Papers by Xiyue Zhu
MedQPA-Gen: Medical Question Proposing and Answering for Report Generation (2026.findings-acl)
Copied to clipboard
Weijie Liang, Xiyue Zhu, Ruike Zhu, Chenhao Li, Cheng Tang, Zhiyu Liu, Zhihua Gong, Shirui Luo, Yudu Li, Volodymyr Kindratenko
| Challenge: | Existing training paradigms fail to explicitly target factual accuracy, resulting in inaccuracies and serious patient safety risks. |
| Approach: | They propose an automatic and generalizable report evaluation technique that uses question proposing and answering to enable controllable, structured reasoning grounded in medical domain knowledge and the factual correctness of the report. |
| Outcome: | The proposed method can improve human preference scores and perform better on downstream tasks. |
Turbocharging Web Automation: The Impact of Compressed History States (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing web automation approaches ignore the importance of history states to accomplish tasks. |
| Approach: | They propose a web history compressor approach to turbocharge web automation using history states by concatenating history states with other inputs. |
| Outcome: | The proposed approach achieves 1.2-5.4% accuracy improvements over baseline methods on Mind2Web and WebLINX datasets. |