In-context retrieval-augmented language models

O Ram, Y Levine, I Dalmedigos, D Muhlgay… - Transactions of the …, 2023 - direct.mit.edu
Abstract Retrieval-Augmented Language Modeling (RALM) methods, which condition a
language model (LM) on relevant documents from a grounding corpus during generation …

Parallel context windows for large language models

N Ratner, Y Levine, Y Belinkov, O Ram, I Magar… - arxiv preprint arxiv …, 2022 - arxiv.org
When applied to processing long text, Large Language Models (LLMs) are limited by their
context window. Existing efforts to address this limitation involve training specialized …

Improving zero-shot reader by reducing distractions from irrelevant documents in open-domain question answering

S Cho, J Seo, S Jeong, JC Park - arxiv preprint arxiv:2310.17490, 2023 - arxiv.org
Large language models (LLMs) enable zero-shot approaches in open-domain question
answering (ODQA), yet with limited advancements as the reader is compared to the …

Chain-of-Rewrite: Aligning Question and Documents for Open-Domain Question Answering

C **n, Y Lu, H Lin, S Zhou, H Zhu… - Findings of the …, 2024 - aclanthology.org
Despite the advancements made with the retrieve-then-read pipeline on open-domain
question answering task, current methods still face challenges stemming from term …

[HTML][HTML] Evaluating LLMs on document-based QA: Exact answer selection and numerical extraction using CogTale dataset

Z Rasool, S Kurniawan, S Balugo, S Barnett… - Natural Language …, 2024 - Elsevier
Abstract Document-based Question-Answering (QA) tasks are crucial for precise information
retrieval. While some existing work focus on evaluating large language model's (LLMs) …

Toward Robust RALMs: Revealing the Impact of Imperfect Retrieval on Retrieval-Augmented Language Models

SI Park, JY Lee - Transactions of the Association for Computational …, 2024 - direct.mit.edu
Abstract Retrieval Augmented Language Models (RALMs) have gained significant attention
for their ability to generate accurate answers and improve efficiency. However, RALMs are …

Towards a small language model powered chain‐of‐reasoning for open‐domain question answering

J Roh, M Kim, K Bae - ETRI Journal, 2024 - Wiley Online Library
We focus on open‐domain question‐answering tasks that involve a chain‐of‐reasoning,
which are primarily implemented using large language models. With an emphasis on cost …