever: Speculative LLM Inference on Smartphones

Large language models (LLMs) are increasingly needed for interactive mobile applications, but high-quality models exceed the limited DRAM available on smartphones. Flash storage can hold larger models, yet flash-backed inference is slow because autoregressive decoding repeatedly invokes the target model and incurs costly I/O. We observe that speculative decoding is a natural fit for this setting: a small draft model can remain in DRAM, while a larger flash-resident target model verifies multiple candidate tokens per invocation. However, existing methods assume server-class accelerators and fail to account for prolonged I/O latency, limited computation parallelism, and irregular speculation execution. We present Lever, an end-to-end system for efficient flash-backed LLM inference on smartphones. Lever jointly optimizes the three stages of speculative decoding under mobile constraints. For drafting, it builds token trees using an I/O- and compute-aware gain-cost objective. For verification, it prunes low-value branches through early-exit prediction to reduce target-model computation. For execution, it maps speculation efficiently across mobile CPU-NPU hardware to improve utilization. Comprehensive evaluations show that Lever reduces inference latency by an average of 2.93x over baseline flash-offloaded inference and 1.50x over conventional speculative decoding, narrowing the latency gap between flash-backed and memory-resident LLM inference.

cs.LG↗

arXiv2026

Mosaic: Cross-Modal Clustering for Efficient Video Understanding

Large vision-language models (VLMs) are enabling interactive video reasoning, giving rise to streaming long-video understanding. In this setting, frames arrive continuously, while the system preserves long-term context and generates responses under strict latency constraints. A central challenge is KVCache management: as video streams grow, KVCache expands rapidly, increasing computation and memory overhead. Existing retrieval-based approaches exploit attention sparsity and offload inactive KVCache from GPU to CPU memory, but their token-level design causes high management overhead and fragmented data movement. We present Mosaic, the first cluster-driven VLM inference system for streaming long-video understanding. Our key insight is that VLM KVCache exhibits an implicit cross-modal clustering structure: retrieved KV states form groups jointly shaped by visual coherence and semantic relevance. Based on this observation, Mosaic uses cross-modal clusters as the basic unit of KVCache organization, maintenance, and retrieval. Evaluations show that Mosaic outperforms state-of-the-art baselines, achieving up to 1.38x speedup.

cs.PF↗

arXiv2026

Swarm: Co-Activation Aware KVCache Offloading Across Multiple SSDs

The key-value (KV) cache has become the dominant contributor to memory consumption in large language model (LLM) inference. Although offloading KVCache from GPU high-bandwidth memory (HBM) to CPU DRAM alleviates device memory pressure, DRAM remains capacity-limited and costly for large, persistent workloads. Solid-state drives (SSDs) provide a cost-effective alternative, but naive SSD-based paging is fundamentally bandwidth-bound due to limited PCIe throughput and per-device bandwidth constraints. In this paper, we observe that KVCache activations in real-world workloads exhibit strong and stable correlations. We term this phenomenon KVCache Co-Activation, where accessing a KV entry is often accompanied by a stable and recurring set of other KV entries. Leveraging this property, we present Swarm, an SSD-based KVCache offloading system that converts bandwidth-bound single-device access into parallel I/O across multiple SSDs. Specifically, Swarm clusters co-activated KV entries offline and distributes the resulting clusters across SSDs using graph-based placement with selective replication to maximize parallel I/O bandwidth. At runtime, Swarm performs load-balanced cluster retrieval and dynamically adapts clustering and caching decisions to sustain high bandwidth utilization under evolving access patterns. Evaluations show that Swarm reduces I/O time by 2.41x and improves effective bandwidth utilization by 2.72x.

cs.PF↗

arXiv2025

DynaKV: Enabling Accurate and Efficient Long-Sequence LLM Decoding on Smartphones

As the demand for human-like reasoning, multi-turn dialogues, and long-form responses grows, large language models (LLMs) are increasingly expected to support efficient and effective long-sequence decoding. However, due to limited DRAM capacity, long-seuqence LLM decoding on smartphones is constrained by the key-value cache (KVCache), whose memory footprint increases linearly with sequence length. Retrieval-based methods mitigate DRAM pressure by offloading KVCache to flash and retrieving query-relevant entries through cluster-based indexing. Unfortunately, as decoding progresses, KVCache distribution shifts render static or local cluster updates progressively misaligned, excluding essential entries or fetching redundant ones. These issues are further exacerbated by smartphone-specific limitations in bandwidth, IOPS, and memory capacity. We propose DynaKV, the first adaptive KVCache management approach that jointly addresses accuracy and efficiency for long-sequence decoding on smartphones. DynaKV integrates three key techniques: (1) Migration-Free Cluster Adaptation, which adaptively splits clusters during retrieval without incurring additional transfers; (2) Continuity-Centric Flash Management, which co-locates correlated entries and clusters and employs a dual-head layout for efficient updates; and (3) Memory-Efficient Cache Design, which virtualizes cache space across DRAM and flash and extends replacement policies to align with cluster-level access patterns. Evaluations demonstrate that DynaKV improves retrieval accuracy and reduces end-to-end latency compared to state-of-the-art solutions, achieving average gains of $1.38\times$ in accuracy and $1.47\times$ speedups. Furthermore, the insights of DynaKV naturally extend to other long-context workloads and multi-tier memory hierarchies, underscoring its broader applicability.

cs.DC↗

arXiv2025

Long Exposure: Accelerating Parameter-Efficient Fine-Tuning for LLMs under Shadowy Sparsity

The adaptation of pre-trained large language models (LLMs) to diverse downstream tasks via fine-tuning is critical for numerous applications. However, the inefficiency of parameter-efficient fine-tuning (PEFT) techniques presents significant challenges in terms of time investments and operational costs. In this paper, we first introduce a nuanced form of sparsity, termed Shadowy Sparsity, which is distinctive in fine-tuning and has not been adequately addressed for acceleration. Under Shadowy Sparsity, we propose Long Exposure, an efficient system to accelerate PEFT for LLMs. Long Exposure comprises three key components: Shadowy-sparsity Exposer employs a prolonged sensing range to capture more sparsity details under shadowy sparsity; Sequence-oriented Predictor provides efficient yet accurate predictions to handle large sequence inputs and constantly-evolving parameters; and Dynamic-aware Operator facilitates more structured computational patterns and coalesced memory accesses, addressing dynamic sparse operations. Extensive evaluations show that Long Exposure outperforms state-of-the-arts with up to a $2.49\times$ speedup in end-to-end fine-tuning, offering promising advancements in accelerating PEFT for LLMs.

cs.LG↗

arXiv2025

Scaling LLM Test-Time Compute with Mobile NPU on Smartphones

Deploying Large Language Models (LLMs) on mobile devices faces the challenge of insufficient performance in smaller models and excessive resource consumption in larger ones. This paper highlights that mobile Neural Processing Units (NPUs) have underutilized computational resources, particularly their matrix multiplication units, during typical LLM inference. To leverage this wasted compute capacity, we propose applying parallel test-time scaling techniques on mobile NPUs to enhance the performance of smaller LLMs. However, this approach confronts inherent NPU challenges, including inadequate hardware support for fine-grained quantization and low efficiency in general-purpose computations. To overcome these, we introduce two key techniques: a hardware-aware tile quantization scheme that aligns group quantization with NPU memory access patterns, and efficient LUT-based replacements for complex operations such as Softmax and dequantization. We design and implement an end-to-end inference system that leverages the NPU's compute capability to support test-time scaling on Qualcomm Snapdragon platforms. Experiments show our approach brings significant speedups: up to 19.0 for mixed-precision GEMM and 2.2 for Softmax. More importantly, we demonstrate that smaller models using test-time scaling can match or exceed the accuracy of larger models, achieving a new performance-cost Pareto frontier.

cs.DC↗

arXiv2025

LeMo: Enabling LEss Token Involvement for MOre Context Fine-tuning

The escalating demand for long-context applications has intensified the necessity of extending the LLM context windows. Despite recent fine-tuning approaches successfully expanding context lengths, their high memory footprints, especially for activations, present a critical practical limitation. Current parameter-efficient fine-tuning methods prioritize reducing parameter update overhead over addressing activation memory constraints. Similarly, existing sparsity mechanisms improve computational efficiency but overlook activation memory optimization due to the phenomenon of Shadowy Activation. In this paper, we propose LeMo, the first LLM fine-tuning system that explores and exploits a new token-level sparsity mechanism inherent in long-context scenarios, termed Contextual Token Sparsity. LeMo minimizes redundant token involvement by assessing the informativeness of token embeddings while preserving model accuracy. Specifically, LeMo introduces three key techniques: (1) Token Elimination, dynamically identifying and excluding redundant tokens across varying inputs and layers. (2) Pattern Prediction, utilizing well-trained predictors to approximate token sparsity patterns with minimal overhead. (3) Kernel Optimization, employing permutation-free and segment-based strategies to boost system performance. We implement LeMo as an end-to-end fine-tuning system compatible with various LLM architectures and other optimization techniques. Comprehensive evaluations demonstrate that LeMo reduces memory consumption by up to 1.93x and achieves up to 1.36x speedups, outperforming state-of-the-art fine-tuning systems.

cs.CL↗

arXiv2024

Matryoshka: Optimization of Dynamic Diverse Quantum Chemistry Systems via Elastic Parallelism Transformation

AI infrastructures, predominantly GPUs, have delivered remarkable performance gains for deep learning. Conversely, scientific computing, exemplified by quantum chemistry systems, suffers from dynamic diversity, where computational patterns are more diverse and vary dynamically, posing a significant challenge to sponge acceleration off GPUs. In this paper, we propose Matryoshka, a novel elastically-parallel technique for the efficient execution of quantum chemistry system with dynamic diversity on GPU. Matryoshka capitalizes on Elastic Parallelism Transformation, a property prevalent in scientific systems yet underexplored for dynamic diversity, to elastically realign parallel patterns with GPU architecture. Structured around three transformation primitives (Permutation, Deconstruction, and Combination), Matryoshka encompasses three core components. The Block Constructor serves as the central orchestrator, which reformulates data structures accommodating dynamic inputs and constructs fine-grained GPU-efficient compute blocks. Within each compute block, the Graph Compiler operates offline, generating high-performance code with clear computational path through an automated compilation process. The Workload Allocator dynamically schedules workloads with varying operational intensities to threads online. It achieves highly efficient parallelism for compute-intensive operations and facilitates fusion with neighboring memory-intensive operations automatically. Extensive evaluation shows that Matryoshka effectively addresses dynamic diversity, yielding acceleration improvements of up to 13.86x (average 9.41x) over prevailing state-of-the-art approaches on 13 quantum chemistry systems.

cs.DC↗

arXiv2024

Neuralink: Fast LLM Inference on Smartphones with Neuron Co-Activation Linking

Large Language Models (LLMs) have achieved remarkable success across various domains, yet deploying them on mobile devices remains an arduous challenge due to their extensive computational and memory demands. While lightweight LLMs have been developed to fit mobile environments, they suffer from degraded model accuracy. In contrast, sparsity-based techniques minimize DRAM usage by selectively transferring only relevant neurons to DRAM while retaining the full model in external storage, such as flash. However, such approaches are critically limited by numerous I/O operations, particularly on smartphones with severe IOPS constraints. In this paper, we propose Neuralink, a novel approach that accelerates LLM inference on smartphones by optimizing neuron placement in flash memory. Neuralink leverages the concept of Neuron Co-Activation, where neurons frequently activated together are linked to facilitate continuous read access and optimize I/O efficiency. Our approach incorporates a two-stage solution: an offline stage that reorganizes neuron placement based on co-activation patterns, and an online stage that employs tailored data access and caching strategies to align well with hardware characteristics. Evaluations conducted on a variety of smartphones and LLMs demonstrate that Neuralink achieves on average $1.49\times$ improvements in end-to-end latency compared to the state-of-the-art. As the first solution to optimize storage placement under sparsity, Neuralink explores a new optimization space at the intersection of sparsity-driven algorithm and storage-level system co-design for LLM inference.

cs.LG↗

arXiv2022

OLLIE: Derivation-based Tensor Program Optimizer

Boosting the runtime performance of deep neural networks (DNNs) is critical due to their wide adoption in real-world tasks. Existing approaches to optimizing the tensor algebra expression of a DNN only consider expressions representable by a fixed set of predefined operators, missing possible optimization opportunities between general expressions. We propose OLLIE, the first derivation-based tensor program optimizer. OLLIE optimizes tensor programs by leveraging transformations between general tensor algebra expressions, enabling a significantly larger expression search space that includes those supported by prior work as special cases. OLLIE uses a hybrid derivation-based optimizer that effectively combines explorative and guided derivations to quickly discover highly optimized expressions. Evaluation on seven DNNs shows that OLLIE can outperform existing optimizers by up to 2.73$\times$ (1.46$\times$ on average) on an A100 GPU and up to 2.68$\times$ (1.51$\times$) on a V100 GPU, respectively.

cs.LG↗