Diversify Your Datasets: Analyzing Generalization via Controlled Variance in Adversarial Datasets

By Ohad Rozen, Vered Shwartz, Roee Aharoni, Ido Dagan. 2019

39 citations recorded by the source corpus.

Phenomenon-specific “adversarial” datasets have been recently designed to perform targeted stress-tests for particular inference types. Recent work (Liu et al., 2019a) proposed that such datasets can be utilized for training NLI and other types of models, often allowing to learn the phenomenon in focus and improve on the challenge dataset, indicating a “blind spot” in the original training data. Yet, although a model can improve in such a training process, it might still be vulnerable to other challenge datasets targeting the same phenomenon but drawn from a different distribution, such as having a different syntactic complexity level. In this work, we extend this method to drive conclusions about a model’s ability to learn and generalize a target phenomenon rather than to “learn” a dataset, by controlling additional aspects in the adversarial datasets. We demonstrate our approach on two inference phenomena – dative alternation and numerical reasoning, elaborating, and in some cases contradicting, the results of Liu et al.. Our methodology enables building better challenge datasets for creating more robust models, and may yield better model understanding and subsequent overarching improvements.

Subfields: Artificial Intelligence, Materials Chemistry, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Materials Chemistry

In Evaluation & Benchmarks

Citations in the corpus

Teaching Autoregressive Language Models Complex Tasks By Demonstration (2021, background): “In some more constrained domains, such as reasoning about quantitative comparisons (more than, less than, etc.), even relatively small numbers of synthesized training examples have proven useful [23], [24].”
Injecting Numerical Reasoning Skills into Language Models (2020, methodology): “A related line of work has been analyzing the mathematical reasoning abilities of neural models over text (Wallace et al., 2019; Rozen et al., 2019; Ravichander et al., 2019), and on arithmetic problems (Saxton et al., 2019; Amini et al., 2019; Lample and Charton, 2020).”
Dynabench: Rethinking Benchmarking in NLP (2021, background): “Yet, challenge sets may suffer from performance instability (Liu et al., 2019a; Rozen et al., 2019; Zhou et al., 2020) and often lack sufficient statistical power (Card et al.”
The Curse of Performance Instability in Analysis Datasets: Consequences, Source, and Suggestions (2020, background): “Rozen et al. (2019) further explain how to improve challenging datasets and why diversity matters.”
What Does My QA Model Know? Devising Controlled Probes Using Expert Knowledge (2019, background): “This shows the high capacity for transformer models to absorb new tasks with minimal costs, as also observed in Richardson et al. (2020) for NLI.”
ANLIzing the Adversarial Natural Language Inference Dataset (2020, background): “Rozen et al. (2019)’s tagset also has some overlap with our tagset, although none directly.”