Papers by Siddhant Garg

13 papers
Knowledge Transfer from Answer Ranking to Answer Generation (2022.emnlp-main)

Copied to clipboard

Challenge: Recent studies show that Question Answering (QA) based on Answer Sentence Selection (AS2) can be improved by generating an improved answer from the top-k ranked answer sentences.
Approach: They propose to train a GenQA model by transferring knowledge from a trained AS2 model . they use top ranked candidate as the generation target and next k top rated candidates as context .
Outcome: The proposed model outperforms existing models on public and industrial datasets.
Pre-training Transformer Models with Sentence-Level Objectives for Answer Sentence Selection (2022.emnlp-main)

Copied to clipboard

Challenge: Existing models for answer sentence selection (AS2) are not yet available for AS2 .
Approach: They propose to incorporate paragraph-level semantics within and across documents to improve transformers for AS2 . they propose to use a dataset to predict whether two sentences are extracted from the same paragraph .
Outcome: The proposed model outperforms baseline models on public and industrial datasets on three public and one industrial dataset.
ProMISe: A Proactive Multi-turn Dialogue Dataset for Information-seeking Intent Resolution (2024.findings-eacl)

Copied to clipboard

Challenge: Work done during internship at Amazon Alexa AI.
Approach: They propose to use iterative suggested question-answering conversation to improve the trade-off between satisfaction of the user’s intent and keeping the information exchange natural.
Outcome: The proposed proposed question-answering conversation improves the satisfaction of the user’s intent while keeping the information exchange natural and cognitive load of the interaction minimal on the users.
Measuring Retrieval Complexity in Question Answering Systems (2024.findings-acl)

Copied to clipboard

Challenge: a new metric, retrieval complexity (RC), measures the difficulty of answering questions.
Approach: They propose a retrieval complexity metric conditioned on the completeness of retrieved documents . they propose an unsupervised pipeline to measure RC given an arbitrary retrieval system .
Outcome: The proposed pipeline measures RC more accurately than alternative estimators on six challenging QA benchmarks.
Towards Improved Multi-Source Attribution for Long-Form Answer Generation (2024.naacl-long)

Copied to clipboard

Challenge: Current LLMs struggle with attribution for long-form answers which require reasoning over multiple evidence sources.
Approach: They propose to improve attribution capability of large language models for long-form answer generation to multiple sources with multiple citations per sentence.
Outcome: The proposed model improves on a wide range of attribution benchmark datasets on PolitiICite, a multi-source attribution dataset based on PolitIcite articles .
Surprisingly Easy Hard-Attention for Sequence to Sequence Learning (D18-1)

Copied to clipboard

Challenge: Existing attention mechanisms are hard and hard, but they are more accurate when trained.
Approach: They propose to use a beam approximation of the joint distribution between attention and output to train sequence to sequence learning.
Outcome: The proposed method is compared to existing attention mechanisms on five translation tasks and shows consistent gains on the same tasks.
Context-Aware Transformer Pre-Training for Answer Sentence Selection (2023.acl-short)

Copied to clipboard

Challenge: Existing approaches to perform Answer Sentence Selection (AS2) using only the candidate sentence are sub-optimal.
Approach: They propose to use pre-trained transformers to perform contextual AS2 fine-tuning . they propose to apply pre-training objectives to local contextual AS2.
Outcome: The proposed methods improve baseline AS2 accuracy by up to 8% on some datasets.
Learning Answer Generation using Supervision from Automatic Question Answering Evaluators (2023.acl-long)

Copied to clipboard

Challenge: Recent studies show sentence-level extractive QA is outperformed by Generation-based QA (GenQA) models.
Approach: They propose a training paradigm for GenQA using automatic QA evaluation models . they augment training data with answers generated by the GenQA model and labelled by GAVA .
Outcome: The proposed training paradigm improves answering accuracy over existing models.
Will this Question be Answered? Question Filtering via Answer Model Distillation for Efficient Question Answering (2021.emnlp-main)

Copied to clipboard

Challenge: Existing methods to improve QA efficiency do not take specific answers into account.
Approach: They propose a transformer-based approach to improve QA efficiency by filtering out questions that will not be answered by the system.
Outcome: The proposed model can approximate the Precision/Recall curves of the target QA system.
Memory-QA: Answering Recall Questions Based on Multimodal Memories (2025.emnlp-main)

Copied to clipboard

Challenge: Memory-QA is a real-world task that involves answering recall questions about visual content from previously stored multimodal memories.
Approach: They propose a memory-QA task that involves answering recall questions about visual content from previously stored multimodal memories.
Outcome: The proposed solution improves memory recording, compression, storage, and search accuracy over state-of-the-art solutions.
BAE: BERT-based Adversarial Examples for Text Classification (2020.emnlp-main)

Copied to clipboard

Challenge: Recent studies have exposed the vulnerability of text classification models to adversarial examples . perturbed versions of the original text are indiscernible by humans and misclassified by the model .
Approach: They propose a black box attack for generating adversarial examples using contextual perturbations from a BERT-masked language model.
Outcome: The proposed attack produces examples with improved grammaticality and semantic coherence compared to previous work.
Beyond Fine-tuning: Few-Sample Sentence Embedding Transfer (2020.aacl-main)

Copied to clipboard

Challenge: Fine-tuning (FT) pre-trained sentence embedding models on small datasets has been shown to have limitations.
Approach: They propose to combine embeddings from a pre-trained model with a simple sentence embeddable model.
Outcome: The proposed approach outperforms FT on small datasets with negligible computational overhead.
Paragraph-based Transformer Pre-training for Multi-Sentence Inference (2022.naacl-main)

Copied to clipboard

Challenge: Recent studies show that pre-trained transformers perform poorly for multi-candidate inference tasks.
Approach: They propose a pre-training objective that models paragraph-level semantics across multiple input sentences.
Outcome: The proposed model outperforms existing models on three AS2 and one fact verification datasets.

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