Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userExposing Length Divergence Bias of Textual Matching Models
107
0
0.0
(
0
)
Added by
Lan Jiang
Publication date
2021
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
Despite the remarkable success deep models have achieved in Textual Matching (TM), their robustness issue is still a topic of concern. In this work, we propose a new perspective to study this issue -- via the length divergence bias of TM models. We conclude that this bias stems from two parts: the label bias of existing TM datasets and the sensitivity of TM models to superficial information. We critically examine widely used TM datasets, and find that all of them follow specific length divergence distributions by labels, providing direct cues for predictions. As for the TM models, we conduct adversarial evaluation and show that all models performances drop on the out-of-distribution adversarial test sets we construct, which demonstrates that they are all misled by biased training sets. This is also confirmed by the textit{SentLen} probing task that all models capture rich length information during training to facilitate their performances. Finally, to alleviate the length divergence bias in TM models, we propose a practical adversarial training method using bias-free training data. Our experiments indicate that we successfully improve the robustness and generalization ability of models at the same time.
rate research
Read More
Adversarial attacks have shown the vulnerability of machine learning models, however, it is non-trivial to conduct textual adversarial attacks on natural language processing tasks due to the discreteness of data. Most previous approaches conduct attacks with the atomic textit{replacement} operation, which usually leads to fixed-length adversarial examples and therefore limits the exploration on the decision space. In this paper, we propose variable-length textual adversarial attacks~(VL-Attack) and integrate three atomic operations, namely textit{insertion}, textit{deletion} and textit{replacement}, into a unified framework, by introducing and manipulating a special textit{blank} token while attacking. In this way, our approach is able to more comprehensively find adversarial examples around the decision boundary and effectively conduct adversarial attacks. Specifically, our method drops the accuracy of IMDB classification by $96%$ with only editing $1.3%$ tokens while attacking a pre-trained BERT model. In addition, fine-tuning the victim model with generated adversarial samples can improve the robustness of the model without hurting the performance, especially for length-sensitive models. On the task of non-autoregressive machine translation, our method can achieve $33.18$ BLEU score on IWSLT14 German-English translation, achieving an improvement of $1.47$ over the baseline model.
Neural dialog models are known to suffer from problems such as generating unsafe and inconsistent responses. Even though these problems are crucial and prevalent, they are mostly manually identified by model designers through interactions. Recently, some research instructs crowdworkers to goad the bots into triggering such problems. However, humans leverage superficial clues such as hate speech, while leaving systematic problems undercover. In this paper, we propose two methods including reinforcement learning to automatically trigger a dialog model into generating problematic responses. We show the effect of our methods in exposing safety and contradiction issues with state-of-the-art dialog models.
We study two problems in neural machine translation (NMT). First, in beam search, whereas a wider beam should in principle help translation, it often hurts NMT. Second, NMT has a tendency to produce translations that are too short. Here, we argue that these problems are closely related and both rooted in label bias. We show that correcting the brevity problem almost eliminates the beam problem; we compare some commonly-used methods for doing this, finding that a simple per-word reward works well; and we introduce a simple and quick way to tune this reward using the perceptron algorithm.
Medical systems in general, and patient treatment decisions and outcomes in particular, are affected by bias based on gender and other demographic elements. As language models are increasingly applied to medicine, there is a growing interest in building algorithmic fairness into processes impacting patient care. Much of the work addressing this question has focused on biases encoded in language models -- statistical estimates of the relationships between concepts derived from distant reading of corpora. Building on this work, we investigate how word choices made by healthcare practitioners and language models interact with regards to bias. We identify and remove gendered language from two clinical-note datasets and describe a new debiasing procedure using BERT-based gender classifiers. We show minimal degradation in health condition classification tasks for low- to medium-levels of bias removal via data augmentation. Finally, we compare the bias semantically encoded in the language models with the bias empirically observed in health records. This work outlines an interpretable approach for using data augmentation to identify and reduce the potential for bias in natural language processing pipelines.
Unintended biases in machine learning (ML) models are among the major concerns that must be addressed to maintain public trust in ML. In this paper, we address process fairness of ML models that consists in reducing the dependence of models on sensitive features, without compromising their performance. We revisit the framework FixOut that is inspired in the approach fairness through unawareness to build fairer models. We introduce several improvements such as automating the choice of FixOuts parameters. Also, FixOut was originally proposed to improve fairness of ML models on tabular data. We also demonstrate the feasibility of FixOuts workflow for models on textual data. We present several experimental results that illustrate the fact that FixOut improves process fairness on different classification settings.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
