義 資 料 LLM 東 京 大 学 松 尾 研 究 室 LLM 大規模言語モデル講座 講義資料 © 2024 by 東京大学松尾研究室 Reference 47 [1]Fiona Macpherson (ed.),Dimitris Platchias (ed.) “Hallucination: Philosophy and Psychology” The MIT Press:9 August 2013 [2] Katja Filippova. 2020. Controlled Hallucinations: Learning to Generate Faithfully from Noisy Data. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 864–870, Online. Association for Computational Linguistics. [3] Joshua Maynez et al., (2020). On Faithfulness and Factuality in Abstractive Summarization. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1906–1919, Online. Association for Computational Linguistics. [4] Sourav Banerjee et al.,(2024) “LLMs Will Always Hallucinate, and We Need to Live With This” arXiv:2409.05746v1 [5] Lei Huang et al.,(2023) “A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions” arXiv: 2311.05232 [6] Common Crawl Common Crawl Get Started アクセス日 2024.1011 [7] 松尾岩澤研究室 大規模言語モデルDeep Learning応用講座2024|Fall アクセス日2024.1011 [8] Stephanie Lin et al.,(2021) “TruthfulQA: Measuring How Models Mimic Human Falsehoods” arXiv: 2109.07958 [9] Danny Hernandez et al.,(2022) “Scaling Laws and Interpretability of Learning from Repeated Data” arXiv: 2205.10487 [10] Faisal Ladhak et al.,(2023) ”When Do Pre-Training Biases Propagate to Downstream Tasks? A Case Study in Text Summarization” In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, pages 3206–3219, Dubrovnik, Croatia. Association for Computational Linguistics. [11] Cheongwoong Kang and Jaesik Choi. 2023. Impact of Co-occurrence on Factual Knowledge of Large Language Models. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 7721–7735, Singapore. Association for Computational Linguistics. [12] Kandpal, N et al.,(2023). “Large Language Models Struggle to Learn Long-Tail Knowledge”Proceedings of the 40th International Conference on Machine Learning, Proceedings of Machine Learning Research :202:15696-15707 [13] Kewei Cheng et al.,(2024) “Inductive or Deductive? Rethinking the Fundamental Reasoning Abilities of LLMs” arXiv:2408.00144 [14] Zuchao Li et al.,(2023) “BatGPT: A Bidirectional Autoregressive Talker from Generative Pre-Trained Transformer” arXiv: 2307.00360 [15] Bingbin Liu et al.,(2023) “Exposing Attention Glitches with Flip-Flop Language Modeling” arXiv: 2306.00946 [16] Tianzhu Ye et al.,(2024) “Differential Transformer” arXiv: 2410.05258 [17] John Schulman “John Schulman –Reinforcement Learning from Human Feedback: Progress and Challenges” YouTube Berkely EECS Channel [18] Ajeya Cotra(2021) “Why AI alignment could be hard with modern deep learning” Cold Takes アクセス日 2024.0920 [19] Ethan Perez et al.,(2022) “Discovering Language Model Behaviors with Model-Written Evaluations” arXiv:2212.09251 [20] Jerry Wei et al.,(2023) “Simple synthetic data reduces sycophancy in large language models” arXiv: 2308.03958 [21] Nouha Dziri et al.,(2021) “Nueral path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding” arXiv:2104.08455 [22] Yung-Sung Chuang et al.,(2023) “DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models” arXiv: 2309.03883 [23] Yung-Sung Chuang et al.,(2024) “Lookback Lens: Detecting and Mitigating Contextual Hallucinations in Large Language Models Using Only Attention Maps” arXiv: 2407.07071 [24] Selvan Sunitha Ravi et al.,(2024) “Lynx: An Open Source Hallucination Evaluation Model” arXiv:2407.08488 [25] OpenAI “Finding GPT-4’s mistakes with GPT-4” アクセス日 2024.0920 [26] Nat McAleese et al.,(2024) “LLM Critics Help Catch LLM Bugs” arXiv: 2407.00215 [27] Potsawee Manakul et al.,(2023) “SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models” arXiv:2303.08896