Papers by Haotong Wang
AnaScore: Understanding Semantic Parallelism in Proportional Analogies (2025.naacl-long)
Copied to clipboard
| Challenge: | AnaScore metric aims to evaluate the strength of semantic parallelism in sentence analogies. |
| Approach: | They propose an automatic metric to evaluate the strength of semantic parallelism in sentence analogies. |
| Outcome: | The proposed metric shows that formally explainable examples are more beneficial for analogical reasoning, whereas ambiguous analogies with no clear criterion tend to hinder inference. |
Search to Pass Messages for Temporal Knowledge Graph Completion (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies on missing facts in temporal knowledge graphs are based on hand-designed architectures and fail to explore the diverse topological and temporal properties of TKGs. |
| Approach: | They propose to use neural architecture search to design a data-specific message passing architecture for TKG completion. |
| Outcome: | The proposed architectures achieve the state-of-the-art performance on three benchmark datasets. |
Continued Pre-training on Sentence Analogies for Translation with Small Data (2024.lrec-main)
Copied to clipboard
| Challenge: | Using 10 times fewer instances, CPoA can achieve gains of +1.4 and +1.3 BLEU points over the original model. |
| Approach: | They propose to train models with analogical abilities on sentence analogies retrieved from corpus . they use a weighting scalar to adjust the influence of closer analogies while diminishing impact of far ones . |
| Outcome: | The proposed approach improves translation performance on a low-resource translation task in german-upper sorbian . it uses 10 times fewer instances to achieve gains of +1.4 and +1.3 BLEU points over the original model . |
A.S.E: A Repository-Level Benchmark for Evaluating Security in AI-Generated Code (2026.findings-acl)
Copied to clipboard
Keke Lian, Wang Bin, Lei Zhang, Libo Chen, Junjie Wang, Ziming Zhao, Yujiu Yang, Miaoqian Lin, Haotong Duan, Haoran Zhao, Shuang Liao, Mingda Guo, Quan Jiazheng, Yilu Zhong, Chenhao He, Chen Zichuan, Jie Wu, Haoling Li, Zhaoxuan Li, Jiongchi Yu, Hui LI, Dong Zhang
| Challenge: | Existing security evaluation benchmarks lack relevance to real-world AI programming tasks . current LLMs struggle with secure coding, research shows . |
| Approach: | They propose a repository-level evaluation benchmark to assess security of AI-generated code. |
| Outcome: | The proposed framework mirrors real-world AI programming tasks and offers valuable insights into the state of AI code generation. |