Papers by William Chen
ParaCrawl: Web-Scale Acquisition of Parallel Corpora (2020.acl-main)
Copied to clipboard
Marta Bañón, Pinzhen Chen, Barry Haddow, Kenneth Heafield, Hieu Hoang, Miquel Esplà-Gomis, Mikel L. Forcada, Amir Kamran, Faheem Kirefu, Philipp Koehn, Sergio Ortiz Rojas, Leopoldo Pla Sempere, Gema Ramírez-Sánchez, Elsa Sarrías, Marek Strelec, Brian Thompson, William Waites, Dion Wiggins, Jaume Zaragoza
| Challenge: | We describe methods to create the largest publicly available parallel corpora by crawling the web . parallel corpus is essential for building highquality machine translation systems . |
| Approach: | They describe methods to create largest publicly available parallel corpora by crawling web sites . they empirically compare alternative methods and publish benchmark data sets . |
| Outcome: | The proposed methods improve state-of-the-art results on common benchmarks, the authors show . the pipeline has been tested on Russian, Sinhala, Nepali, Tagalog, Swahili, and Somali . |
EDIS: Entity-Driven Image Search over Multimodal Web Content (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing image retrieval methods require large datasets and a large candidate set. |
| Approach: | They propose a news-domain dataset for cross-modal image search with 1 million web images . they propose combining multimodal image-text pairs with a million candidates . |
| Outcome: | The proposed dataset challenges state-of-the-art methods with dense entities and the large-scale candidate set. |
A Unified View of Evaluation Metrics for Structured Prediction (2023.emnlp-main)
Copied to clipboard
| Challenge: | a framework that unifies evaluation metrics for structured prediction tasks is presented . metric design decisions are motivated by specific characteristics of tasks, and we suggest modifications to existing metrics to meet those motivations. |
| Approach: | They propose a framework that unifies a variety of evaluation metrics for different structured prediction tasks. |
| Outcome: | The proposed framework can be used to create new metrics based on the output structure of a number of tasks. |
MuRAG: Multimodal Retrieval-Augmented Generator for Open Question Answering over Images and Text (2022.emnlp-main)
Copied to clipboard
| Challenge: | Pre-trained language models store a massive amount of world knowledge implicitly in their parameters, but large models often fail to encode information about rare entities and events. |
| Approach: | They propose a retrieval-augmented model which accesses an external non-parametric memory to augment language generation. |
| Outcome: | The proposed model outperforms existing models by 10-20% absolute on two datasets and under distractor and full-wiki settings. |
Design Challenges for a Multi-Perspective Search Engine (2022.findings-naacl)
Copied to clipboard
| Challenge: | a document retrieval system fails to deliver diverse and direct responses to controversial questions . classical document retrievals provide a ranked list of references to relevant but not necessarily trustworthy web documents . |
| Approach: | They propose a perspective-oriented document retrieval paradigm to address these challenges . they propose sponses with different perspectives within topically-related web documents . |
| Outcome: | The proposed system is based on a user survey and a prototype . it will be used to assess the utility and understanding of the system . |
On the Risk of Misinformation Pollution with Large Language Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | a recent study demonstrates that large language models can be misused for generating credible-sounding misinformation . however, the ability to produce credible text raises concerns regarding their potential misuse . |
| Approach: | They propose three defense strategies to mitigate misinformation generated by Large Language Models . they propose a threat model and simulate potential misuse scenarios . |
| Outcome: | The proposed defense strategies have shown promising results, albeit with costs. |
MultiMUC: Multilingual Template Filling on MUC-4 (2024.eacl-long)
Copied to clipboard
William Gantt, Shabnam Behzad, Hannah An, Yunmo Chen, Aaron White, Benjamin Van Durme, Mahsa Yarmohammadi
| Challenge: | We present multilingual parallel template filling datasets for MUCs . systems were required to extract one template per incident, containing details about perpetrators, victims, weapons used . |
| Approach: | They introduce MultiMUC, the first multilingual parallel corpus for template filling . they obtain automatic translations from a strong multilingual machine translation system . |
| Outcome: | The proposed dataset includes translations of the classic MUC-4 template filling benchmark into Arabic, Chinese, Farsi, Korean, and Russian. |
Evaluating Self-Supervised Speech Representations for Indigenous American Languages (2024.lrec-main)
Copied to clipboard
| Challenge: | a recent study focused on the use of self-supervised learning to learn speech representations for indigenous languages . aaron e. scott: the vast linguistic diversity represented by indigenous languages remains unexplored . by expanding the scope of language processing to include indigenous languages, we can foster linguistic inclusivity, he says . |
| Approach: | They benchmark the efficacy of large-scale self-supervised learning models on indigenous American languages. |
| Outcome: | The proposed model can generalize to real-world data, showing strong performance . evaluators found that the model performed better than monolingual models on indigenous languages . |
On Event Individuation for Document-Level Information Extraction (2023.findings-emnlp)
Copied to clipboard
| Challenge: | a bomb exploded in a restaurant in Lima, and a second device was deactivated by the police . |
| Approach: | They argue that the task demands definitive answers to thorny questions of *event individuation* they argue that even human experts disagree on the task . |
| Outcome: | The proposed task demands definitive answers to thorny questions of *event individuation* . the proposed task also raises concerns about the usefulness of template filling metrics . |
Towards Robust Speech Representation Learning for Thousands of Languages (2024.emnlp-main)
Copied to clipboard
William Chen, Wangyou Zhang, Yifan Peng, Xinjian Li, Jinchuan Tian, Jiatong Shi, Xuankai Chang, Soumi Maiti, Karen Livescu, Shinji Watanabe
| Challenge: | XEUS is a cross-lingual encoder for universal speech that can be trained on 1 million hours of data across 4057 languages. |
| Approach: | They propose a Cross-lingual Encoder for Universal Speech that can be trained on 1 million hours of data across 4057 languages and a newly created corpus of 7400+ hours from 4057 . |
| Outcome: | The proposed model outperforms state-of-the-art models on several benchmarks and outperfies MMS 1B and w2v-BERT 2.0 v2 by 0.8% and 4.4% respectively. |
Empowering Psychotherapy with Large Language Models: Cognitive Distortion Detection through Diagnosis of Thought Prompting (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing systems for mental health support are shallow and heuristic, e.g., analyzing emotions and generating comforting responses. |
| Approach: | They propose to use cognitive distortion detection to perform diagnosis on the patient’s speech via three stages: subjectivity assessment to separate the facts and the thoughts; contrastive reasoning to elicit the reasoning processes supporting and contradicting the thoughts and schema analysis to summarize the cognition schemas. |
| Outcome: | The proposed system improves on ChatGPT for cognitive distortion detection while generating high-quality rationales approved by human experts. |
Outlier Dimensions Encode Task Specific Knowledge (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have shown that fine-tuning outlier dimensions is detrimental to the representational quality of embeddings. |
| Approach: | They investigate how fine-tuning impacts outlier dimensions by testing their hypothesis that a single outlier dimension can complete downstream tasks with a minimal error rate. |
| Outcome: | The proposed model can encode crucial task-specific knowledge and the value of a representation in a single outlier dimension drives downstream model decisions. |
Iterative Document-level Information Extraction via Imitation Learning (2023.eacl-main)
Copied to clipboard
| Challenge: | Documents may feature zero or more instances of a template of any given type, and the task of template extraction entails identifying the templates in a document and extracting each template’s slot values. |
| Approach: | They propose to use iterative extraction to extract complex relations, i.e., N-tuples representing a mapping from named slots to spans of text within a document. |
| Outcome: | The proposed model leads to state-of-the-art results on two established benchmarks and a strong baseline on the new BETTER Granular task. |
Reasoning with Graphs: Structuring Implicit Knowledge to Enhance LLMs Reasoning (2025.findings-acl)
Copied to clipboard
Haoyu Han, Yaochen Xie, Hui Liu, Xianfeng Tang, Sreyashi Nag, William Headden, Yang Li, Chen Luo, Shuiwang Ji, Qi He, Jiliang Tang
| Challenge: | Large language models (LLMs) have demonstrated remarkable success across a wide range of tasks, however, they still face challenges in reasoning tasks that require understanding and inferring relationships between distinct pieces of information within text sequences. |
| Approach: | They propose to construct explicit graphs from context and leverage them to enhance LLM reasoning performance on reasoning tasks. |
| Outcome: | Extensive experiments show that the proposed method improves both logical reasoning and multi-hop question answering tasks. |
On the Evaluation of Speech Foundation Models for Spoken Language Understanding (2024.findings-acl)
Copied to clipboard
Siddhant Arora, Ankita Pasad, Chung-Ming Chien, Jionghao Han, Roshan Sharma, Jee-weon Jung, Hira Dhamyal, William Chen, Suwon Shon, Hung-yi Lee, Karen Livescu, Shinji Watanabe
| Challenge: | Spoken language understanding evaluation (SLUE) benchmarks are used to benchmark complex spoken language understanding tasks on natural speech. |
| Approach: | They propose a set of benchmark tasks to evaluate spoken language understanding on natural speech . they use pre-trained speech foundation models to evaluate the utility of different SFMs . |
| Outcome: | The proposed framework outperforms pre-trained speech foundation models on natural speech . the proposed framework also outperformed self-supervised SFMs on the sequence generation tasks . |
Breaking Bad Tokens: Detoxification of LLMs Using Sparse Autoencoders (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) are ubiquitous in user-facing applications, yet they still generate undesirable toxic outputs, including profanity, vulgarity, and derogatory remarks. |
| Approach: | They leverage sparse autoencoders to identify toxicity-related directions in residual stream of large language models and perform targeted activation steering using the corresponding decoder vectors. |
| Outcome: | The proposed models surpass baselines in reducing toxicity by up to 20%, though fluency can degrade noticeably on GPT-2 Small and Gemma-2-2B. |