Papers by Yuqing Sun

5 papers
End-to-end Adversarial Sample Generation for Data Augmentation (2023.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for generating adversarial samples have deceived many neural inference models, such as text classification and machine translation.
Approach: They propose an adversarial sample generator that consists of a conditioned paraphrasing model and a condition generator and introduce a pretrained discriminator to help the adversarial sample generator adapt to the data characteristics.
Outcome: The proposed approach improves the performance of the trained model on several tasks and is robust for various attacking techniques.
Rectified Sparse Attention for Efficient Long-Sequence Generation (2026.findings-acl)

Copied to clipboard

Challenge: Recent sparse decoding methods improve efficiency but suffer from KV cache misalignment, resulting in performance degradation.
Approach: They propose a method that combines block-sparse attention with periodic dense rectification to bound error accumulation and preserve alignment with the pretraining distribution.
Outcome: Experiments on math reasoning, language modeling, and retrieval tasks show that ReSA achieves near-lossless generation quality with significantly improved efficiency.
Inducing Argument Facets for Faithful Opinion Summarization (2025.findings-emnlp)

Copied to clipboard

Challenge: Faithful opinion summarization task involves generating a summary that covers the majority and minority opinions in documents.
Approach: They propose a facets-guided opinion summarization method that induces facets and partitions documents into multiple facet-specific sets.
Outcome: The proposed method outperforms state-of-the-art methods and multiple LLMs on two representative datasets and shows it can be used in specialty domains.
Applying Contrastive Learning to Code Vulnerability Type Classification (2024.emnlp-main)

Copied to clipboard

Challenge: Recent approaches to classification of vulnerabilities ignore their relationships and treat each class in isolation, resulting in non-scalable code vector representations.
Approach: They propose a hierarchical contrastive learning framework to bring vector representations of related CWEs closer together and use max-pooling to enable the model to handle longer vulnerability code inputs.
Outcome: The proposed framework outperforms state-of-the-art methods by 2.97%-17.90% on accuracy and 0.98%-22.27% on weighted-F1 with even better performance on higher-quality datasets.
CoLLiE: Collaborative Training of Large Language Models in an Efficient Way (2023.emnlp-demo)

Copied to clipboard

Challenge: Large language models (LLMs) are increasingly pivotal in a wide range of tasks . however, the resources required for training these models necessitate efficient solutions .
Approach: They propose a library that facilitates collaborative training of large language models . they use 3D parallelism, parameter-efficient fine-tuning methods and optimizers .
Outcome: The proposed library has proven superior training efficiency in comparison with prevalent solutions in pre-training and fine-tuning scenarios.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations