快速访问
文章信息
参考文献
[1]Sharma M, Tong M, Korbak T, et al. Towards understanding sycophancy in language models[C]// International Conference on Learning Representations. 2024, 2024: 110-144.
[2]滕斌圣, 曹欣蓓. 当大模型开始“谄媚”，我们该如何保持清醒？[J]. 商学院, 2025(7): 12-12.
[3]邓天奇, 李达军. 从技术顺从到意义偏置： AI谄媚的传播表征、结构性后果与制度调适[J/OL]. 新闻爱好者, 1-12 [2026-03-07]. https://doi.org/10.16017/j.cnki.xwahz.20260113.001.
[4]Malmqvist L. Sycophancy in large language models: Causes and mitigations[C]// Intelligent Computing-Proceedings of the Computing Conference. Cham: Springer Nature Switzerland, 2025: 61-74.
[5]Zhao Y, Zhang R, Xiao J, et al. Towards analyzing and mitigating sycophancy in large vision-language models[J]. arXiv preprint arXiv:2408.11261, 2024, 3.
[6]Zhao Y, Zhang R, Xiao J, et al. Sycophancy in vision-language models: A systematic analysis and an inference-time mitigation framework[J]. Neurocomputing, 2025: 131217.
[7]Laban P, Murakhovs'ka L, Xiong C, et al. Are You Sure? Challenging LLMs Leads to Performance Drops in the FlipFlop Experiment[J]. arXiv preprint arXiv:2311.08596, 2023.
[8]Aswin R R V, Tyagi N, Uddin M N, et al. Chaos with keywords: Exposing large language models sycophantic hallucination to misleading keywords and evaluating defense strategies[J]. Preprint, 2024.
[9]Li Y, Du Y, Zhou K, et al. Evaluating Object Hallucination in Large Vision-Language Models[C]// Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. 2023: 292-305.
[10]Wang J, Wang Y, Xu G, et al. AMBER: An LLM-free Multi-dimensional Benchmark for MLLMs Hallucination Evaluation[J]. arXiv preprint arXiv:2311.07397, 2023.
[11]Wei J W, Huang D, Lu Y, et al. Simple Synthetic Data Reduces Sycophancy in Large Language Models[J]. arXiv preprint arXiv:2308.03958, 2023.
[12]Chen W, Huang Z, Xie L, et al. From Yes-Men to Truth-Tellers: Addressing Sycophancy in Large Language Models with Pinpoint Tuning[J]. arXiv preprint arXiv:2409.01658, 2024.
[13]Khan A A, Alam S, Wang X, et al. Mitigating sycophancy in large language models via direct preference optimization[C]// 2024 IEEE International Conference on Big Data (BigData). IEEE, 2024: 1664-1671.
[14]Cheng M, Yu S, Lee C, et al. Social sycophancy: A broader understanding of LLM sycophancy[J]. arXiv preprint arXiv:2505.13995, 2025.
[15]马儒宁, 涂小坡, 丁军娣, 等. 视觉显著性凸显目标的评价[J]. 自动化学报, 2012, 38(5): 870-876.
[16]Said M N, Zaidi A, Usman R, et al. Deconstructing bias: A multifaceted framework for diagnosing cultural and compositional inequities in text-to-image generative models[J]. arXiv preprint arXiv:2505.01430, 2025.
[17]Itti L, Koch C, Niebur E. A model of saliency-based visual attention for rapid scene analysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1998, 20(11): 1254-1259.
[18]Elsharif W, Alzubaidi M, Agus M. Cultural bias in text-to-image models: A systematic review of bias identification, evaluation and mitigation strategies[J]. IEEE Access, 2025.
[19]Huang K, Sun K, Xie E, et al. T2i-compbench: A comprehensive benchmark for open-world compositional text-to-image generation[J]. Advances in Neural Information Processing Systems, 2023, 36: 78723-78747.
[20]Luo H, Deng Z, Chen R, et al. Faintbench: A holistic and precise benchmark for bias evaluation in text-to-image models[J]. arXiv preprint arXiv:2405.17814, 2024.
[21]Holland E, Wolf E B, Looser C, et al. Visual attention to powerful postures: People avert their gaze from nonverbal dominance displays[J]. Journal of Experimental Social Psychology, 2017, 68: 60-67.
[22]Schubert T W. Grounding of rank: embodiment, space, and magnitude[J]. Current Opinion in Psychology, 2020, 33: 222-226.
[23]Solomon Mathialagan C, Gallagher A C, Batra D. Vip: Finding important people in images[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015: 4858-4866.
版权与开放获取声明
作为一本开放获取的学术期刊，所有文章均遵循 Creative Commons Attribution 4.0 International License (CC BY 4.0) 协议发布，允许用户在署名原作者的前提下自由共享与再利用内容。所有文章均可免费供读者和机构阅读、下载、引用与传播，EWA Publishing 不会通过期刊的出版发行向读者或机构收取任何费用。