arXiv subjects

Olivia Hsu

Publications and source records attributed to Olivia Hsu.

11 recordsLinked to original sources

arXiv2026

M*: A Modular, Extensible, Serving System for Multimodal Models

We are entering a new era of composite model architectures that integrate diverse components such as vision encoders, language backbones, diffusion and flow heads, audio codecs, action generators, and world-model predictors. Such architectures underpin a broad class of multimodal models, including unified multimodal models, omni models, speech-language models, vision-language-action policies, and world models. However, existing model serving frameworks were built on narrow assumptions about model structure, making them ill-suited to accommodate this new architectural diversity. Here we present M*, a universal serving system for efficient serving of composite AI models. M* represents models as dataflow graphs, processing requests spanning diverse modalities and tasks as traversals over these graphs. The core insight is a modular abstraction that supports arbitrary composition of model components, flexible placement onto a physical cluster, and model-agnostic optimizations within a distributed runtime. We call this abstraction the Walk Graph and show how it can concisely capture composite models from a broad range of families. We instantiate M* on representative models and find that it achieves, on average, 20% lower end-to-end latency than vLLM-Omni for text-to-image workloads on BAGEL, while delivering up to 2.9x lower real-time factor and 2.7x higher throughput for text-to-speech workloads on Qwen3-Omni. M* also outperforms the V-JEPA 2-AC rollout baseline for robotic planning by up to 12.5x. Thus, our work paves the road towards more efficient serving of complex models with minimal developer effort.

cs.LG↗

arXiv2026

Streaming Tensor Programs: A Streaming Abstraction for Dynamic Parallelism

Dynamic behaviors are becoming prevalent in tensor applications, like machine learning, where many widely used models contain data-dependent tensor shapes and control flow. However, the limited expressiveness of prior programming abstractions for spatial dataflow accelerators (SDAs) forces these dynamic behaviors to be implemented statically and/or unoptimized. To address these challenges, we present Streaming Tensor Programs (STeP), a streaming abstraction that enables dynamic tensor workloads to run efficiently on SDAs. STeP introduces flexible routing operators, an explicit memory hierarchy, and symbolic-shape semantics that expose dynamic data rates and tensor dimensions. These capabilities unlock new optimizations, like dynamic tiling, dynamic parallelization, and configuration time-multiplexing, that adapt SDA execution to dynamic behaviors while preserving dataflow efficiency. Using a cycle-approximate simulator on representative LLM layers and a full model with real-world traces, STeP enables: dynamic tiling that breaks the Pareto-optimal frontier from prior work, dynamic parallelization that improves latency by ~2.72x, and configuration time-multiplexing that increases compute utilization by ~2.64x over prior SDA abstractions and their implementations.

cs.PL↗

arXiv2026

FuseFlow: A Fusion-Centric Compilation Framework for Sparse Deep Learning on Streaming Dataflow

As deep learning models scale, sparse computation and specialized dataflow hardware have emerged as powerful solutions to address efficiency. We propose FuseFlow, a compiler that converts sparse machine learning models written in PyTorch to fused sparse dataflow graphs for reconfigurable dataflow architectures (RDAs). FuseFlow is the first compiler to support general cross-expression fusion of sparse operations. In addition to fusion across kernels (expressions), FuseFlow also supports optimizations like parallelization, dataflow ordering, and sparsity blocking. It targets a cycle-accurate dataflow simulator for microarchitectural analysis of fusion strategies. We use FuseFlow for design-space exploration across four real-world machine learning applications with sparsity, showing that full fusion (entire cross-expression fusion across all computation in an end-to-end model) is not always optimal for sparse models-fusion granularity depends on the model itself. FuseFlow also provides a heuristic to identify and prune suboptimal configurations. Using Fuseflow, we achieve performance improvements, including a ~2.7x speedup over an unfused baseline for GPT-3 with BigBird block-sparse attention.

cs.LG↗

arXiv2025

Adaptive Self-improvement LLM Agentic System for ML Library Development

ML libraries, often written in architecture-specific programming languages (ASPLs) that target domain-specific architectures, are key to efficient ML systems. However, writing these high-performance ML libraries is challenging because it requires expert knowledge of ML algorithms and the ASPL. Large language models (LLMs), on the other hand, have shown general coding capabilities. However, challenges remain when using LLMs for generating ML libraries using ASPLs because 1) this task is complicated even for experienced human programmers and 2) there are limited code examples because of the esoteric and evolving nature of ASPLs. Therefore, LLMs need complex reasoning with limited data in order to complete this task. To address these challenges, we introduce an adaptive self-improvement agentic system. In order to evaluate the effectiveness of our system, we construct a benchmark of a typical ML library and generate ASPL code with both open and closed-source LLMs on this benchmark. Our results show improvements of up to $3.9\times$ over a baseline single LLM.

cs.CL↗

arXiv2025

Ember: A Compiler for Efficient Embedding Operations on Decoupled Access-Execute Architectures

Irregular embedding lookups are a critical bottleneck in recommender models, sparse large language models, and graph learning models. In this paper, we first demonstrate that, by offloading these lookups to specialized access units, Decoupled Access-Execute (DAE) processors achieve 2.6$\times$ higher performance and 6.4$\times$ higher performance/watt than GPUs on end-to-end models. Then, we propose the Ember compiler for automatically generating optimized DAE code from PyTorch and TensorFlow. Conversely from other DAE compilers, Ember features multiple intermediate representations specifically designed for different optimization levels. In this way, Ember can implement all optimizations to match the performance of hand-written code, unlocking the full potential of DAE architectures at scale.

cs.AR↗

arXiv2024

DFModel: Design Space Optimization of Large-Scale Systems Exploiting Dataflow Mappings

We propose DFModel, a modeling framework for mapping dataflow computation graphs onto large-scale systems. Mapping a workload to a system requires optimizing dataflow mappings at various levels, including the inter-chip (between chips) level and the intra-chip (within a chip) level. DFModel is, to the best of our knowledge, the first framework to perform the optimization at multiple levels of the memory hierarchy and the interconnection network hierarchy. We use DFModel to explore a wide range of workloads on a variety of systems. Evaluated workloads include two state-of-the-art machine learning applications (Large Language Models and Deep Learning Recommendation Models) and two high-performance computing applications (High Performance LINPACK and Fast Fourier Transform). System parameters investigated span the combination of dataflow and traditional accelerator architectures, memory technologies (DDR, HBM), interconnect technologies (PCIe, NVLink), and interconnection network topologies (torus, DGX, dragonfly). For a variety of workloads on a wide range of systems, the DFModel provided a mapping that predicts an average of 1.25X better performance compared to the ones measured on real systems. DFModel shows that for large language model training, dataflow architectures achieve 1.52X higher performance, 1.59X better cost efficiency, and 1.6X better power efficiency compared to non-dataflow architectures. On an industrial system with dataflow architectures, the DFModel-optimized dataflow mapping achieves a speedup of 6.13X compared to non-dataflow mappings from previous performance models such as Calculon, and 1.52X compared to a vendor provided dataflow mapping.

cs.AR↗

arXiv2024

Compilation of Modular and General Sparse Workspaces

Recent years have seen considerable work on compiling sparse tensor algebra expressions. This paper addresses a shortcoming in that work, namely how to generate efficient code (in time and space) that scatters values into a sparse result tensor. We address this shortcoming through a compiler design that generates code that uses sparse intermediate tensors (sparse workspaces) as efficient adapters between compute code that scatters and result tensors that do not support random insertion. Our compiler automatically detects sparse scattering behavior in tensor expressions and inserts necessary intermediate workspace tensors. We present an algorithm template for workspace insertion that is the backbone of our code generation algorithm. Our algorithm template is modular by design, supporting sparse workspaces that span multiple user-defined implementations. Our evaluation shows that sparse workspaces can be up to 27.12$\times$ faster than the dense workspaces of prior work. On the other hand, dense workspaces can be up to 7.58$\times$ faster than the sparse workspaces generated by our compiler in other situations, which motivates our compiler design that supports both. Our compiler produces sequential code that is competitive with hand-optimized linear and tensor algebra libraries on the expressions they support, but that generalizes to any other expression. Sparse workspaces are also more memory efficient than dense workspaces as they compress away zeros. This compression can asymptotically decrease memory usage, enabling tensor computations on data that would otherwise run out of memory.

cs.PL↗

arXiv2023

BaCO: A Fast and Portable Bayesian Compiler Optimization Framework

We introduce the Bayesian Compiler Optimization framework (BaCO), a general purpose autotuner for modern compilers targeting CPUs, GPUs, and FPGAs. BaCO provides the flexibility needed to handle the requirements of modern autotuning tasks. Particularly, it deals with permutation, ordered, and continuous parameter types along with both known and unknown parameter constraints. To reason about these parameter types and efficiently deliver high-quality code, BaCO uses Bayesian optimiza tion algorithms specialized towards the autotuning domain. We demonstrate BaCO's effectiveness on three modern compiler systems: TACO, RISE & ELEVATE, and HPVM2FPGA for CPUs, GPUs, and FPGAs respectively. For these domains, BaCO outperforms current state-of-the-art autotuners by delivering on average 1.36x-1.56x faster code with a tiny search budget, and BaCO is able to reach expert-level performance 2.9x-3.9x faster.

cs.PL↗

arXiv2023

The Sparse Abstract Machine

We propose the Sparse Abstract Machine (SAM), an abstract machine model for targeting sparse tensor algebra to reconfigurable and fixed-function spatial dataflow accelerators. SAM defines a streaming dataflow abstraction with sparse primitives that encompass a large space of scheduled tensor algebra expressions. SAM dataflow graphs naturally separate tensor formats from algorithms and are expressive enough to incorporate arbitrary iteration orderings and many hardware-specific optimizations. We also present Custard, a compiler from a high-level language to SAM that demonstrates SAM's usefulness as an intermediate representation. We automatically bind from SAM to a streaming dataflow simulator. We evaluate the generality and extensibility of SAM, explore the performance space of sparse tensor algebra optimizations using SAM, and show SAM's ability to represent dataflow hardware.

cs.AR↗

arXiv2023

Inclusive Study Group Formation At Scale

Underrepresented students face many significant challenges in their education. In particular, they often have a harder time than their peers from majority groups in building long-term high-quality study groups. This challenge is exacerbated in remote-learning scenarios, where students are unable to meet face-to-face and must rely on pre-existing networks for social support. We present a scalable system that removes structural obstacles faced by underrepresented students and supports all students in building inclusive and flexible study groups. One of our main goals is to make the traditionally informal and unstructured process of finding study groups for homework more equitable by providing a uniform but lightweight structure. We aim to provide students from underrepresented groups an experience that is similar in quality to that of students from majority groups. Our process is unique in that it allows students the opportunity to request group reassignments during the semester if they wish. Unlike other collaboration tools our system is not mandatory and does not use peer-evaluation. We trialed our approach in a 1000+ student introductory Engineering and Computer Science course that was conducted entirely online during the COVID-19 pandemic. We find that students from underrepresented backgrounds were more likely to ask for group-matching support compared to students from majority groups. At the same time, underrepresented students that we matched into study groups had group experiences that were comparable to students we matched from majority groups. B-range students in high-comfort and high-quality groups had improved learning outcomes.

cs.CY↗

arXiv2022

Stardust: Compiling Sparse Tensor Algebra to a Reconfigurable Dataflow Architecture

We introduce Stardust, a compiler that compiles sparse tensor algebra to reconfigurable dataflow architectures (RDAs). Stardust introduces new user-provided data representation and scheduling language constructs for mapping to resource-constrained accelerated architectures. Stardust uses the information provided by these constructs to determine on-chip memory placement and to lower to the Capstan RDA through a parallel-patterns rewrite system that targets the Spatial programming model. The Stardust compiler is implemented as a new compilation path inside the TACO open-source system. Using cycle-accurate simulation, we demonstrate that Stardust can generate more Capstan tensor operations than its authors had implemented and that it results in 138$\times$ better performance than generated CPU kernels and 41$\times$ better performance than generated GPU kernels.

cs.PL↗