Papers by Tu Vu
Encouraging Paragraph Embeddings to Remember Sentence Identity Improves Classification (P19-1)
Copied to clipboard
| Challenge: | Existing paragraph embedding methods do not capture basic linguistic properties, but their performance is limited. |
| Approach: | They propose a paragraph embedding method that can't tell whether a sentence occurs in a given paragraph. |
| Outcome: | The proposed method outperforms reconstruction-based methods on a semi-supervised dataset and improves on benchmark datasets. |
HiCOT: Improving Neural Topic Models via Optimal Transport and Contrastive Learning (2025.findings-acl)
Copied to clipboard
| Challenge: | Recent advances in neural topic models (NTMs) have improved topic quality but still face challenges: weak document-topic alignment, high inference costs due to large pretrained language models, and limited modeling of hierarchical topic structures. |
| Approach: | They propose a framework that integrates hierarchical clustering and contrastive learning to refine document-topic relationships using compact PLM-based embeddings. |
| Outcome: | The proposed framework improves topic coherence, topic performance, representation quality and computational efficiency over existing NTMs. |
EMO: Embedding Model Distillation via Intra-Model Relation and Optimal Transport Alignments (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for knowledge distillation focus on direct output alignment, neglecting this crucial structural information. |
| Approach: | They propose a framework for knowledge distillation that maps tokens one-to-one and aligns attention matrix patterns using Centered Kernel Alignment. |
| Outcome: | The proposed framework significantly outperforms existing CTKD baselines. |
SPoT: Better Frozen Model Adaptation through Soft Prompt Transfer (2022.acl-long)
Copied to clipboard
| Challenge: | Recent studies show that pre-trained language models can be more efficient when they are larger than they are in their size. |
| Approach: | They propose a prompt-based transfer learning approach called SPoT: Soft Prompt Transfer that learns a soft prompt on one or more source tasks and initializes it for a target task. |
| Outcome: | The proposed approach outperforms Prompt Tuning and MODELTUNING on superGLUE benchmarks while using up to 27,000 fewer task-specific parameters. |
FreshLLMs: Refreshing Large Language Models with Search Engine Augmentation (2024.findings-acl)
Copied to clipboard
Tu Vu, Mohit Iyyer, Xuezhi Wang, Noah Constant, Jerry Wei, Jason Wei, Chris Tar, Yun-Hsuan Sung, Denny Zhou, Quoc Le, Thang Luong
| Challenge: | Modern large language models often "hallucinate" plausible but factually incorrect information, which reduces their trustworthiness especially in settings where accurate and up-to-date information is critical. |
| Approach: | They develop a human evaluation procedure to measure correctness and hallucination and use it to benchmark both closed and open-source LLMs. |
| Outcome: | The proposed method outperforms both competing search engine-augmented prompting methods and commercial systems on search-augmented QA. |
Efficient Model Development through Fine-tuning Transfer (2025.emnlp-main)
Copied to clipboard
| Challenge: | Modern large language models face a major bottleneck: each new version of a pre-trained model requires expensive and repetitive alignment. |
| Approach: | They propose a method that transfers fine-tuning updates across model versions . they extract the diff vector, which is the difference in parameters induced by fine-uning, from a source model and apply it to the base of a different target model. |
| Outcome: | The proposed method reduces training costs while maintaining model performance. |
Sentence Simplification with Memory-Augmented Neural Networks (N18-2)
Copied to clipboard
| Challenge: | Sentence simplification aims to simplify the content and structure of complex sentences . prior work has focused on monolingual machine translation (MT) and tree-based MT (TBMT). |
| Approach: | They adapt an architecture with augmented memory capacities called Neural Semantic Encoders for sentence simplification. |
| Outcome: | The proposed architecture improves on different datasets and improves human judgments. |
GroundingGPT: Language Enhanced Multi-modal Grounding Model (2024.acl-long)
Copied to clipboard
Zhaowei Li, Qi Xu, Dong Zhang, Hang Song, YiQing Cai, Qi Qi, Ran Zhou, Junting Pan, Zefeng Li, Vu Tu, Zhida Huang, Tao Wang
| Challenge: | Existing multi-modal large language models focus on capturing global information while neglecting the fine-grained local information in multimodal inputs. |
| Approach: | They propose an end-to-end language enhanced multi-modal grounding model that performs fine-grained grounding tasks for image, video and audio. |
| Outcome: | The proposed model achieves impressive fine-grained understanding of multi-modal inputs while maintaining or improving its global comprehension capabilities. |
Topic Modeling for Short Texts via Optimal Transport-Based Clustering (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing approaches to topic modeling are based on probabilistic graphical models or non-negative matrix factorization techniques. |
| Approach: | They propose a method that aligns global clusters with topics to discover topics and learn document representations in topic space. |
| Outcome: | The proposed method outperforms state-of-the-art techniques in short-text topic modeling across commonly used metrics. |
STraTA: Self-Training with Task Augmentation for Better Few-shot Learning (2021.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in NLP demonstrate the effectiveness of applying large-scale pre-trained language models to downstream tasks. |
| Approach: | They propose a method that uses task augmentation to fine-tune unlabeled data. |
| Outcome: | The proposed approach improves sample efficiency across 12 few-shot benchmarks. |
O_O-VC: Synthetic Data-Driven One-to-One Alignment for Any-to-Any Voice Conversion (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Traditional voice conversion methods attempt to separate speaker identity and linguistic information into distinct representations, but this method often leads to information loss during training. |
| Approach: | They propose a method that leverages synthetic speech data generated by a pretrained model . synthetic data pairs that share the same linguistic content are used as input-output pairs . |
| Outcome: | The proposed method outperforms state-of-the-art methods in speaker-to-voice conversions. |
Overcoming Catastrophic Forgetting in Zero-Shot Cross-Lingual Generation (2022.emnlp-main)
Copied to clipboard
| Challenge: | generative multilingual models fine-tuned on English forget to generate non-English data when labeled data is only available in English . generative models fine tuned on English fail to generate multilingual summarization tasks when labeling data is available in other languages . |
| Approach: | They propose to use prompt tuning to overcome catastrophic forgetting in a generative task in . they assume a strict setting with no parallel data or machine translation . |
| Outcome: | The proposed method can overcome catastrophic forgetting to enable zero-shot cross-lingual generation. |
Leveraging QA Datasets to Improve Generative Data Augmentation (2022.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in NLP have substantially improved the capability of pretrained language models to generate high-quality text. |
| Approach: | They propose to reformulate data generation as context generation for a given question-answer (QA) pair and leverage QA datasets for training context generators. |
| Outcome: | The proposed approach improves performance for few-shot and zero-shot classification datasets on multiple classification dataset. |
Introducing a Large-Scale Dataset for Vietnamese POS Tagging on Conversational Texts (2020.lrec-1)
Copied to clipboard
| Challenge: | POS taggers are trained on informal texts which contain many informal inputs such as acronyms, abbreviations, out-of-vocabulary words, etc. |
| Approach: | They propose a large-scale human-labeled dataset for the Vietnamese POS tagging task on conversational texts and develop an annotation guideline to manually annotate 16.310K sentences using this guideline. |
| Outcome: | The proposed tagging scheme achieved 93.36% accuracy score and higher than the model with handcrafted features and fine-tuning BERT. |
Dialect-robust Evaluation of Generated Text (2023.acl-long)
Copied to clipboard
Jiao Sun, Thibault Sellam, Elizabeth Clark, Tu Vu, Timothy Dozat, Dan Garrette, Aditya Siddhant, Jacob Eisenstein, Sebastian Gehrmann
| Challenge: | Existing evaluation metrics that are not robust to dialect variation are difficult to measure for many groups of users and can penalize systems for producing text in lower-resource dialects. |
| Approach: | They propose a dialect-robust evaluation metric that produces the same score for system outputs that share the same semantics but are expressed in different dialects. |
| Outcome: | The proposed method significantly improves dialect robustness while preserving the correlation between automated metrics and human ratings. |
ViDeBERTa: A powerful pre-trained language model for Vietnamese (2023.findings-eacl)
Copied to clipboard
| Challenge: | Existing models for Vietnamese that perform well on downstream tasks, such as Question answering, are based on Transformer. |
| Approach: | They propose a pre-trained monolingual Vietnamese model with three versions . they fine-tune and evaluate the model on three important natural language downstream tasks, Part-of-speech tagging, Named-entity recognition, and Question answering. |
| Outcome: | The proposed model outperforms the existing model on three important natural language downstream tasks, Part-of-speech tagging, Named-entity recognition, and Question answering. |
Foundational Autoraters: Taming Large Language Models for Better Automatic Evaluation (2024.emnlp-main)
Copied to clipboard
| Challenge: | evaluating large language models' output is difficult due to the high cost of human evaluation. |
| Approach: | They propose a family of foundational large autorater models that train on over 100 quality assessment tasks. |
| Outcome: | The proposed model outperforms models on 8 of 12 autorater benchmarks on 53 quality assessment tasks. |
Exploring and Predicting Transferability across NLP Tasks (2020.emnlp-main)
Copied to clipboard
Tu Vu, Tong Wang, Tsendsuren Munkhdalai, Alessandro Sordoni, Adam Trischler, Andrew Mattarella-Micke, Subhransu Maji, Mohit Iyyer
| Challenge: | Recent advances in NLP demonstrate the effectiveness of training large-scale language models and transferring them to downstream tasks. |
| Approach: | They conduct an extensive study of the transferability between 33 NLP tasks across three broad classes of problems. |
| Outcome: | The proposed model can improve performance even with low-data source tasks that differ substantially from the target task. |