Personalization of Large Language Models: A Survey

Personalization of Large Language Models (LLMs) has recently become increasingly important with a wide range of applications. Despite the importance and recent progress, most existing works on personalized LLMs have focused either entirely on (a) personalized text generation or (b) leveraging LLMs for personalization-related downstream applications, such as recommendation systems. In this work, we bridge the gap between these two separate main directions for the first time by introducing a taxonomy for personalized LLM usage and summarizing the key differences and challenges. We provide a formalization of the foundations of personalized LLMs that consolidates and expands notions of personalization of LLMs, defining and discussing novel facets of personalization, usage, and desiderata of personalized LLMs. We then unify the literature across these diverse fields and usage scenarios by proposing systematic taxonomies for the granularity of personalization, personalization techniques, datasets, evaluation methods, and applications of personalized LLMs. Finally, we highlight challenges and important open problems that remain to be addressed. By unifying and surveying recent research using the proposed taxonomies, we aim to provide a clear guide to the existing literature and different facets of personalization in LLMs, empowering both researchers and practitioners.

A Framework for Fine-Tuning LLMs using Heterogeneous Feedback

Large language models (LLMs) have been applied to a wide range of tasks, including text summarization, web navigation, and chatbots. They have benefitted from supervised fine-tuning (SFT) and reinforcement learning from human feedback (RLHF) following an unsupervised pretraining. These datasets can be difficult to collect, limited in scope, and vary in sample quality. Additionally, datasets can vary extensively in supervision format, from numerical to binary as well as multi-dimensional with many different values. We present a framework for fine-tuning LLMs using heterogeneous feedback, which has two main components. First, we combine the heterogeneous feedback data into a single supervision format, compatible with methods like SFT and RLHF. Next, given this unified feedback dataset, we extract a high-quality and diverse subset to obtain performance increases potentially exceeding the full dataset. We conduct extensive experiments to understand the effectiveness of these techniques for incorporating heterogeneous feedback, and demonstrate improvements from using a high-quality and diverse subset of the data. We find that our framework is able to improve models in multiple areas simultaneously, such as in instruction following and bias reduction.

Approximate Caching for Efficiently Serving Diffusion Models

Text-to-image generation using diffusion models has seen explosive popularity owing to their ability in producing high quality images adhering to text prompts. However, production-grade diffusion model serving is a resource intensive task that not only require high-end GPUs which are expensive but also incurs considerable latency. In this paper, we introduce a technique called approximate-caching that can reduce such iterative denoising steps for an image generation based on a prompt by reusing intermediate noise states created during a prior image generation for similar prompts. Based on this idea, we present an end to end text-to-image system, Nirvana, that uses the approximate-caching with a novel cache management-policy Least Computationally Beneficial and Frequently Used (LCBFU) to provide % GPU compute savings, 19.8% end-to-end latency reduction and 19% dollar savings, on average, on two real production workloads. We further present an extensive characterization of real production text-to-image prompts from the perspective of caching, popularity and reuse of intermediate states in a large production environment.

Reinforced Approximate Exploratory Data Analysis

Exploratory data analytics (EDA) is a sequential decision making process where analysts choose subsequent queries that might lead to some interesting insights based on the previous queries and corresponding results. Data processing systems often execute the queries on samples to produce results with low latency. Different downsampling strategy preserves different statistics of the data and have different magnitude of latency reductions. The optimum choice of sampling strategy often depends on the particular context of the analysis flow and the hidden intent of the analyst. In this paper, we are the first to consider the impact of sampling in interactive data exploration settings as they introduce approximation errors. We propose a Deep Reinforcement Learning (DRL) based framework which can optimize the sample selection in order to keep the analysis and insight generation flow intact. Evaluations with 3 real datasets show that our technique can preserve the original insight generation flow while improving the interaction latency, compared to baseline methods.

Analysis of Distributed Deep Learning in the Cloud

We aim to resolve this problem by introducing a comprehensive distributed deep learning (DDL) profiler, which can determine the various execution "stalls" that DDL suffers from while running on a public cloud. We have implemented the profiler by extending prior work to additionally estimate two types of communication stalls - interconnect and network stalls. We train popular DNN models using the profiler to characterize various AWS GPU instances and list their advantages and shortcomings for users to make an informed decision. We observe that the more expensive GPU instances may not be the most performant for all DNN models and AWS may sub-optimally allocate hardware interconnect resources. Specifically, the intra-machine interconnect can introduce communication overheads up to 90% of DNN training time and network-connected instances can suffer from up to 5x slowdown compared to training on a single instance. Further, we model the impact of DNN macroscopic features such as the number of layers and the number of gradients on communication stalls. Finally, we propose a measurement-based recommendation model for users to lower their public cloud monetary costs for DDL, given a time budget.

Electra: Conditional Generative Model based Predicate-Aware Query Approximation

The goal of Approximate Query Processing (AQP) is to provide very fast but "accurate enough" results for costly aggregate queries thereby improving user experience in interactive exploration of large datasets. Recently proposed Machine-Learning based AQP techniques can provide very low latency as query execution only involves model inference as compared to traditional query processing on database clusters. However, with increase in the number of filtering predicates(WHERE clauses), the approximation error significantly increases for these methods. Analysts often use queries with a large number of predicates for insights discovery. Thus, maintaining low approximation error is important to prevent analysts from drawing misleading conclusions. In this paper, we propose ELECTRA, a predicate-aware AQP system that can answer analytics-style queries with a large number of predicates with much smaller approximation errors. ELECTRA uses a conditional generative model that learns the conditional distribution of the data and at runtime generates a small (~1000 rows) but representative sample, on which the query is executed to compute the approximate result. Our evaluations with four different baselines on three real-world datasets show that ELECTRA provides lower AQP error for large number of predicates compared to baselines.

ApproxDet: Content and Contention-Aware Approximate Object Detection for Mobiles

Advanced video analytic systems, including scene classification and object detection, have seen widespread success in various domains such as smart cities and autonomous transportation. With an ever-growing number of powerful client devices, there is incentive to move these heavy video analytics workloads from the cloud to mobile devices to achieve low latency and real-time processing and to preserve user privacy. However, most video analytic systems are heavyweight and are trained offline with some pre-defined latency or accuracy requirements. This makes them unable to adapt at runtime in the face of three types of dynamism -- the input video characteristics change, the amount of compute resources available on the node changes due to co-located applications, and the user's latency-accuracy requirements change. In this paper we introduce ApproxDet, an adaptive video object detection framework for mobile devices to meet accuracy-latency requirements in the face of changing content and resource contention scenarios. To achieve this, we introduce a multi-branch object detection kernel (layered on Faster R-CNN), which incorporates a data-driven modeling approach on the performance metrics, and a latency SLA-driven scheduler to pick the best execution branch at runtime. We couple this kernel with approximable video object tracking algorithms to create an end-to-end video object detection system. We evaluate ApproxDet on a large benchmark video dataset and compare quantitatively to AdaScale and YOLOv3. We find that ApproxDet is able to adapt to a wide variety of contention and content characteristics and outshines all baselines, e.g., it achieves 52% lower latency and 11.1% higher accuracy over YOLOv3.

ApproxNet: Content and Contention Aware Video Analytics System for the Edge

Videos take lot of time to transport over the network, hence running analytics on live video at the edge devices, right where it was captured has become an important system driver. However these edge devices, e.g., IoT devices, surveillance cameras, AR/VR gadgets are resource constrained. This makes it impossible to run state-of-the-art heavy Deep Neural Networks (DNNs) on them and yet provide low and stable latency under various circumstances, such as, changes in the resource availability on the device, the content characteristics, or requirements from the user. In this paper we introduce ApproxNet, a video analytics system for the edge. It enables novel dynamic approximation techniques to achieve desired inference latency and accuracy trade-off under different system conditions and resource contentions, variations in the complexity of the video contents and user requirements. It achieves this by enabling two approximation knobs within a single DNN model, rather than creating and maintaining an ensemble of models (such as in MCDNN [Mobisys-16]). Ensemble models run into memory issues on the lightweight devices and incur large switching penalties among the models in response to runtime changes. We show that ApproxNet can adapt seamlessly at runtime to video content changes and changes in system dynamics to provide low and stable latency for object detection on a video stream. We compare the accuracy and the latency to ResNet [2015], MCDNN, and MobileNets [Google-2017].

DeepPlace: Learning to Place Applications in Multi-Tenant Clusters

Large multi-tenant production clusters often have to handle a variety of jobs and applications with a variety of complex resource usage characteristics. It is non-trivial and non-optimal to manually create placement rules for scheduling that would decide which applications should co-locate. In this paper, we present DeepPlace, a scheduler that learns to exploits various temporal resource usage patterns of applications using Deep Reinforcement Learning (Deep RL) to reduce resource competition across jobs running in the same machine while at the same time optimizing for overall cluster utilization.

Linear Antenna Array with Suppressed Sidelobe and Sideband Levels using Time Modulation

In this paper, the goal is to achieve an ultra low sidelobe level (SLL) and sideband levels (SBL) of a time modulated linear antenna array. The approach followed here is not to give fixed level of excitation to the elements of an array, but to change it dynamically with time. The excitation levels of the different array elements over time are varied to get the low sidelobe and sideband levels. The mathematics of getting the SLL and SBL furnished in detail and simulation is done using the mathematical results. The excitation pattern over time is optimized using Genetic Algorithm (GA). Since, the amplitudes of the excitations of the elements are varied within a finite limit, results show it gives better sidelobe and sideband suppression compared to previous time modulated arrays with uniform amplitude excitations.