Papers by Yuepei Li

6 papers
GenDecider: Integrating “None of the Candidates” Judgments in Zero-Shot Entity Linking Re-ranking (2024.naacl-short)

Copied to clipboard

Challenge: Existing methods for zero-shot reranking assume the correct entity is always among the retrieved candidates.
Approach: They propose a novel re-ranking approach for Zero-Shot Entity Linking . they use the Llama model to detect scenarios where the correct entity is not retrieved .
Outcome: The proposed approach significantly improves disambiguation and accuracy on the ZESHEL dataset.
Distantly Supervised Named Entity Recognition via Confidence-Based Multi-Class Positive and Unlabeled Learning (2022.acl-long)

Copied to clipboard

Challenge: Existing methods for named entity recognition suffer from incomplete annotations due to incompleteness of external knowledge bases.
Approach: They propose a method to solve the named entity recognition problem under distant supervision using dictionaries and knowledge bases.
Outcome: The proposed method outperforms existing methods on two benchmark datasets labeled by various knowledge bases.
Investigating Context Faithfulness in Large Language Models: The Roles of Memory Strength and Evidence Style (2025.findings-acl)

Copied to clipboard

Challenge: Retrieval-augmented generation improves Large Language Models (LLMs) by integrating external information into the response generation process.
Approach: They investigate the impact of memory strength and evidence presentation on LLMs’ receptiveness to external evidence by measuring the divergence in LLM responses to different paraphrases of the same question.
Outcome: The proposed method improves Large Language Models (LLMs) by integrating external information into the response generation process.
Towards a More Generalized Approach in Open Relation Extraction (2025.acl-long)

Copied to clipboard

Challenge: Existing OpenRE methods assume unlabeled data is a mixture of known and novel instances.
Approach: They propose a generalized OpenRE setting that considers unlabeled data as a mixture of known and novel instances.
Outcome: The proposed framework outperforms baselines in relation classification and clustering on three benchmark datasets.
Improving Unsupervised Relation Extraction by Augmenting Diverse Sentence Pairs (2023.emnlp-main)

Copied to clipboard

Challenge: Recent studies on relation representation learning focus on contrastive learning strategies, but these studies overlook important aspects.
Approach: They propose to use within-sentence pairs augmentation and cross-sentent pairs extraction to increase diversity of positive pairs and strengthen the discriminative power of contrastive learning.
Outcome: The proposed task increases diversity of positive pairs and strengthens discriminative power . it overcomes limitations of traditional Relation Extraction tasks, which require manual annotations .
Re-Examine Distantly Supervised NER: A New Benchmark and a Simple Approach (2025.coling-main)

Copied to clipboard

Challenge: Existing DS-NER approaches rely on large validation sets and test set for tuning inappropriately.
Approach: They propose a method where training data is annotated using domain dictionaries and test data is analyzed by domain experts.
Outcome: The proposed method reduces the need for labor-intensive manual annotations but rely on large human labeled validation set.

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