Papers by Simran Arora

3 papers
Reasoning over Public and Private Data in Retrieval-Based Systems (2023.tacl-1)

Copied to clipboard

Challenge: Existing retrieval systems assume relevant corpora are fully (e.g., publicly) accessible, but users are often unwilling to expose their private data to entities hosting public data.
Approach: They propose a split iterative retrieval problem involving iterating retrieval over multiple privacy scopes and propose 'concurrentQA' benchmark to test this problem.
Outcome: The proposed method improves on the existing retrieval methods but still suffers performance degradations when applied to a dataset from a public and private distribution.
Metadata Shaping: A Simple Approach for Knowledge-Enhanced Language Models (2022.findings-acl)

Copied to clipboard

Challenge: Existing methods to capture entity knowledge with factual knowledge are limited . despite its simplicity, metadata shaping is quite effective .
Approach: They propose a method which inserts substrings corresponding to readily available entity metadata into examples at train and inference time based on mutual information.
Outcome: The proposed method exceeds the baseline model by 4.3 F1 points and achieves state-of-the-art results.
Contextual Embeddings: When Are They Worth It? (2020.acl-main)

Copied to clipboard

Challenge: In recent years, rich contextual embeddings have enabled rapid progress on benchmarks like GLUE, but require significant computational resources during pretraining and during downstream task training and inference.
Approach: They empirically compare contextual embeddings with classic pretrained embedders and a random word embeddable with a simple baseline.
Outcome: The proposed models perform within 5 to 10% accuracy on industry-scale data.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations