# Probing Natural Language Inference Models through Semantic Fragments

arXiv: https://arxiv.org/abs/1909.07521 · HTML: https://mlsanity.com/paper/1909.07521/

## Key facts

- Cited by 64 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Sep 2019.

## Built on by (most cited first, 30 of 64)

- [Probing BERT in Hyperbolic Spaces](https://mlsanity.com/paper/2104.03869/) (2021)
- [Do Transformers Parse while Predicting the Masked Word?](https://mlsanity.com/paper/2303.08117/) (2023)
- [Evaluating Entity Disambiguation and the Role of Popularity in Retrieval-Based NLP](https://mlsanity.com/paper/2106.06830/) (2021)
- [Commonsense-Focused Dialogues for Response Generation: An Empirical Study](https://mlsanity.com/paper/2109.06427/) (2021)
- [Figurative Language in Recognizing Textual Entailment](https://mlsanity.com/paper/2106.01195/) (2021)
- [IndoNLI: A Natural Language Inference Dataset for Indonesian](https://mlsanity.com/paper/2110.14566/) (2021)
- [Pushing the Limits of Rule Reasoning in Transformers through Natural Language Satisfiability](https://mlsanity.com/paper/2112.09054/) (2021)
- [Learning Deductive Reasoning from Synthetic Corpus based on Formal Logic](https://mlsanity.com/paper/2308.07336/) (2023)
- [Exploring Transitivity in Neural NLI Models through Veridicality](https://mlsanity.com/paper/2101.10713/) (2021)
- [Thinking Aloud: Dynamic Context Generation Improves Zero-Shot Reasoning Performance of GPT-2](https://mlsanity.com/paper/2103.13033/) (2021)
- [multiPRover: Generating Multiple Proofs for Improved Interpretability in Rule Reasoning](https://mlsanity.com/paper/2106.01354/) (2021)
- [Multi-Step Deductive Reasoning Over Natural Language: An Empirical Study on Out-of-Distribution Generalisation](https://mlsanity.com/paper/2207.14000/) (2022)
- [Visual Comparison of Language Model Adaptation](https://mlsanity.com/paper/2208.08176/) (2022)
- [Evaluating Large Language Models with NeuBAROCO: Syllogistic Reasoning Ability and Human-like Biases](https://mlsanity.com/paper/2306.12567/) (2023)
- [Semantic Sensitivities and Inconsistent Predictions: Measuring the Fragility of NLI Models](https://mlsanity.com/paper/2401.14440/) (2024)
- [Exploring Reasoning Biases in Large Language Models Through Syllogism: Insights from the NeuBAROCO Dataset](https://mlsanity.com/paper/2408.04403/) (2024)
- [Probing Commonsense Explanation in Dialogue Response Generation](https://mlsanity.com/paper/2104.09574/) (2021)
- [Quantifying Adaptability in Pre-trained Language Models with 500 Tasks](https://mlsanity.com/paper/2112.03204/) (2021)
- [Scope Ambiguities in Large Language Models](https://mlsanity.com/paper/2404.04332/) (2024)
- [Is My Model Using The Right Evidence? Systematic Probes for Examining Evidence-Based Tabular Reasoning](https://mlsanity.com/paper/2108.00578/) (2021)
- [Neuro-symbolic Natural Logic with Introspective Revision for Natural Language Inference](https://mlsanity.com/paper/2203.04857/) (2022)
- [Negation, Coordination, and Quantifiers in Contextualized Language Models](https://mlsanity.com/paper/2209.07836/) (2022)
- [Trusting RoBERTa over BERT: Insights from CheckListing the Natural Language Inference Task](https://mlsanity.com/paper/2107.07229/) (2021)
- [Can Transformers Reason in Fragments of Natural Language?](https://mlsanity.com/paper/2211.05417/) (2022)
- [SyGNS: A Systematic Generalization Testbed Based on Natural Language Semantics](https://mlsanity.com/paper/2106.01077/) (2021)
- [Dyna-bAbI: unlocking bAbI's potential with dynamic synthetic benchmarking](https://mlsanity.com/paper/2112.00086/) (2021)
- [Probing neural language models for understanding of words of estimative probability](https://mlsanity.com/paper/2211.03358/) (2022)
- [Evaluating Paraphrastic Robustness in Textual Entailment Models](https://mlsanity.com/paper/2306.16722/) (2023)
- [Linguistic Interpretability of Transformer-based Language Models: a systematic review](https://mlsanity.com/paper/2504.08001/) (2025)
- [Analyzing the Effects of Reasoning Types on Cross-Lingual Transfer Performance](https://mlsanity.com/paper/2110.02386/) (2021)

## Abstract (arXiv)

Do state-of-the-art models for language understanding already have, or can they easily learn, abilities such as boolean coordination, quantification, conditionals, comparatives, and monotonicity reasoning (i.e., reasoning about word substitutions in sentential contexts)? While such phenomena are involved in natural language inference (NLI) and go beyond basic linguistic understanding, it is unclear the extent to which they are captured in existing NLI benchmarks and effectively learned by models. To investigate this, we propose the use of semantic fragments---systematically generated datasets that each target a different semantic phenomenon---for probing, and efficiently improving, such capabilities of linguistic models. This approach to creating challenge datasets allows direct control over the semantic diversity and complexity of the targeted linguistic phenomena, and results in a more precise characterization of a model's linguistic behavior. Our experiments, using a library of 8 such semantic fragments, reveal two remarkable findings: (a) State-of-the-art models, including BERT, that are pre-trained on existing NLI benchmark datasets perform poorly on these new fragments, even though the phenomena probed here are central to the NLI task. (b) On the other hand, with only a few minutes of additional fine-tuning---with a carefully selected learning rate and a novel variation of "inoculation"---a BERT-based model can master all of these logic and monotonicity fragments while retaining its performance on established NLI benchmarks.