Sorting through the noise: Testing robustness of information processing in pre-trained language models
Open accessThe DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
Pre-trained LMs have shown impressive performance on downstream NLP tasks, but we have yet to establish a clear understanding of their sophistication when it comes to processing, retaining, and applying information presented in their input.In this paper we tackle a component of this question by examining robustness of models' ability to deploy relevant context information in the face of distracting content.We present models with cloze tasks requiring use of critical context information, and introduce distracting content to test how robustly the models retain and use that critical information for prediction.We also systematically manipulate the nature of these distractors, to shed light on dynamics of models' use of contextual cues.We find that although models appear in simple contexts to make predictions based on understanding and applying relevant facts from prior context, the presence of distracting but irrelevant content has clear impact in confusing model predictions.In particular, models appear particularly susceptible to factors of semantic similarity and word position.The findings are consistent with the conclusion that LM predictions are driven in large part by superficial contextual cues, rather than by robust representations of context meaning.
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
Know What You Don't Know: Unanswerable Questions for SQuAD
Rajpurkar, P., Jia, R., & Liang, P. (2018) · arXiv · doi:10.48550/arxiv.1806.03822 · Cited by 2,242
Know What You Don’t Know: Unanswerable Questions for SQuAD
Rajpurkar, P., Jia, R., & Liang, P. (2018) · Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) · doi:10.18653/v1/p18-2124 · Cited by 2,240
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 · cited 2×
“In focusing on models’ ability to extract, retain, and deploy information conveyed in text, our work also relates to tasks in reading comprehension question answering (Rajpurkar et al., 2018; Kočiskỳ et al., 2018; Mostafazadeh et al., 2017; Yang et al., 2018; Richardson et al., 2013).”
BackgroundA primer in BERTology: What we know about how BERT works
Rogers, A., Kovaleva, O., & Rumshisky, A. (2020) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00349 · Cited by 1,552 · cited 1×
“A good deal of prior work has focused on testing for linguistic knowledge in language models (Rogers et al., 2020).”
BackgroundAdversarial Examples for Evaluating Reading Comprehension Systems
Jia, R., & Liang, P. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1215 · Cited by 1,410 · cited 2×
“, 2016), use insertion of additional material to make the tasks generally more difficult— a tactic that also parallels the related method of adversarial testing (Jia and Liang, 2017; McCoy et al., 2019; Nie et al., 2020).”
MethodHow Can We Know What Language Models Know?
Jiang, Z., Xu, F. F., Araki, J., & Neubig, G. (2020) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00324 · Cited by 1,002
Right 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
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 2×
“Some such work, like the bAbI dataset (Weston et al., 2016) and CBT (Hill et al., 2016), use insertion of additional material to make the tasks generally more difficult— a tactic that also parallels the related method of adversarial testing (Jia and Liang, 2017; McCoy et al., 2019; Nie et al.,…”
Method
Cited by
Eight Things to Know about Large Language Models
Bowman, S. R. (2023) · arXiv · doi:10.48550/arxiv.2304.00612 · Cited by 137
“There are many sound scientific results showing that recent LLMs fail at language and commonsense reasoning tasks, sometimes relatively simple ones, under good-faith attempts to elicit good behavior (Pandia & Ettinger, 2021; Schuster & Linzen, 2022).”
Large Language Models Can Be Easily Distracted by Irrelevant Context
Shi, F., Chen, X., Misra, K., Scales, N., Dohan, D., Chi, E., Schärli, N., & Zhou, D. (2023) · arXiv · doi:10.48550/arxiv.2302.00093 · Cited by 134
“Similar issues have been found for general-purpose pretrained language models, on the tasks of factual reasoning (Kassner & Sch¨utze, 2020; Pandia & Ettinger, 2021; Misra et al., 2023; Li et al., 2022), code generation (Jones & Steinhardt, 2022), and syntactic generalization (Chaves & Richter,…”
BackgroundLarge language models demonstrate the potential of statistical learning in language
Contreras Kallens, P., Kristensen‐McLachlan, R. D., & Christiansen, M. H. (2023) · Cognitive Science · doi:10.1111/cogs.13256 · Cited by 130
“Some of this work has already been done in the context of encoder-only masked language models, such as BERT and its related descendants (Ettinger, 2020; Pandia & Ettinger, 2021; Rogers, Kovaleva, & Rumshisky, 2020).”
Language Model Behavior: A Comprehensive Survey
Chang, T. A., & Bergen, B. K. (2024) · Computational Linguistics · doi:10.1162/coli_a_00492 · Cited by 105
Metadata
- Authors
- Pandia, L., & Ettinger, A.
- Venue
- Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
- Published
- 2021
- Pages
- 1583-1596
- OpenAlex
- W3202546170
- Cited by
- 22
- References
- 15
- FWCI
- 2.47
- 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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Sorting through the noise: Testing robustness of information processing in pre-trained language models. https://app.cochat.ai/papers/10.18653/v1/2021.emnlp-main.119
- The paper
- Pandia, L., & Ettinger, A. (2021). Sorting through the noise: Testing robustness of information processing in pre-trained language models. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. https://doi.org/10.18653/v1/2021.emnlp-main.119
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.