Papers
Topics
Authors
Recent
Search
2000 character limit reached

Characterizing the Role of Similarity in the Property Inferences of Language Models

Published 29 Oct 2024 in cs.CL | (2410.22590v2)

Abstract: Property inheritance -- a phenomenon where novel properties are projected from higher level categories (e.g., birds) to lower level ones (e.g., sparrows) -- provides a unique window into how humans organize and deploy conceptual knowledge. It is debated whether this ability arises due to explicitly stored taxonomic knowledge vs. simple computations of similarity between mental representations. How are these mechanistic hypotheses manifested in contemporary LLMs? In this work, we investigate how LMs perform property inheritance with behavioral and causal representational analysis experiments. We find that taxonomy and categorical similarities are not mutually exclusive in LMs' property inheritance behavior. That is, LMs are more likely to project novel properties from one category to the other when they are taxonomically related and at the same time, highly similar. Our findings provide insight into the conceptual structure of LLMs and may suggest new psycholinguistic experiments for human subjects.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (77)
  1. Can language models encode perceptual structure without grounding? a case study in color. In Proceedings of the 25th Conference on Computational Natural Language Learning, pages 109–132, Online. Association for Computational Linguistics.
  2. CausalGym: Benchmarking causal interpretability methods on linguistic tasks. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 14638–14663, Bangkok, Thailand. Association for Computational Linguistics.
  3. Lawrence W. Barsalou. 1983. Ad hoc categories. Memory & cognition, 11(3):211–227.
  4. Emily M. Bender and Alexander Koller. 2020. Climbing towards NLU: On meaning, form, and understanding in the age of data. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 5185–5198, Online. Association for Computational Linguistics.
  5. Sudeep Bhatia. 2023. Inductive reasoning in minds and machines. Psychological Review.
  6. Sudeep Bhatia and Russell Richie. 2021. Transformer networks of human concept knowledge. Psychological Review.
  7. Towards monosemanticity: Decomposing language models with dictionary learning. Transformer Circuits Thread.
  8. Gabriella Chronis and Katrin Erk. 2020. When is a bishop not like a rook? when it’s like a rabbi! multi-prototype BERT embeddings for estimating semantic relationships. In Proceedings of the 24th Conference on Computational Natural Language Learning, pages 227–244, Online. Association for Computational Linguistics.
  9. Evaluating the Ripple Effects of Knowledge Editing in Language Models. Transactions of the Association for Computational Linguistics, 12:283–298.
  10. Allan M Collins and M Ross Quillian. 1969. Retrieval time from semantic memory. Journal of verbal learning and verbal behavior, 8(2):240–247.
  11. The Llama 3 Herd of Models. arXiv preprint arXiv:2407.21783.
  12. Toy models of superposition. Preprint, arXiv:2209.10652.
  13. Jonathan St BT Evans. 1989. Bias in human reasoning: Causes and consequences. Lawrence Erlbaum Associates, Inc.
  14. Jiahai Feng and Jacob Steinhardt. 2024. How do Language Models Bind Entities in Context? In The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024. OpenReview.net.
  15. Causal analysis of syntactic agreement mechanisms in neural language models. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 1828–1843, Online. Association for Computational Linguistics.
  16. Causal abstraction: A theoretical foundation for mechanistic interpretability. Preprint, arXiv:2301.04709.
  17. Neural natural language inference models partially embed theories of lexical entailment and negation. In Proceedings of the Third BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP, pages 163–173, Online. Association for Computational Linguistics.
  18. Finding alignments between interpretable causal variables and distributed neural representations. In Causal Learning and Reasoning, 1-3 April 2024, Los Angeles, California, USA, volume 236 of Proceedings of Machine Learning Research, pages 160–187. PMLR.
  19. Arnold L Glass and Keith J Holyoak. 1974. Alternative conceptions of semantic theory. Cognition, 3(4):313–339.
  20. Semantic projection recovers rich human knowledge of multiple object features from word embeddings. Nature Human Behaviour, pages 1–13.
  21. Human-like property induction is a challenge for large language models. In Proceedings of the 44th Annual Conference of the Cognitive Science Society.
  22. Inductive reasoning in humans and large language models. Cognitive Systems Research, 83:101155.
  23. Brett K Hayes and Evan Heit. 2018. Inductive reasoning 2.0. Wiley Interdisciplinary Reviews: Cognitive Science, 9(3):e1459.
  24. Things-data, a multimodal collection of large-scale datasets for investigating object representations in human brain and behavior. eLife, 12:e82580.
  25. Things: A database of 1,854 object concepts and more than 26,000 naturalistic object images. PloS one, 14(10):e0223792.
  26. Evan Heit and Joshua Rubinstein. 1994. Similarity and property effects in inductive reasoning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 20(2):411.
  27. John Hewitt and Percy Liang. 2019. Designing and interpreting probes with control tasks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 2733–2743, Hong Kong, China. Association for Computational Linguistics.
  28. Distributed representations, page 77–109. MIT Press, Cambridge, MA, USA.
  29. Rigorously assessing natural language explanations of neurons. In Proceedings of the 6th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP, pages 317–331, Singapore. Association for Computational Linguistics.
  30. RAVEL: Evaluating interpretability methods on disentangling language model representations. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8669–8687, Bangkok, Thailand. Association for Computational Linguistics.
  31. Sparse autoencoders find highly interpretable features in language models. In The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024. OpenReview.net.
  32. Mistral 7b. arXiv:2310.06825.
  33. A high-throughput approach for the efficient prediction of perceived similarity of natural objects. bioRxiv, pages 2024–06.
  34. Charles Kemp. 2011. Inductive reasoning about chimeric creatures. Advances in Neural Information Processing Systems, 24:316–324.
  35. Charles Kemp and Alan Jern. 2014. A Taxonomy of Inductive Problems. Psychonomic Bulletin & Review, 21(1):23–46.
  36. Charles Kemp and Joshua B Tenenbaum. 2009. Structured Statistical Models of Inductive Reasoning. Psychological Review, 116(1):20.
  37. Brenden M Lake and Gregory L Murphy. 2021. Word meaning in minds and machines. Psychological Review.
  38. Language models, like humans, show content effects on reasoning tasks. PNAS nexus, 3(7).
  39. ALBERT: A Lite BERT for Self-supervised Learning of Language Representations. In International Conference on Learning Representations.
  40. Lmms reloaded: Transformer-based sense embeddings for disambiguation and beyond. Artificial Intelligence, 305:103661.
  41. Gary Lupyan and Molly Lewis. 2019. From words-as-mappings to words-as-cues: The role of language in semantic knowledge. Language, Cognition and Neuroscience, 34(10):1319–1337.
  42. Sparse feature circuits: Discovering and editing interpretable causal graphs in language models. CoRR, abs/2403.19647.
  43. Right for the wrong reasons: Diagnosing syntactic heuristics in natural language inference. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 3428–3448, Florence, Italy. Association for Computational Linguistics.
  44. Kanishka Misra. 2022. minicons: Enabling flexible behavioral and representational analyses of transformer language models. arXiv:2203.13112.
  45. Experimental Contexts Can Facilitate Robust Semantic Property Inference in Language Models, but Inconsistently. EMNLP 2024.
  46. Do language models learn typicality judgments from text? In Proceedings of the 43rd Annual Conference of the Cognitive Science Society.
  47. COMPS: Conceptual minimal pair sentences for testing robust property knowledge and its inheritance in pre-trained language models. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, pages 2928–2949, Dubrovnik, Croatia. Association for Computational Linguistics.
  48. A property induction framework for neural language models. In Proceedings of the 44th Annual Conference of the Cognitive Science Society.
  49. The quest for the right mediator: A history, survey, and theoretical grounding of causal interpretability. CoRR, abs/2408.01416.
  50. In-context learning generalizes, but not always robustly: The case of syntax. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 4761–4779, Mexico City, Mexico. Association for Computational Linguistics.
  51. Gregory L Murphy. 2002. The Big Book of Concepts. MIT press.
  52. Semantic memory redux: An experimental test of hierarchical category representation. Journal of Memory and Language, 67(4):521–539.
  53. Category-based Induction. Psychological Review, 97(2):185.
  54. The geometry of categorical and hierarchical concepts in large language models. CoRR, abs/2406.01506.
  55. Roma Patel and Ellie Pavlick. 2022. Mapping language models to grounded conceptual spaces. In International Conference on Learning Representations.
  56. GloVe: Global vectors for word representation. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1532–1543, Doha, Qatar. Association for Computational Linguistics.
  57. Steven T Piantadosi and Felix Hill. 2022. Meaning without reference in large language models. arXiv:2208.02957.
  58. TAXI: Evaluating categorical knowledge editing for language models. In Findings of the Association for Computational Linguistics ACL 2024, pages 15343–15352, Bangkok, Thailand and virtual meeting. Association for Computational Linguistics.
  59. Lance J Rips. 1975. Inductive judgments about natural categories. Journal of Verbal Learning and Verbal Behavior, 14(6):665–681.
  60. Gemma 2: Improving open language models at a practical size. CoRR, abs/2408.00118.
  61. Timothy T Rogers and James L McClelland. 2004. Semantic Cognition: A Parallel Distributed Processing Approach. MIT press.
  62. Eleanor Rosch. 1975. Cognitive representations of semantic categories. Journal of experimental psychology: General, 104(3):192.
  63. Measuring inductive biases of in-context learning with underspecified demonstrations. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11289–11310, Toronto, Canada. Association for Computational Linguistics.
  64. Steven A Sloman. 1993. Feature-based induction. Cognitive psychology, 25(2):231–280.
  65. Steven A Sloman. 1998. Categorical inference is not a tree: The myth of inheritance hierarchies. Cognitive Psychology, 35(1):1–33.
  66. Edward E Smith and William K Estes. 1978. Theories of semantic memory. Handbook of learning and cognitive processes, 6:1–56.
  67. P. Smolensky. 1986. Neural and conceptual interpretation of PDP models, page 390–431. MIT Press, Cambridge, MA, USA.
  68. Investigating gender bias in language models using causal mediation analysis. In Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual.
  69. Editing conceptual knowledge for large language models. CoRR, abs/2403.06259.
  70. Peter C Wason. 1968. Reasoning about a rule. Quarterly journal of experimental psychology, 20(3):273–281.
  71. Transformers: State-of-the-art natural language processing. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, pages 38–45, Online. Association for Computational Linguistics.
  72. Do PLMs know and understand ontological knowledge? In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3080–3101, Toronto, Canada. Association for Computational Linguistics.
  73. pyvene: A library for understanding and improving PyTorch models via interventions. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 3: System Demonstrations), pages 158–165, Mexico City, Mexico. Association for Computational Linguistics.
  74. Interpretability at scale: Identifying causal mechanisms in alpaca. In Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023.
  75. Zhibiao Wu and Martha Palmer. 1994. Verb Semantics and Lexical Selection. In 32nd Annual Meeting of the Association for Computational Linguistics, pages 133–138, Las Cruces, New Mexico, USA. Association for Computational Linguistics.
  76. Fei Xu and Joshua B Tenenbaum. 2007. Word learning as Bayesian inference. Psychological Review, 114(2):245.
  77. Revealing interpretable object representations from human behavior. In 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019. OpenReview.net.

Summary

No one has generated a summary of this paper yet.

Sign Up to Summarize

Paper to Video (Beta)

No one has generated a video about this paper yet.

Sign Up to Generate All Videos Subscribe on YouTube

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Sign Up to Generate

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Generate Now

Continue Learning

We haven't generated follow-up questions for this paper yet.

Generate Now

Collections

Sign up for free to add this paper to one or more collections.

Sign Up

Tweets

Sign up for free to view the 3 tweets with 6 likes about this paper.

Sign Up for Free