相关论文: Pushdown Compression
Learned video compression methods have demonstrated great promise in catching up with traditional video codecs in their rate-distortion (R-D) performance. However, existing learned video compression schemes are limited by the binding of the…
Text compression for large language model (LLM) systems is usually framed as token deletion, retrieval, summarization, or exact reconstruction. We study a more aggressive but explicitly lossy setting: compress text into compact codes that…
Present day computational fluid dynamics simulations generate extremely large amounts of data, sometimes on the order of TB/s. Often, a significant fraction of this data is discarded because current storage systems are unable to keep pace.…
User data confidentiality protection is becoming a rising challenge in the present deep learning research. Without access to data, conventional data-driven model compression faces a higher risk of performance degradation. Recently, some…
In the modern paradigm of multi-agent networks, communication has become one of the main bottlenecks for decentralized optimization, where a large number of agents are involved in minimizing the average of the local cost functions. In this…
In a system that performs both encryption and lossy compression, the conventional way is to compress first and then encrypt the compressed data. This separation approach proves to be optimal. In certain applications where sensitive…
Our goal is compression of massive-scale grid-structured data, such as the multi-terabyte output of a high-fidelity computational simulation. For such data sets, we have developed a new software package called TuckerMPI, a parallel C++/MPI…
Split computing ($\neq$ split learning) is a promising approach to deep learning models for resource-constrained edge computing systems, where weak sensor (mobile) devices are wirelessly connected to stronger edge servers through channels…
Text encoding is one of the most important steps in Natural Language Processing (NLP). It has been done well by the self-attention mechanism in the current state-of-the-art Transformer encoder, which has brought about significant…
In-context learning has established itself as an important learning paradigm for Large Language Models (LLMs). In this paper, we demonstrate that LLMs can learn encoding keys in-context and perform analysis directly on encoded…
As the memory channel count is confined by physical dimensions, memory expanders appear to be a promising approach to extending memory capacity and channels by augmenting the existing I/O interface (e.g., PCIe) with memory-semantic…
Advances in large language models have driven strong performance across many tasks, but their memory and compute costs still hinder deployment. SVD-based compression reduces storage and can speed up inference via low-rank factors, yet…
Modern smart distribution system requires storage, transmission and processing of big data generated by sensors installed in electric meters. On one hand, this data is essentially required for intelligent decision making by smart grid but…
Computing problems that handle large amounts of data necessitate the use of lossless data compression for efficient storage and transmission. We present a novel lossless universal data compression algorithm that uses parallel computational…
Efficient key-value (KV) cache management is crucial for the practical deployment of large language models (LLMs), yet existing compression techniques often incur a trade-off between performance degradation and computational overhead. We…
Pushdown systems (PDSs) and recursive state machines (RSMs), which are linearly equivalent, are standard models for interprocedural analysis. Yet RSMs are more convenient as they (a) explicitly model function calls and returns, and (b)…
With ever-increasing volumes of scientific data produced by HPC applications, significantly reducing data size is critical because of limited capacity of storage space and potential bottlenecks on I/O or networks in writing/reading or…
Several emerging non-volatile (NV) memory technologies are rising as interesting alternatives to build the Last-Level Cache (LLC). Their advantages, compared to SRAM memory, are higher density and lower static power, but write operations…
Zero-shot Composed Image Retrieval (ZS-CIR) enables image search using a reference image and a text prompt without requiring specialized text-image composition networks trained on large-scale paired data. However, current ZS-CIR approaches…
Error-bounded lossy compression has been a critical technique to significantly reduce the sheer amounts of simulation datasets for high-performance computing (HPC) scientific applications while effectively controlling the data distortion…