Papers by Yuanyuan Lei
Where Reasoning Breaks: Logic-Aware Path Selection by Controlling Logical Connectives in LLMs Reasoning Chains (2026.findings-acl)
Copied to clipboard
| Challenge: | a single transition error can propagate through the entire reasoning chain, leading to unstable performance. |
| Approach: | They propose a framework that intervenes at logical connective junctions to improve LLMs' reasoning. |
| Outcome: | The proposed framework achieves favorable accuracy–efficiency trade-off compared to global inference time scaling methods like beam search and self-consistency. |
Boosting Logical Fallacy Reasoning in LLMs via Logical Structure Tree (2024.emnlp-main)
Copied to clipboard
| Challenge: | Logical fallacy is the use of invalid or flawed reasoning in the construction of a statement. |
| Approach: | They propose to build a logical structure tree to represent hierarchical logic flow among relation connectives and their arguments in a statement. |
| Outcome: | The proposed model significantly improves accuracy and recall for fallacy detection and fallacy classification. |
MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing (2026.acl-industry)
Copied to clipboard
Junbo Niu, Zheng Liu, Zhuangcheng Gu, Bin Wang, Linke Ouyang, Zhiyuan Zhao, Tao Chu, Tianyao He, Fan Wu, Qintong Zhang, Zhenjiang Jin, Guang Liang, Rui Zhang, Wenzheng Zhang, Yuan Qu, Zhifei Ren, Yuefeng Sun, Zirui Tang, Boyu Niu, Yuanhong Zheng, Dongsheng Ma, Ziyang Miao, Hejun Dong, Siyi Qian, Junyuan Zhang, Fangdong Wang, Jingzhou Chen, Xiaomeng Zhao, Liqun Wei, Wei Li, Shasha Wang, RuiLiang Xu, Yuanyuan Cao, Lu Chen, Qianqian Wu, Huaiyu Gu, Lindong Lu, Dechen Lin, null Shenguanlin, Xuanhe Zhou, Linfeng Zhang, Yuhang Zang, Xiaoyi Dong, Jiaqi Wang, Bo Zhang, Lei Bai, Pei Chu, Weijia Li, Jiang Wu, Lijun Wu, Zhenxiang Li, Guangyu Wang, Zhongying Tu, Chao Xu, Kai Chen, Bowen Zhou, Dahua Lin, Wentao Zhang, Conghui He
| Challenge: | Document images are characterized by higher resolutions, denser content, and more complex structural layouts. |
| Approach: | They propose a 1.2B-parameter document parsing vision-language model that decouples layout analysis from local content recognition. |
| Outcome: | The proposed model surpasses general-purpose and domain-specific models on multiple benchmarks while maintaining significantly lower computational overhead. |
Multi-document Summarization through Multi-document Event Relation Graph Reasoning in LLMs: a case study in Framing Bias Mitigation (2025.acl-long)
Copied to clipboard
| Challenge: | a recent study has focused on detecting media bias in news articles . a multi-document event relation graph is used to generate a neutralized summary . |
| Approach: | They propose to generate a neutralized summary given multiple articles presenting different ideological views. |
| Outcome: | The proposed method mitigates media bias and improves content preservation. |
Knowledge Vector of Logical Reasoning in Large Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Logical reasoning is a central capability in LLMs, but understanding their abilities remains poorly understood. |
| Approach: | They propose to refine the knowledge representations of each reasoning type in LLMs to encourage complementarity . they propose to use complementary loss and subspace constraint loss to enhance complementarities . |
| Outcome: | The proposed framework encourages complementarity between the different types of reasoning in LLMs. |
Evaluating Gender Bias of LLMs in Making Morality Judgements (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have shown remarkable capabilities in a multitude of NLP tasks, but are still not immune to limitations such as gender bias. |
| Approach: | They propose to use a dataset to examine whether LLMs possess gender bias when asked to give moral opinions. |
| Outcome: | The proposed models show that they are biased when asked to give moral opinions. |
EMONA: Event-level Moral Opinions in News Articles (2024.naacl-long)
Copied to clipboard
Yuanyuan Lei, Md Messal Monem Miah, Ayesha Qamar, Sai Ramana Reddy, Jonathan Tong, Haotian Xu, Ruihong Huang
| Challenge: | Recent work on news articles has focused on social media short texts, but little has explored moral sentiment within news articles. |
| Approach: | They propose to extract event-level moral opinions from news articles using a new dataset . they use annotated event-based moral opinions to analyze news articles . |
| Outcome: | The proposed dataset consists of 400 news articles containing over 10k sentences and 45k events, among which 9,613 events received moral foundation labels. |
Identifying Conspiracy Theories News based on Event Relation Graph (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Conspiracy theories are narratives that explains an event or situation in an irrational or malicious manner. |
| Approach: | They propose to integrate an event relation graph into conspiracy theory identification by using soft labels. |
| Outcome: | The proposed approach improves precision and recall of conspiracy theory identification, and generalizes well for new unseen media sources. |
Prompting Large Language Models with Chain-of-Thought for Few-Shot Knowledge Base Question Generation (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for question generation over knowledge bases rely on annotated data for fine-tuning . emergence of Large Language Models (LLMs) has shown impressive generalization ability in few-shot tasks. |
| Approach: | They propose to use a logical form to generate a question in a reasoning problem . they propose to extend the prompting method into a method that can generate questions in logical forms . |
| Outcome: | The proposed method outperforms baselines on three public KBQG datasets. |
Discourse Structures Guided Fine-grained Propaganda Identification (2023.emnlp-main)
Copied to clipboard
| Challenge: | Using teacher-predicted probabilities and knowledge distillation frameworks to identify propaganda content is important. |
| Approach: | They propose to integrate local and global discourse structures for propaganda discovery and construct two teacher models for identifying PDTB-style discourse relations between nearby sentences and common discourse roles of sentences in a news article respectively. |
| Outcome: | The proposed models improve accuracy and recall of propaganda content identification at sentence-level and token-level. |
Sentence-level Media Bias Analysis Informed by Discourse Structures (2022.emnlp-main)
Copied to clipboard
| Challenge: | Recent work on detecting media bias at the level of individual articles is limited to single sentences. |
| Approach: | They propose to use a news discourse structure and PDTB discourse relations to identify bias sentences within an article that can illuminate and explain the overall bias of the entire article. |
| Outcome: | The proposed model can detect bias at the level of individual articles and a single sentence can explain it. |
Sentence-level Media Bias Analysis with Event Relation Graph (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing studies on media bias at the article level have identified media biases but only a few have been done on article level. |
| Approach: | They propose to construct an event relation graph to explicitly reason about event-event relations for sentence-level bias identification. |
| Outcome: | The proposed model improves both precision and recall of bias sentence identification. |
Few-Shot (Dis)Agreement Identification in Online Discussions with Regularized and Augmented Meta-Learning (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing annotated datasets do not cover all topics of interest. |
| Approach: | They propose a metric-based meta-learning approach that trains a meta-learner with two key abilities: decoding and generalizing domains. |
| Outcome: | The proposed approach can be quickly applied to analyze opinions for new topics with few labeled instances. |
Polarity Calibration for Opinion Summarization (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing opinions summarization models emphasize the majority opinions while ignoring the minority opinions. |
| Approach: | They propose a method to align output summary and input text to achieve polarity calibration. |
| Outcome: | The proposed model can mitigate the polarity mismatch between output summary and input text, and maintain the content semantic and language quality. |
Discovering a Shared Logical Subspace: Steering LLM Logical Reasoning via Alignment of Natural-Language and Symbolic Views (2026.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to multistep logical reasoning are limited by natural language refinement or external symbolic solvers. |
| Approach: | They propose a logical subspace that captures logical reasoning capabilities in LLMs that are shared across views while remaining independent of surface forms. |
| Outcome: | The proposed approach improves accuracy by 11 percentage points and generalizes well on out-of-domain problems. |