Bryan Li
YOU?
Author Swipe
View article: Digital twins for accurate prediction beyond routine operation
Digital twins for accurate prediction beyond routine operation Open
View article: Leveraging Domain Knowledge at Inference Time for LLM Translation: Retrieval versus Generation
Leveraging Domain Knowledge at Inference Time for LLM Translation: Retrieval versus Generation Open
View article: Multilingual Retrieval Augmented Generation for Culturally-Sensitive Tasks: A Benchmark for Cross-lingual Robustness
Multilingual Retrieval Augmented Generation for Culturally-Sensitive Tasks: A Benchmark for Cross-lingual Robustness Open
View article: Multilingual Retrieval Augmented Generation for Culturally-Sensitive Tasks: A Benchmark for Cross-lingual Robustness
Multilingual Retrieval Augmented Generation for Culturally-Sensitive Tasks: A Benchmark for Cross-lingual Robustness Open
The paradigm of retrieval-augmented generated (RAG) helps mitigate hallucinations of large language models (LLMs). However, RAG also introduces biases contained within the retrieved documents. These biases can be amplified in scenarios whi…
View article: Uncovering Differences in Persuasive Language in Russian versus English Wikipedia
Uncovering Differences in Persuasive Language in Russian versus English Wikipedia Open
We study how differences in persuasive language across Wikipedia articles, written in either English and Russian, can uncover each culture's distinct perspective on different subjects. We develop a large language model (LLM) powered system…
View article: Eliciting Better Multilingual Structured Reasoning from LLMs through Code
Eliciting Better Multilingual Structured Reasoning from LLMs through Code Open
The development of large language models (LLM) has shown progress on reasoning, though studies have largely considered either English or simple reasoning tasks. To address this, we introduce a multilingual structured reasoning and explanat…
View article: Large Language Models as Sous Chefs: Revising Recipes with GPT-3
Large Language Models as Sous Chefs: Revising Recipes with GPT-3 Open
With their remarkably improved text generation and prompting capabilities, large language models can adapt existing written information into forms that are easier to use and understand. In our work, we focus on recipes as an example of com…
View article: This Land is {Your, My} Land: Evaluating Geopolitical Biases in Language Models
This Land is {Your, My} Land: Evaluating Geopolitical Biases in Language Models Open
Do the Spratly Islands belong to China, the Philippines, or Vietnam? A pretrained large language model (LLM) may answer differently if asked in the languages of each claimant country: Chinese, Tagalog, or Vietnamese. This contrasts with a …
View article: PAXQA: Generating Cross-lingual Question Answering Examples at Training Scale
PAXQA: Generating Cross-lingual Question Answering Examples at Training Scale Open
Existing question answering (QA) systems owe much of their success to large, high-quality training data. Such annotation efforts are costly, and the difficulty compounds in the cross-lingual setting. Therefore, prior cross-lingual QA work …
View article: Multilingual Bidirectional Unsupervised Translation through Multilingual Finetuning and Back-Translation
Multilingual Bidirectional Unsupervised Translation through Multilingual Finetuning and Back-Translation Open
We propose a two-stage approach for training a single NMT model to translate unseen languages both to and from English. For the first stage, we initialize an encoder-decoder model to pretrained XLM-R and RoBERTa weights, then perform multi…
View article: Proceedings of the The Sixth Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2023)
Proceedings of the The Sixth Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2023) Open
In many humanitarian scenarios, translation into severely low resource languages often does not require a universal translation engine, but a dedicated text-specific translation engine.For example, healthcare records, hygienic procedures, …
View article: PAXQA: Generating Cross-lingual Question Answering Examples at Training Scale
PAXQA: Generating Cross-lingual Question Answering Examples at Training Scale Open
Existing question answering (QA) systems owe much of their success to large, high-quality training data. Such annotation efforts are costly, and the difficulty compounds in the cross-lingual setting. Therefore, prior cross-lingual QA work …
View article: Enhancing Human Summaries for Question-Answer Generation in Education
Enhancing Human Summaries for Question-Answer Generation in Education Open
Hannah Gonzalez, Liam Dugan, Eleni Miltsakaki, Zhiqi Cui, Jiaxuan Ren, Bryan Li, Shriyash Upadhyay, Etan Ginsberg, Chris Callison-Burch. Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2…
View article: Word Alignment in the Era of Deep Learning: A Tutorial
Word Alignment in the Era of Deep Learning: A Tutorial Open
The word alignment task, despite its prominence in the era of statistical machine translation (SMT), is niche and under-explored today. In this two-part tutorial, we argue for the continued relevance for word alignment. The first part prov…
View article: CREATIVESUMM: Shared Task on Automatic Summarization for Creative Writing
CREATIVESUMM: Shared Task on Automatic Summarization for Creative Writing Open
This paper introduces the shared task of summarizing documents in several creative domains, namely literary texts, movie scripts, and television scripts. Summarizing these creative documents requires making complex literary interpretations…
View article: Bidirectional Language Models Are Also Few-shot Learners
Bidirectional Language Models Are Also Few-shot Learners Open
Large language models such as GPT-3 (Brown et al., 2020) can perform arbitrary tasks without undergoing fine-tuning after being prompted with only a few labeled examples. An arbitrary task can be reformulated as a natural language prompt, …
View article: Multilingual Bidirectional Unsupervised Translation Through Multilingual Finetuning and Back-Translation
Multilingual Bidirectional Unsupervised Translation Through Multilingual Finetuning and Back-Translation Open
We propose a two-stage approach for training a single NMT model to translate unseen languages both to and from English. For the first stage, we initialize an encoder-decoder model to pretrained XLM-R and RoBERTa weights, then perform multi…
View article: $\rm{C {\small IS}}^2$: A Simplified Commonsense Inference Evaluation for Story Prose
$\rm{C {\small IS}}^2$: A Simplified Commonsense Inference Evaluation for Story Prose Open
Transformers have been showing near-human performance on a variety of tasks, but they are not without their limitations. We discuss the issue of conflating results of transformers that are instructed to do multiple tasks simultaneously. In…
View article: CIS²: A Simplified Commonsense Inference Evaluation for Story Prose
CIS²: A Simplified Commonsense Inference Evaluation for Story Prose Open
Transformers have been showing near-human performance on a variety of tasks, but they are not without their limitations. We discuss the issue of conflating results of transformers that are instructed to do multiple tasks simultaneously. In…
View article: CCDC 2099928: Experimental Crystal Structure Determination
CCDC 2099928: Experimental Crystal Structure Determination Open
An entry from the Cambridge Structural Database, the world’s repository for small molecule crystal structures. The entry contains experimental data from a crystal diffraction study. The deposited dataset for this entry is freely available …
View article: Exploring Content Selection in Summarization of Novel Chapters
Exploring Content Selection in Summarization of Novel Chapters Open
We present a new summarization task, generating summaries of novel chapters using summary/chapter pairs from online study guides. This is a harder task than the news summarization task, given the chapter length as well as the extreme parap…
View article: Cantonese Automatic Speech Recognition Using Transfer Learning from Mandarin
Cantonese Automatic Speech Recognition Using Transfer Learning from Mandarin Open
We propose a system to develop a basic automatic speech recognizer(ASR) for Cantonese, a low-resource language, through transfer learning of Mandarin, a high-resource language. We take a time-delayed neural network trained on Mandarin, and…