updated abstract

README.md

@@ -1,5 +1,8 @@
 # Abstract
-Efficiently optimizing multi-model inference pipelines for fast, accurate, and cost-effective inference is a crucial challenge in ML production systems, given their tight end-to-end latency requirements. To simplify the exploration of the vast and intricate trade-off space of accuracy and cost in inference pipelines, providers frequently opt to consider one of them. However, the challenge lies in reconciling accuracy and cost trade-offs. To address this challenge and propose a solution to efficiently manage model variants in inference pipelines, we present IPA, an online deep-learning Inference Pipeline Adaptation system that efficiently leverages model variants for each deep learning task. Model variants are different versions of pre-trained models for the same deep learning task with variations in resource requirements, latency, and accuracy. IPA dynamically configures batch size, replication, and model variants to optimize accuracy, minimize costs, and meet user-defined latency SLAs using Integer Programming. It supports multi-objective settings for achieving different trade-offs between accuracy and cost objectives while remaining adaptable to varying workloads and dynamic traffic patterns. Extensive experiments on a Kubernetes implementation with five real-world inference pipelines demonstrate that IPA improves normalized accuracy by up to 35% with a minimal cost increase of less than 5%.
+Efficiently optimizing multi-model inference pipelines for fast, accurate, and cost-effective inference is a crucial challenge in machine learning production systems, given their tight end-to-end latency requirements. To simplify the exploration of the vast and intricate trade-off space of accuracy and cost in inference pipelines, providers frequently opt to consider one of them. However, the challenge lies in reconciling accuracy and cost trade-offs.
+
+To address this challenge and propose a solution to efficiently manage model variants in inference pipelines, we present IPA, an online deep learning **I**nference **P**ipeline **A**daptation system that efficiently leverages model variants for each deep learning task. Model variants are different versions of pre-trained models for the same deep learning task with variations in resource requirements, latency, and accuracy. IPA dynamically configures batch size, replication, and model variants to optimize accuracy, minimize costs, and meet user-defined latency Service Level Agreements (SLA) using Integer Programming. It supports multi-objective settings for achieving different trade-offs between accuracy and cost objectives while remaining adaptable to varying workloads and dynamic traffic patterns. This adaptability to a wider variety of configurations enables IPA to achieve better trade-offs between the cost and accuracy objectives.
+Extensive experiments on a Kubernetes~ implementation with five real-world inference pipelines demonstrate that IPA improves end-to-end accuracy by up to 21\% with a minimal cost increase.
 
 # 1 Project Setup Steps (Automated Artifact Evaluation - For JSys Reviewers)
 1. Go to the [infrastructure](/infrastructure/README.md) for the guide to set up the K8S cluster and related depandancies, the complete installtion takes ~30 minutes.