Are Pretrained Language Models Symbolic Reasoners over Knowledge?
Open accessThe DOI matches its Crossref record; 16 of the 16 references it cites were found in a registry or an index.
Abstract
How can pretrained language models (PLMs) learn factual knowledge from the training set?We investigate the two most important mechanisms: reasoning and memorization.Prior work has attempted to quantify the number of facts PLMs learn, but we present, using synthetic data, the first study that investigates the causal relation between facts present in training and facts learned by the PLM.For reasoning, we show that PLMs seem to learn to apply some symbolic reasoning rules correctly but struggle with others, including two-hop reasoning.Further analysis suggests that even the application of learned reasoning rules is flawed.For memorization, we identify schema conformity (facts systematically supported by other facts) and frequency as key factors for its success.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Understanding deep learning requires rethinking generalization
Zhang, C., Bengio, S., Hardt, M., Recht, B., & Vinyals, O. (2016) · arXiv · doi:10.48550/arxiv.1611.03530 · Cited by 2,838
Language Models as Knowledge Bases?
Petroni, F., Rocktäschel, T., Riedel, S., Lewis, P., Bakhtin, A., Wu, Y., & Miller, A. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1250 · Cited by 1,834
HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering
Yang, Z., Qi, P., Zhang, S., Bengio, Y., Cohen, W., Salakhutdinov, R., & Manning, C. D. (2018) · Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d18-1259 · Cited by 1,804
RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space
Sun, Z., Deng, Z.-H., Nie, J.-Y., & Tang, J. (2019) · arXiv · doi:10.48550/arxiv.1902.10197 · Cited by 1,414
A Closer Look at Memorization in Deep Networks
Arpit, D., Jastrzębski, S., Ballas, N., Krueger, D., Bengio, E., Kanwal, M. S., Maharaj, T., Fischer, A., Courville, A., Bengio, Y., et al. (2017) · arXiv · doi:10.48550/arxiv.1706.05394 · Cited by 1,307 · cited 1×
“Based on perceptrons and convolutional neural networks, Arpit et al. (2017); Zhang et al. (2017) study the relation of generalizing from real structured data vs. memorizing random noise in the image domain, similar to our study of schema-conformant facts and outliers.”
MethodRight for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference
McCoy, R. T., Pavlick, E., & Linzen, T. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1334 · Cited by 944
Annotation Artifacts in Natural Language Inference Data
Gururangan, S., Swayamdipta, S., Levy, O., Schwartz, R., Bowman, S., & Smith, N. A. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers) · doi:10.18653/v1/n18-2017 · Cited by 907
Towards AI-Complete Question Answering: A Set of Prerequisite Toy Tasks
Weston, J., Bordes, A., Chopra, S., Rush, A. M., van Merriënboer, B., Joulin, A., & Mikolov, T. (2015) · arXiv · doi:10.48550/arxiv.1502.05698 · Cited by 670 · cited 1×
“A unique feature of our approach compared to prior work (Sinha et al., 2019; Richardson et al., 2020; Weston et al., 2016; Clark et al., 2020) is that we do not gather all relevant facts and present them to the model at inference time.”
What BERT Is Not: Lessons from a New Suite of Psycholinguistic Diagnostics for Language Models
Ettinger, A. (2020) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00298 · Cited by 511
DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs
Dua, D., Wang, Y., Dasigi, P., Stanovsky, G., Singh, S., & Gardner, M. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1246 · Cited by 393 · cited 1×
“Another line of work (Gururangan et al., 2018; Kaushik and Lipton, 2018; Dua et al., 2019; Mc-Coy et al., 2019) shows that much of PLMs’ performance on reasoning tasks is due to statistical artifacts in datasets and does not exhibit true reasoning and generalization capabilities.”
BackgroundDROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs
Dua, D., Wang, Y., Dasigi, P., Stanovsky, G., Singh, S., & Gardner, M. (2019) · arXiv · doi:10.48550/arxiv.1903.00161 · Cited by 238
Compositionality Decomposed: How do Neural Networks Generalise?
Hupkes, D., Dankers, V., Mul, M., & Bruni, E. (2020) · Journal of Artificial Intelligence Research · doi:10.1613/jair.1.11674 · Cited by 194
Transformers as Soft Reasoners over Language
Clark, P., Tafjord, O., & Richardson, K. (2020) · Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2020/537 · Cited by 176
CLUTRR: A Diagnostic Benchmark for Inductive Reasoning from Text
Sinha, K., Sodhani, S., Dong, J., Pineau, J., & Hamilton, W. L. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1458 · Cited by 110 · cited 2×
“For complex reasoning QA benchmarks (Yang et al., 2018; Sinha et al., 2019), PLMs are finetuned to the downstream tasks.”
Background
Cited by
Measuring and improving consistency in pretrained language models
Elazar, Y., Kassner, N., Ravfogel, S., Ravichander, A., Hovy, E., Schütze, H., & Goldberg, Y. (2021) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00410 · Cited by 220
“…2020) perform on linguistic tasks (Goldberg, 2019; Hewitt and Man-ning, 2019; Tenney et al., 2019; Elazar et al., 2021), commonsense (Forbes et al., 2019; Da and Kasai, 2019; Zhang et al., 2020) and reasoning (Talmor et al., 2020; Kassner et al., 2020), usually assessed by measures of accuracy.”
BackgroundSelf-citationLanguage Model Behavior: A Comprehensive Survey
Chang, T. A., & Bergen, B. K. (2024) · Computational Linguistics · doi:10.1162/coli_a_00492 · Cited by 105
Impact of Pretraining Term Frequencies on Few-Shot Numerical Reasoning
Razeghi, Y., Logan IV, R. L., Gardner, M., & Singh, S. (2022) · Findings of the Association for Computational Linguistics: EMNLP 2022 · doi:10.18653/v1/2022.findings-emnlp.59 · Cited by 89
“Impact of Frequency on LM Performance Kassner et al. (2020) and Wei et al. (2021) perform controlled experiments varying pretraining data to characterize the extent to which pretraining affects LMs’ ability to memorize and reason with facts as well as learn generalizable syntax rules.”
MethodAND does not mean OR: Using Formal Languages to Study Language Models’ Representations
Traylor, A., Feiman, R., & Pavlick, E. (2021) · Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers) · doi:10.18653/v1/2021.acl-short.21 · Cited by 19
“Most closely re-lated is work that assesses whether models can perform symbolic reasoning about language (Kassner et al., 2020) e.g., quantifiers or negation (Talmor et al., 2020; Ettinger, 2020; Kassner and Sch¨utze, 2020; Wang et al., 2018) or by measuring the sys-tematicity of models’ inferences…”
Background
Metadata
- Authors
- Kassner, N., Krojer, B., & Schütze, H.
- Venue
- Proceedings of the 24th Conference on Computational Natural Language Learning
- Published
- 2020
- Pages
- 552-564
- OpenAlex
- W3091922147
- Cited by
- 35
- References
- 16
- FWCI
- 3.02
- Type
- conference paper
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.