Related papers: SPARTA: A Divide and Conquer Approach to Address T…

Heterogeneous Memory Architecture (HMA) aims to optimize memory usage by leveraging a combination of memory types, such as high-bandwidth memory (HBM), commodity DRAM, and non-volatile memory (NVM), when utilized as main memory. To achieve…

Pre-trained speech Transformers in speech translation (ST) have facilitated state-of-the-art (SotA) results; yet, using such encoders is computationally expensive. To improve this, we present a novel Reducer Adaptor block, RedApt, that…

Fine-tuning large language models (LLMs) is often limited by the memory available on commodity GPUs. Parameter-efficient fine-tuning (PEFT) methods such as QLoRA reduce the number of trainable parameters, yet still incur high memory usage…

Transformers are ubiquitous in Natural Language Processing (NLP) tasks, but they are difficult to be deployed on hardware due to the intensive computation. To enable low-latency inference on resource-constrained hardware platforms, we…

Efficient large-scale inference of transformer-based large language models (LLMs) remains a fundamental systems challenge, frequently requiring multi-GPU parallelism to meet stringent latency and throughput targets. Conventional tensor…

Finetuning large language models (LLMs) in federated learning (FL) settings has become increasingly important as it allows resource-constrained devices to finetune a model using private data. However, finetuning LLMs using backpropagation…

Index structures often materialize one or multiple levels of explicit indirections (aka pointers) to allow for a quick traversal to the data of interest. Unfortunately, dereferencing a pointer to go from one level to the other is costly…

Limited memory bandwidth is a critical bottleneck in modern systems. 3D-stacked DRAM enables higher bandwidth by leveraging wider Through-Silicon-Via (TSV) channels, but today's systems cannot fully exploit them due to the limited internal…

This paper proposes a novel approach to address the challenge that pretrained VLA models often fail to effectively improve performance and reduce adaptation costs during standard supervised finetuning (SFT). Some advanced finetuning methods…

In-memory computing is a promising approach to addressing the processor-memory data transfer bottleneck in computing systems. We propose Spin-Transfer Torque Compute-in-Memory (STT-CiM), a design for in-memory computing with Spin-Transfer…

The attention mechanism is becoming increasingly popular in Natural Language Processing (NLP) applications, showing superior performance than convolutional and recurrent architectures. However, attention becomes the compution bottleneck…

This paper proposes multiple extensions to the popular bicriterion transit routing approach -- Trip-Based Transit Routing (TBTR). Specifically, building on the premise of the HypRAPTOR algorithm, we first extend TBTR to its partitioning…

The tensor-train (TT) format is a data-sparse tensor representation commonly used in high dimensional function approximations arising from computational and data sciences. Various sequential and parallel TT decomposition algorithms have…

Sparse coding is a core building block in many data analysis and machine learning pipelines. Typically it is solved by relying on generic optimization techniques, such as the Iterative Soft Thresholding Algorithm and its accelerated version…

Simultaneous speech translation is an essential communication task difficult for humans whereby a translation is generated concurrently with oncoming speech inputs. For such a streaming task, transformers using block processing to break an…

Memory management operations that modify page-tables, typically performed during memory allocation/deallocation, are infamous for their poor performance in highly threaded applications, largely due to process-wide TLB shootdowns that the OS…

Graph partitioning has long been seen as a viable approach to address Graph DBMS scalability. A partitioning, however, may introduce extra query processing latency unless it is sensitive to a specific query workload, and optimised to…

Soft Prompt Tuning (SPT) is a parameter-efficient method for adapting pre-trained language models (PLMs) to specific tasks by inserting learnable embeddings, or soft prompts, at the input layer of the PLM, without modifying its parameters.…

Modern large-scale deep learning workloads highlight the need for parallel execution across many devices in order to fit model data into hardware accelerator memories. In these settings, array redistribution may be required during a…

Number Theoretic Transform (NTT) is an essential mathematical tool for computing polynomial multiplication in promising lattice-based cryptography. However, costly division operations and complex data dependencies make efficient and…