NLP Service APIs and Models for Efficient Registration of New Clients

by   Sahil Shah, et al.

State-of-the-art NLP inference uses enormous neural architectures and models trained for GPU-months, well beyond the reach of most consumers of NLP. This has led to one-size-fits-all public API-based NLP service models by major AI companies, serving large numbers of clients. Neither (hardware deficient) clients nor (heavily subscribed) servers can afford traditional fine tuning. Many clients own little or no labeled data. We initiate a study of adaptation of centralized NLP services to clients, and present one practical and lightweight approach. Each client uses an unsupervised, corpus-based sketch to register to the service. The server uses an auxiliary network to map the sketch to an abstract vector representation, which then informs the main labeling network. When a new client registers with its sketch, it gets immediate accuracy benefits. We demonstrate the success of the proposed architecture using sentiment labeling, NER, and predictive language modeling


page 1

page 2

page 3

page 4


A Scheme to Verify Services with Unboundedly many Clients using NuSMV

We study model checking of client - server systems, where the servers of...

Inference-Time Personalized Federated Learning

In Federated learning (FL), multiple clients collaborate to learn a mode...

XONN: XNOR-based Oblivious Deep Neural Network Inference

Advancements in deep learning enable cloud servers to provide inference-...

Comfetch: Federated Learning of Large Networks on Memory-Constrained Clients via Sketching

A popular application of federated learning is using many clients to tra...

Learning Private Neural Language Modeling with Attentive Aggregation

Mobile keyboard suggestion is typically regarded as a word-level languag...

Advising Agent for Service-Providing Live-Chat Operators

Call centers, in which human operators attend clients using textual chat...

Data Analysis: Communicating with Offshore Vendors using Instant Messaging Services

The purpose of this study is to find whether the choice of correct analy...