Elamur Research⌕
☰
home / papers / 2023 / MoCa: Measuring Human-Language Model Alignment on Causal and Moral Judgment Tasks

MoCa: Measuring Human-Language Model Alignment on Causal and Moral Judgment Tasks

Paper · 2023 · Allen Nie, Yuhui Zhang, Atharva Amdekar et al.

Summary

Large language models are increasingly used as stand-ins for people, yet aggregate accuracy scores reveal little about whether they reason like humans. This NeurIPS 2023 work tests whether LLMs judge causation and moral permissibility the way people do. Instead of crowd-sourced scenarios, the authors drew carefully controlled vignettes from 24 cognitive science studies. In these stories, small wording changes manipulate specific factors, such as norm violations, abnormal events, causal structure, personal force, and whether harm is avoidable. Experts labeled each story with these latent factors, and the authors collected fresh human judgments for the stories. They then compared many LLMs, including models of different sizes, RLHF-tuned systems, and the instruction-distilled Alpaca-7B. Overall agreement with humans improves in newer, larger, and RLHF-tuned models. However, a conjoint analysis based on the Average Marginal Component Effect (AMCE) shows that models weigh the individual factors quite differently from people. The study argues that structured, theory-driven challenge sets can expose hidden model tendencies that headline metrics miss.

Key points

  • The dataset contains a causal judgment set and a moral judgment set built from 24 cognitive science papers. It includes 5150 collected human responses and expert factor annotations. Code and data are available at github.com/cicl-stanford/moca.
  • AMCE quantifies how strongly each factor shifts yes/no judgments for humans and for models. Notably, no model clearly prefers harm as a side effect over harm used as a means, even though humans do.
  • Alignment with human tendencies does not grow steadily with model size. Models trained with similar preference-tuning methods (e.g., Claude-v1 vs. GPT3.5-turbo) also show different implicit tendencies.
  • As a few-shot annotator, text-davinci-v2 labeled factor attributes with 85.5% accuracy on causal stories and 70.9% on moral stories.
  • When explaining their errors, smaller models hallucinated story facts more often (text-curie-001: 8/10 causal, 3/10 moral) than larger ones (GPT-4: 0/10 on both). Larger models seemed to follow fixed moral principles and often took a passive stance, while humans weighed more of the story's nuances.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Fields, topics and concepts

📚 Shelves 8The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksSocial Sciences · 1,772 worksSafety Research · 543 worksNeuroscience · 165 worksCognitive Neuroscience · 164 works

🧭 Topics 1Research subjects and languages the work is about.Language models & pretraining · 3,320 works

Cite

Cite as Allen Nie, Yuhui Zhang, Atharva Amdekar, Chris Piech, Tatsunori Hashimoto and Tobias Gerstenberg. 2023. MoCa: Measuring Human-Language Model Alignment on Causal and Moral Judgment Tasks. arXiv:2310.19677.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2310.19677 · arXiv 2310.19677

@misc{nie2023moca,
  title = {{MoCa: Measuring Human-Language Model Alignment on Causal and Moral Judgment Tasks}},
  author = {Allen Nie, Yuhui Zhang, Atharva Amdekar, Chris Piech, Tatsunori Hashimoto and Tobias Gerstenberg},
  year = {2023},
  doi = {10.48550/arxiv.2310.19677},
  eprint = {2310.19677},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2310.19677}
}

References

110 references, 22 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Mark D Alicke, David Rose, and Dori Bloom. Causation, norm violation, and culpable control. The Journal of Philosophy, 108(12):670–696, 2011. find
  2. Guilherme FCF Almeida, José Luiz Nunes, Neele Engelmann, Alex Wiegmann, and Marcelo de Araújo. Exploring the psychology of gpt-4’s moral and legal reasoning. arXiv preprint arXiv:2308.01264, 2023. find
  3. Edmond Awad, Sohan Dsouza, Richard Kim, Jonathan Schulz, Joseph Henrich, Azim Shariff, Jean-François Bonnefon, and Iyad Rahwan. The moral machine experiment. Nature, 563(7729):59–64, 2018. find
  4. Yuntao Bai, Andy Jones, Kamal Ndousse, Amanda Askell, Anna Chen, Nova DasSarma, Dawn Drain, Stanislav Fort, Deep Ganguli, Tom Henighan, et al. Training a helpful and harmless assistant with reinforcement learning from human feedback. arXiv preprint arXiv:2204.05862, 2022a. find
  5. Yuntao Bai, Saurav Kadavath, Sandipan Kundu, Amanda Askell, Jackson Kernion, Andy Jones, Anna Chen, Anna Goldie, Azalia Mirhoseini, Cameron McKinnon, et al. Constitutional ai: Harmlessness from ai feedback. arXiv preprint arXiv:2212.08073, 2022b.
  6. Chris L. Baker, Julian Jara-Ettinger, Rebecca Saxe, and Joshua B. Tenenbaum. Rational quantitative attribution of beliefs, desires and percepts in human mentalizing. Nature Human Behaviour, 1(4):0064, 2017. doi: 10.1038/s41562-017-0064. URL https://doi.org/10.1038% 2Fs41562-017-0064. — [doi:10.1038/s41562-017-0064](https://doi.org/10.1038/s41562-017-0064) find
  7. Jonathan Baron and Ilana Ritov. Omission bias, individual differences, and normality. Organizational Behavior and Human Decision Processes, 94(2):74–85, 2004. find
  8. Peter W Battaglia, Jessica B Hamrick, and Joshua B Tenenbaum. Simulation as an engine of physical scene understanding. Proceedings of the National Academy of Sciences, 110(45):18327–18332, 2013. find
  9. Paul Bloomfield. Morality and Self-interest. Oxford University Press, 2007. find
  10. Rishi Bommasani, Drew A Hudson, Ehsan Adeli, Russ Altman, Simran Arora, Sydney von Arx, Michael S Bernstein, Jeannette Bohg, Antoine Bosselut, Emma Brunskill, et al. On the opportunities and risks of foundation models. arXiv preprint arXiv:2108.07258, 2021.
  11. Antoine Bosselut, Hannah Rashkin, Maarten Sap, Chaitanya Malaviya, Asli Celikyilmaz, and Yejin Choi. COMET: Commonsense transformers for automatic knowledge graph construction. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pp. 4762–4779, Florence, Italy,…
  12. Tom B Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. Language models are few-shot learners. arXiv preprint arXiv:2005.14165, 2020.
  13. Shyamal Buch, Cristóbal Eyzaguirre, Adrien Gaidon, Jiajun Wu, Li Fei-Fei, and Juan Carlos Niebles. Revisiting the" video" in video-language understanding. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 2917–2927, 2022. find
  14. Stephen Cave, Rune Nyrup, Karina Vold, and Adrian Weller. Motivations and risks of machine ethics. Proceedings of the IEEE, 107(3):562–574, 2019. doi: 10.1109/JPROC.2018.2865996. — [doi:10.1109/jproc.2018.2865996](https://doi.org/10.1109/jproc.2018.2865996) find
  15. Julia F Christensen, Albert Flexas, Margareta Calabrese, Nadine K Gut, and Antoni Gomila. Moral judgment reloaded: a moral dilemma validation study. Frontiers in psychology, 5:607, 2014. find
  16. Kevin Clark, Minh-Thang Luong, Quoc V Le, and Christopher D Manning. Electra: Pre-training text encoders as discriminators rather than generators. arXiv preprint arXiv:2003.10555, 2020. find
  17. Randolph Clarke, Joshua Shepherd, John Stigall, Robyn Repko Waller, and Chris Zarpentine. Causation, norms, and omissions: A study of causal judgments. Philosophical Psychology, 28(2): 279–293, 2015. find
  18. Aida Mostafazadeh Davani, Mark Díaz, and Vinodkumar Prabhakaran. Dealing with disagreements: Looking beyond the majority vote in subjective annotations. Transactions of the Association for Computational Linguistics, 10:92–110, 2022.
  19. Ernest Davis and Gary Marcus. Commonsense reasoning and commonsense knowledge in artificial intelligence. Communications of the ACM, 58(9):92–103, 2015.
  20. Peter DeScioli, Rebecca Bruening, and Robert Kurzban. The omission effect in moral cognition: Toward a functional explanation. Evolution and Human Behavior, 32(3):204–215, 2011. find
  21. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volum…
  22. Danica Dillion, Niket Tandon, Yuling Gu, and Kurt Gray. Can ai language models replace human participants? Trends in Cognitive Sciences, 2023. find
  23. Claude Eatherly and Günther Anders. Burning Conscience: The Case Of The Hiroshima Pilot Claude Eatherly. Pickle Partners Publishing, 2015. find
  24. Jennifer L Eberhardt. Biased: Uncovering the hidden prejudice that shapes what we see, think, and do. Penguin Books, 2020. find
  25. Denis Emelin, Ronan Le Bras, Jena D. Hwang, Maxwell Forbes, and Yejin Choi. Moral stories: Situated reasoning about norms, intents, actions, and their consequences. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pp. 698–718, Online and Punta Cana, Dominica… find
  26. Philippa Foot. The problem of abortion and the doctrine of the double effect. Oxford review, 5, 1967. find
  27. Maxwell Forbes, Jena D. Hwang, Vered Shwartz, Maarten Sap, and Yejin Choi. Social chemistry 101: Learning to reason about social and moral norms. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 653–670, Online, November 2020. Association for Com… find
  28. Iason Gabriel. Artificial intelligence, values, and alignment. Minds and machines, 30(3):411–437, 2020. find
  29. Tobias Gerstenberg and Thomas F. Icard. Expectations affect physical causation judgments. Journal of Experimental Psychology: General, 149(3):599–607, 2020. find
  30. Tobias Gerstenberg and David A. Lagnado. When contributions make a difference: Explaining order effects in responsibility attributions. Psychonomic Bulletin & Review, 19(4):729–736, 2012. find
  31. Tobias Gerstenberg and Simon Stephan. A counterfactual simulation model of causation by omission. Cognition, 2021. URL https://psyarxiv.com/wmh4c/. ↗
  32. Tobias Gerstenberg and Joshua B. Tenenbaum. Intuitive theories. In Michael Waldmannn (ed.), Oxford Handbook of Causal Reasoning, pp. 515–548. Oxford University Press, 2017. find
  33. Tobias Gerstenberg, Noah D. Goodman, David A. Lagnado, and Joshua B. Tenenbaum. A counterfactual simulation model of causal judgments for physical events. Psychological Review, 128(6): 936–975, 2021. find
  34. Fabrizio Gilardi, Meysam Alizadeh, and Maël Kubli. Chatgpt outperforms crowd-workers for text-annotation tasks. arXiv preprint arXiv:2303.15056, 2023.
  35. Ben Green. Data science as political action: Grounding data science in a politics of justice. Journal of Social Computing, 2(3):249–265, 2021. doi: 10.23919/JSC.2021.0029. — [doi:10.23919/jsc.2021.0029](https://doi.org/10.23919/jsc.2021.0029) find
  36. Joshua D Greene, Fiery A Cushman, Lisa E Stewart, Kelly Lowenberg, Leigh E Nystrom, and Jonathan D Cohen. Pushing moral buttons: The interaction between personal force and intention in moral judgment. Cognition, 111(3):364–371, 2009. find
  37. Jens Hainmueller, Daniel J Hopkins, and Teppei Yamamoto. Causal inference in conjoint analysis: Understanding multidimensional choices via stated preference experiments. Political analysis, 22 (1):1–30, 2014. find
  38. Joseph Y Halpern. Cause, responsibility and blame: a structural-model approach. Law, probability and risk, 14(2):91–118, 2015. find
  39. H. L. A. Hart and T. Honoré. Causation in the law. Oxford University Press, New York, 1959/1985. find
  40. Marc Hauser. Moral minds: How nature designed our universal sense of right and wrong. Ecco/HarperCollins Publishers, 2006. find
  41. Iris Hendrickx, Su Nam Kim, Zornitsa Kozareva, Preslav Nakov, Diarmuid Ó Séaghdha, Sebastian Padó, Marco Pennacchiotti, Lorenza Romano, and Stan Szpakowicz. SemEval-2010 task 8: Multi-way classification of semantic relations between pairs of nominals. In Proceedings of the 5th International Workshop…
  42. Dan Hendrycks, Collin Burns, Steven Basart, Andrew Critch, Jerry Li, Dawn Song, and Jacob Steinhardt. Aligning ai with shared human values. arXiv preprint arXiv:2008.02275, 2020.
  43. Dan Hendrycks, Mantas Mazeika, Andy Zou, Sahil Patel, Christine Zhu, Jesus Navarro, Dawn Song, Bo Li, and Jacob Steinhardt. What would jiminy cricket do? towards agents that behave morally. arXiv preprint arXiv:2110.13136, 2021. find
  44. Paul Henne, Ángel Pinillos, and Felipe De Brigard. Cause by omission and norm: Not watering plants. Australasian Journal of Philosophy, 95(2):270–283, 2017. find
  45. Paul Henne, Laura Niemi, Ángel Pinillos, Felipe De Brigard, and Joshua Knobe. A counterfactual explanation for the action effect in causal judgment. Cognition, 190:157–164, 2019. find
  46. Paul Henne, Aleksandra Kulesza, Karla Perez, and Augustana Houcek. Counterfactual thinking and recency effects in causal judgment. Cognition, 212:104708, 2021. find
  47. D. J. Hilton and B. R. Slugoski. Knowledge-based causal attribution: The abnormal conditions focus model. Psychological Review, 93(1):75–88, 1986. find
  48. D. J. Hilton, J. McClure, and R. M. Sutton. Selecting explanations from causal chains: Do statistical principles explain preferences for voluntary causes. European Journal of Social Psychology, 40(3): 383–400, 2010. find
  49. Christopher Hitchcock and Joshua Knobe. Cause and norm. The Journal of Philosophy, 106(11): 587–612, 2009. find
  50. Bryce Huebner, Marc D Hauser, and Phillip Pettit. How the source, inevitability and means of bringing about harm interact in folk-moral judgments. Mind & Language, 26(2):210–233, 2011. find
  51. Thomas F Icard, Jonathan F Kominsky, and Joshua Knobe. Normality and actual causal strength. Cognition, 161:80–93, 2017. find
  52. Liwei Jiang, Jena D Hwang, Chandra Bhagavatula, Ronan Le Bras, Maxwell Forbes, Jon Borchardt, Jenny Liang, Oren Etzioni, Maarten Sap, and Yejin Choi. Delphi: Towards machine ethics and norms. arXiv preprint arXiv:2110.07574, 2021. find
  53. Zhijing Jin, Sydney Levine, Fernando Gonzalez, Ojasv Kamal, Maarten Sap, Mrinmaya Sachan, Rada Mihalcea, Josh Tenenbaum, and Bernhard Schölkopf. When to make exceptions: Exploring language models as accounts of human moral judgment. arXiv preprint arXiv:2210.01478, 2022. — [arXiv:2210.01478](https://arxiv.org/abs/2210.01478) find
  54. Noa Kallioinen, Maria Pershina, Jannik Zeiser, Farbod Nosrat Nezami, Gordon Pipa, Achim Stephan, and Peter König. Moral judgements on the actions of self-driving cars and human drivers in dilemma situations from different perspectives. Frontiers in psychology, 10:2415, 2019. find
  55. Zachary Kenton, Tom Everitt, Laura Weidinger, Iason Gabriel, Vladimir Mikulik, and Geoffrey Irving. Alignment of language agents. arXiv preprint arXiv:2103.14659, 2021. find
  56. Emre Kıcıman, Robert Ness, Amit Sharma, and Chenhao Tan. Causal reasoning and large language models: Opening a new frontier for causality. arXiv preprint arXiv:2305.00050, 2023. find
  57. Max Kleiman-Weiner, Tobias Gerstenberg, Sydney Levine, and Joshua B Tenenbaum. Inference of intention and permissibility in moral decision making. In CogSci, 2015. find
  58. Joshua Knobe and Ben Fraser. Causal judgment and moral judgment: Two experiments. Moral psychology, 2:441–447, 2008. find
  59. Jonathan F Kominsky and Jonathan Phillips. Immoral professors and malfunctioning tools: Counterfactual relevance accounts explain the effect of norm violations on causal selection. Cognitive science, 43(11):e12792, 2019. find
  60. Jonathan F Kominsky, Jonathan Phillips, Tobias Gerstenberg, David Lagnado, and Joshua Knobe. Causal superseding. Cognition, 137:196–209, 2015. find
  61. Eliza Kosoy, David M Chan, Adrian Liu, Jasmine Collins, Bryanna Kaufmann, Sandy Han Huang, Jessica B Hamrick, John Canny, Nan Rosemary Ke, and Alison Gopnik. Towards understanding how machines can learn causal overhypotheses. arXiv preprint arXiv:2206.08353, 2022. — [arXiv:2206.08353](https://arxiv.org/abs/2206.08353) find
  62. D. A. Lagnado, T. Gerstenberg, and R. Zultan. Causal responsibility and counterfactuals. Cognitive Science, 47:1036–1073, 2013. find
  63. Brenden M Lake, Tomer D Ullman, Joshua B Tenenbaum, and Samuel J Gershman. Building machines that learn and think like people. Behavioral and brain sciences, 40, 2017. find
  64. Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, and Radu Soricut. Albert: A lite bert for self-supervised learning of language representations. arXiv preprint arXiv:1909.11942, 2019.
  65. S Matthew Liao, Alex Wiegmann, Joshua Alexander, and Gerard Vong. Putting the trolley in order: Experimental philosophy and the loop case. Philosophical Psychology, 25(5):661–671, 2012. find
  66. Hairong Liu, Mingbo Ma, Liang Huang, Hao Xiong, and Zhongjun He. Robust neural machine translation with joint textual and phonetic embedding. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pp. 3044–3049, Florence, Italy, July 2019a. Association for Comput… find
  67. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692, 2019b.
  68. Nicholas Lourie, Ronan Le Bras, and Yejin Choi. Scruples: A corpus of community ethical judgments on 32,000 real-life anecdotes. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pp. 13470–13479, 2021. find
  69. David R Mandel. Judgment dissociation theory: An analysis of differences in causal, counterfactual and covariational reasoning. Journal of Experimental Psychology: General, 132(3):419, 2003. find
  70. Ian R McKenzie, Alexander Lyzhov, Michael Pieler, Alicia Parrish, Aaron Mueller, Ameya Prabhu, Euan McLean, Aaron Kirtland, Alexis Ross, Alisa Liu, et al. Inverse scaling: When bigger isn’t better. arXiv preprint arXiv:2306.09479, 2023.
  71. John Mikhail. Universal moral grammar: Theory, evidence and the future. Trends in cognitive sciences, 11(4):143–152, 2007. find
  72. Adam B Moore, Brian A Clark, and Michael J Kane. Who shalt not kill? individual differences in working memory capacity, executive control, and moral judgment. Psychological science, 19(6): 549–557, 2008. find
  73. Adam Morris, Jonathan Phillips, Tobias Gerstenberg, and Fiery Cushman. Quantitative causal selection patterns in token causation. PloS one, 14(8):e0219704, 2019. find
  74. Nasrin Mostafazadeh, Nathanael Chambers, Xiaodong He, Devi Parikh, Dhruv Batra, Lucy Vanderwende, Pushmeet Kohli, and James Allen. A corpus and cloze evaluation for deeper understanding of commonsense stories. In Proceedings of the 2016 Conference of the North American Chapter of the Association for… find
  75. Ahogni N’gbala and Nyla R Branscombe. Mental simulation and causal attribution: When simulating an event does not affect fault assignment. Journal of Experimental Social Psychology, 31(2): 139–162, 1995. find
  76. Kevin O’Neill, Paul Henne, Paul Bello, John Pearson, and Felipe De Brigard. Degrading causation. 2021. find
  77. OpenAI. Gpt-4 technical report. arXiv, 2023.
  78. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35: 27730–27744, 2022.
  79. Alexander Pan, Jun Shern Chan, Andy Zou, Nathaniel Li, Steven Basart, Thomas Woodside, Hanlin Zhang, Scott Emmons, and Dan Hendrycks. Do the rewards justify the means? measuring trade-offs between rewards and ethical behavior in the machiavelli benchmark. In International Conference on Machine Learn… find
  80. Joon Sung Park, Lindsay Popowski, Carrie Cai, Meredith Ringel Morris, Percy Liang, and Michael S Bernstein. Social simulacra: Creating populated prototypes for social computing systems. In Proceedings of the 35th Annual ACM Symposium on User Interface Software and Technology, pp. 1–18, 2022. find
  81. Judea Pearl and Dana Mackenzie. The book of why: the new science of cause and effect. Basic books, 2018. find
  82. Tadeg Quillien and Christopher G Lucas. Counterfactuals and the logic of causal selection. Psychological Review, 2023. find
  83. Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, Ilya Sutskever, et al. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9, 2019. find
  84. Madeline G Reinecke, Yiran Mao, Markus Kunesch, Edgar A Duéñez-Guzmán, Julia Haas, and Joel Z Leibo. The puzzle of evaluating moral cognition in artificial agents. Cognitive Science, 47 (8):e13315, 2023. find
  85. Kevin Reuter, Lara Kirfel, Raphael Van Riel, and Luca Barlassina. The good, the bad, and the timely: how temporal order and moral judgment influence causal selection. Frontiers in psychology, 5: 1336, 2014. find
  86. Ilana Ritov and Jonathan Baron. Status-quo and omission biases. Journal of Risk and Uncertainty, 5 (1), 1992. doi: 10.1007/BF00208786. URL http://dx.doi.org/10.1007/BF00208786. — [doi:10.1007/bf00208786](https://doi.org/10.1007/bf00208786) ↗
  87. Melissa Roemmele, Cosmin Adrian Bejan, and Andrew S Gordon. Choice of plausible alternatives: An evaluation of commonsense causal reasoning. In 2011 AAAI Spring Symposium Series, 2011. find
  88. Jana Samland, Marina Josephs, Michael R Waldmann, and Hannes Rakoczy. The role of prescriptive norms and knowledge in children’s and adults’ causal selection. Journal of Experimental Psychology: General, 145(2):125, 2016. find
  89. Maarten Sap, Ronan Le Bras, Emily Allaway, Chandra Bhagavatula, Nicholas Lourie, Hannah Rashkin, Brendan Roof, Noah A Smith, and Yejin Choi. Atomic: An atlas of machine commonsense for if-then reasoning. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 33-01, pp. 3027–3035, 2… find
  90. Nino Scherrer, Claudia Shi, Amir Feder, and David M Blei. Evaluating the moral beliefs encoded in llms. arXiv preprint arXiv:2307.14324, 2023. find
  91. Steven A Sloman and David Lagnado. Causality in thought. Annual review of psychology, 66: 223–247, 2015. find
  92. B. A. Spellman. Crediting causality. Journal of Experimental Psychology: General, 126(4):323–348, 1997. find
  93. Nisan Stiennon, Long Ouyang, Jeffrey Wu, Daniel Ziegler, Ryan Lowe, Chelsea Voss, Alec Radford, Dario Amodei, and Paul F Christiano. Learning to summarize with human feedback. Advances in Neural Information Processing Systems, 33:3008–3021, 2020. find
  94. Justin Sytsma. Crossed wires: Blaming artifacts for bad outcomes. 2021. find
  95. Zeerak Talat, Hagen Blix, Josef Valvoda, Maya Indira Ganesh, Ryan Cotterell, and Adina Williams. A word on machine ethics: A response to jiang et al. (2021), 2021. find
  96. Alon Talmor, Jonathan Herzig, Nicholas Lourie, and Jonathan Berant. CommonsenseQA: A question answering challenge targeting commonsense knowledge. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume…
  97. Judith Jarvis Thomson. Killing, letting die, and the trolley problem. The Monist, 59(2):204–217, 1976. find
  98. Pedro A Tsividis, Joao Loula, Jake Burga, Nathan Foss, Andres Campero, Thomas Pouncy, Samuel J Gershman, and Joshua B Tenenbaum. Human-level reinforcement learning through theory-based modeling, exploration, and planning. arXiv preprint arXiv:2107.12544, 2021. — [arXiv:2107.12544](https://arxiv.org/abs/2107.12544) find
  99. Tomer D. Ullman, Elizabeth Spelke, Peter Battaglia, and Joshua B. Tenenbaum. Mind games: Game engines as an architecture for intuitive physics. Trends in Cognitive Sciences, 21(9):649–665, 2017. doi: 10.1016/j.tics.2017.05.012. URL https://doi.org/10.1016%2Fj.tics.2017.05.012. find
  100. Michael R Waldmann and Jörn H Dieterich. Throwing a bomb on a person versus throwing a person on a bomb: Intervention myopia in moral intuitions. Psychological science, 18(3):247–253, 2007. find
  101. Alex Wang, Yada Pruksachatkun, Nikita Nangia, Amanpreet Singh, Julian Michael, Felix Hill, Omer Levy, and Samuel R Bowman. Superglue: A stickier benchmark for general-purpose language understanding systems. arXiv preprint arXiv:1905.00537, 2019.
  102. Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc Le, Ed Chi, and Denny Zhou. Self-consistency improves chain of thought reasoning in language models. arXiv preprint arXiv:2203.11171, 2022.
  103. Jason Wei, Yi Tay, Rishi Bommasani, Colin Raffel, Barret Zoph, Sebastian Borgeaud, Dani Yogatama, Maarten Bosma, Denny Zhou, Donald Metzler, et al. Emergent abilities of large language models. arXiv preprint arXiv:2206.07682, 2022a.
  104. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Ed Chi, Quoc Le, and Denny Zhou. Chain of thought prompting elicits reasoning in large language models. arXiv preprint arXiv:2201.11903, 2022b.
  105. H. M. Wellman and S. A. Gelman. Cognitive development: Foundational theories of core domains. Annual Review of Psychology, 43(1):337–375, 1992. find
  106. Gary L Wells and Igor Gavanski. Mental simulation of causality. Journal of personality and social psychology, 56(2):161, 1989. find
  107. Alex Wiegmann, Yasmina Okan, and Jonas Nagel. Order effects in moral judgment. Philosophical Psychology, 25(6):813–836, 2012. find
  108. Phillip Wolff. Representing causation. Journal of experimental psychology: General, 136(1):82, 2007. find
  109. Yongchao Zhou, Andrei Ioan Muresanu, Ziwen Han, Keiran Paster, Silviu Pitis, Harris Chan, and Jimmy Ba. Large language models are human-level prompt engineers. arXiv preprint arXiv:2211.01910, 2022.
  110. Caleb Ziems, Jane A Yu, Yi-Chia Wang, Alon Halevy, and Diyi Yang. The moral integrity corpus: A benchmark for ethical dialogue systems. arXiv preprint arXiv:2204.03021, 2022.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/nie-2023-moca-measuring-human-language-model/