Nicola Cancedda

发表论文 125 篇 · 总被引 1509 次 · h-index 18

代表论文

• HalluLens: LLM Hallucination Benchmark (2025 · Annual Meeting of the Association for Computational Linguistics · 被引 130)
• Robust LLM safeguarding via refusal feature adversarial training (2024 · International Conference on Learning Representations · 被引 70)
• Prediction of Learning Curves in Machine Translation (2012 · Annual Meeting of the Association for Computational Linguistics · 被引 60)
• Calibrating Verbal Uncertainty as a Linear Feature to Reduce Hallucinations (2025 · Conference on Empirical Methods in Natural Language Processing · 被引 43)
• Verifying Chain-of-Thought Reasoning via Its Computational Graph (2025 · arXiv.org · 被引 21)
• Generation of Compound Words in Statistical Machine Translation into Compounding Languages (2013 · International Conference on Computational Logic · 被引 21)