Yumo Xu

I am an NLP scientist at AWS AI Labs researching and building intelligent language services for enterprises.

My current interests lie in the following areas:

Effective Retrieval-Augmented Generation (RAG) systems,
Accurate faithfulness evaluation and source attribution, and
Reliable autonomous agents with generalizable tool-use capabilities.

Prior to Amazon, I received my PhD in NLP at the University of Edinburgh, advised by Prof. Mirella Lapata. My PhD research was on text summarization, the process of condensing a source text into a shorter version while preserving its salient information.

News

Sep 7, 2023	I will serve as an Area Chair for NLG at LREC-COLING’24.
May 1, 2023	Generative modeling for eXtractive summarization (GenX) was accepted by ACL’23.
Apr 18, 2023	Excited to join AWS AI Labs in Seattle as a scientist.
Mar 31, 2023	Happy to join the Organizing Committee of SIGSUMM.
Jan 21, 2023	Working on extractive summarization? Check out our latest work accepted by ICLR’23.

(Recent) Selected Publications

Summarization

QTSumm: Query-Focused Summarization over Tabular Data

In EMNLP 2023

Abs PDF Code

People primarily consult tables to conduct data analysis or answer specific questions. Text generation systems that can provide accurate table summaries tailored to users’ information needs can facilitate more efficient access to relevant data insights. However, existing table-to-text generation studies primarily focus on converting tabular data into coherent statements, rather than addressing information-seeking purposes. In this paper, we define a new query-focused table summarization task, where text generation models have to perform human-like reasoning and analysis over the given table to generate a tailored summary, and we introduce a new benchmark named QTSumm for this task. QTSumm consists of 5,625 human-annotated query-summary pairs over 2,437 tables on diverse topics. Moreover, we investigate state-of-the-art models (i.e., text generation, table-to-text generation, and large language models) on the QTSumm dataset. Experimental results and manual analysis reveal that our benchmark presents significant challenges in table-to-text generation for future research.
Summarization

Abstractive Summarizers are Excellent Extractive Summarizers

Varab, Daniel, and Xu, Yumo

In ACL 2023

Abs PDF Code

Extractive and abstractive summarization designs have historically been fragmented, limiting the benefits that often arise from compatible model architectures. In this paper, we explore the potential synergies of modeling extractive summarization with an abstractive summarization system and propose three novel inference algorithms using the sequence-to-sequence architecture. We evaluate them on the CNN & Dailymail dataset and show that recent advancements in abstractive system designs enable abstractive systems to not only compete, but even surpass the performance of extractive systems with custom architectures. To our surprise, abstractive systems achieve this without being exposed to extractive oracle summaries and, therefore, for the first time allow a single model to produce both abstractive and extractive summaries. This evidence questions our fundamental understanding of extractive system design, and the necessity for extractive labels while pathing the way for promising research directions in hybrid models.
Summarization

Text Summarization with Oracle Expectation

Xu, Yumo, and Lapata, Mirella

In ICLR 2023

Abs PDF Code

Extractive summarization produces summaries by identifying and concatenating the most important sentences in a document. Since most summarization datasets do not come with gold labels indicating whether document sentences are summary-worthy, different labeling algorithms have been proposed to extrapolate oracle extracts for model training. In this work, we identify two flaws with the widely used greedy labeling approach: it delivers suboptimal and deterministic oracles. To alleviate both issues, we propose a simple yet effective labeling algorithm that creates soft, expectation-based sentence labels. We define a new learning objective for extractive summarization which incorporates learning signals from multiple oracle summaries and prove it is equivalent to estimating the oracle expectation for each document sentence. Without any architectural modifications, the proposed labeling scheme achieves superior performance on a variety of summarization benchmarks across domains and languages, in both supervised and zero-shot settings.
Summarization

Document Summarization with Latent Queries

Xu, Yumo, and Lapata, Mirella

TACL 2022

Abs PDF Code

The availability of large-scale datasets has driven the development of neural models that create summaries from single documents, for generic purposes. When using a summarization system, users often have specific intents with various language realizations, which, depending on the information need, can range from a single keyword to a long narrative composed of multiple questions. Existing summarization systems, however, often either fail to support or act robustly on this query focused summarization task. We introduce LaQSum, the first unified text summarization system that learns Latent Queries from documents for abstractive summarization with any existing query forms. Under a deep generative framework, our system jointly optimizes a latent query model and a conditional language model, allowing users to plug-and-play queries of any type at test time. Despite learning from only generic summarization data and requiring no further optimization for downstream summarization tasks, our system robustly outperforms strong comparison systems across summarization benchmarks with different query types, document settings, and target domains.