Papers by Xuanhong Li
TRIDENT: Enhancing Large Language Model Safety with Tri-Dimensional Diversified Red-Teaming Data Synthesis (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) excel in natural language processing tasks but are vulnerable to harmful content and being exploited for malicious purposes. |
| Approach: | They propose a framework to measure the risk coverage of alignment datasets across three dimensions: Lexical Diversity, Malicious Intent, and Jailbreak Tactics. |
| Outcome: | The proposed framework measures risk coverage across Lexical Diversity, Malicious Intent, and Jailbreak Tactics. |
Revisiting Source Context in Nearest Neighbor Machine Translation (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing research does not explicitly consider the source context when retrieving similar examples . |
| Approach: | They propose a method to improve neural machine translation via source context enhancement by integrating a source-aware distance calibration module. |
| Outcome: | The proposed approach can be integrated with representative kNN-MT baselines and achieve significant performance improvements. |