Related papers: CREST: Effectively Compacting a Datastore For Retr…
To mitigate the high inference latency stemming from autoregressive decoding in Large Language Models (LLMs), Speculative Decoding has emerged as a novel decoding paradigm for LLM inference. In each decoding step, this method first drafts…
Since ChatGPT released its API for public use, the number of applications built on top of commercial large language models (LLMs) increase exponentially. One popular usage of such models is leveraging its in-context learning ability and…
Deployment of autoregressive large language models (LLMs) is costly, and as these models increase in size, the associated costs will become even more considerable. Consequently, different methods have been proposed to accelerate the token…
Motivation: High-throughput sequencing (HTS) enables population-scale genomics but generates massive datasets, creating bottlenecks in storage, transfer, and analysis. FASTQ, the standard format for over two decades, stores one byte per…
We present a new data structure called the \emph{Compressed Random Access Memory} (CRAM) that can store a dynamic string $T$ of characters, e.g., representing the memory of a computer, in compressed form while achieving asymptotically…
Contrastive image-text models such as CLIP form the building blocks of many state-of-the-art systems. While they excel at recognizing common generic concepts, they still struggle on fine-grained entities which are rare, or even absent from…
Speculative sampling reduces the latency of autoregressive decoding for target model LLMs without sacrificing inference quality, by using a cheap draft model to suggest a candidate token and a verification criterion to accept or resample…
Speculative decoding (SD) has been demonstrated as an effective technique for lossless LLM inference acceleration. Retrieval-based SD methods, one kind of model-free method, have yielded promising speedup, but they often rely on incomplete…
SRAM-based cache memory faces several scalability limitations in deep nanoscale technologies, e.g., high leakage current, low cell stability, and low density. Emerging Non-Volatile Memory (NVM) technologies have received lots of attention…
Speculative Decoding (SD) ensures that the output matches the target model's distribution exactly. However, we argue that this distribution matching requirement is too stringent and results in unnecessarily low acceptance rates, limiting…
In this work, we study the robust subspace tracking (RST) problem and obtain one of the first two provable guarantees for it. The goal of RST is to track sequentially arriving data vectors that lie in a slowly changing low-dimensional…
Speculative decoding is a widely used technique for accelerating inference in large language models (LLMs), but its performance degrades as input length grows, with significant drops even at moderate lengths. Yet, this early degradation has…
This paper presents an efficient multi-scale vision Transformer, called ResT, that capably served as a general-purpose backbone for image recognition. Unlike existing Transformer methods, which employ standard Transformer blocks to tackle…
The past few years have witnessed a growing interest in LLM-based recommender systems (RSs), although their industrial deployment remains in a preliminary stage. Most existing deployments leverage LLMs offline as feature enhancers,…
Dimension reduction is widely regarded as an effective way for decreasing the computation, storage and communication loads of data-driven intelligent systems, leading to a growing demand for statistical methods that allow analysis (e.g.,…
Maximum-likelihood (ML) decoding can be used to obtain the optimal performance of error correction codes. However, the size of the search space and consequently the decoding complexity grows exponentially, making it impractical to be…
We study semantic compression for text where meanings contained in the text are conveyed to a source decoder, e.g., for classification. The main motivator to move to such an approach of recovering the meaning without requiring exact…
In most natural language inference problems, sentence representation is needed for semantic retrieval tasks. In recent years, pre-trained large language models have been quite effective for computing such representations. These models…
Autoregressive language models demonstrate excellent performance in various scenarios. However, the inference efficiency is limited by its one-step-one-word generation mode, which has become a pressing problem recently as the models become…
In 2018, Yang et al. introduced a novel and effective approach, using maximum distance separable (MDS) codes, to mitigate the impact of elasticity in cloud computing systems. This approach is referred to as coded elastic computing. Some…