United States

The success of language models based on the Transformer architecture appears to be inconsistent with observed anisotropic properties of representations learned by such models. We resolve this by showing, contrary to previous studies, that the representations do not occupy a narrow cone, but rather drift in common directions. At any training step, all the embeddings except for the ground-truth target embedding are updated with gradient in the same direction. Compounded over the training set, the embeddings drift and share common components, manifested in their shape in all the models we have empirically tested. Our experiments show that isotropy can be restored using a simple transformation.

NAACL 2021

Too Much in Common: Shifting of Embeddings in Transformer Language Models and its Implications

2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics.

**Whova App** 
Stay in touch with your fellow conference attendees via the [Whova App](https://whova.com/portal/webapp/nacon_202106/)

**Conference Structure**
https://2021.naacl.org/blog/conference-structure/

**Walkthrough video of how to NAACL 2021** 

Please take a moment to view this video explaining how to navigate the platform, attend sessions network with other attendees. 


<figure class="video_container">
  <iframe src="https://screencast-o-matic.com/watch/crhwbGVh3vx?v=6&ff=1&title=0&controls=1" width=640  height=350  frameborder="0" allowfullscreen="true"> </iframe>
</figure>

<br>
<br>

technical paper

The task of long-form question answering (LFQA) involves retrieving documents relevant to a given question and using them to generate a paragraph-length answer. While many models have recently been proposed for LFQA, we show in this paper that the task formulation raises fundamental challenges regarding evaluation and dataset creation that currently preclude meaningful modeling progress. To demonstrate these challenges, we first design a new system that relies on sparse attention and contrastive retriever learning to achieve state-of-the-art performance on the ELI5 LFQA dataset. While our system tops the public leaderboard, a detailed analysis reveals several troubling trends: (1) our system's generated answers are not actually grounded in the documents that it retrieves; (2) ELI5 contains significant train / validation overlap, as at least 81% of ELI5 validation questions occur in paraphrased form in the training set; (3) ROUGE-L is not an informative metric of generated answer quality and can be easily gamed; and (4) human evaluations used for other text generation tasks are unreliable for LFQA. We offer suggestions to mitigate each of these issues, which we hope will lead to more rigorous LFQA research and meaningful progress in the future.

Hurdles to Progress in Long-form Question Answering

Recent years have seen a flourishing of neural keyphrase generation (KPG) works, including the release of several large-scale datasets and a host of new models to tackle them. Model performance on KPG tasks has increased significantly with evolving deep learning research. However, there lacks a comprehensive comparison among different model designs, and a thorough investigation on related factors that may affect a KPG system’s generalization performance. In this empirical study, we aim to fill this gap by providing extensive experimental results and analyzing the most crucial factors impacting the generalizability of KPG models. We hope this study can help clarify some of the uncertainties surrounding the KPG task and facilitate future research on this topic.

An Empirical Study on Neural Keyphrase Generation

Conditioned dialogue generation suffers from the scarcity of labeled responses. In this work, we exploit labeled non-dialogue text data related to the condition, which are much easier to collect. We propose a multi-task learning approach to leverage both labeled dialogue and text data. The 3 tasks jointly optimize the same pre-trained Transformer -- conditioned dialogue generation task on the labeled dialogue data, conditioned language encoding task and conditioned language generation task on the labeled text data. Experimental results show that our approach outperforms the state-of-the-art models by leveraging the labeled texts, and it also obtains larger improvement in performance comparing to the previous methods to leverage text data.

A Simple and Efficient Multi-Task Learning Approach for Conditioned Dialogue Generation

We consider the problem of learning to simplify medical texts. This is important because most reliable, up-to-date information in biomedicine is dense with jargon and thus practically inaccessible to the lay audience. Furthermore, manual simplification does not scale to the rapidly growing body of biomedical literature, motivating the need for automated approaches. Unfortunately, there are no large-scale resources available for this task. In this work we introduce a new corpus of parallel texts in English comprising technical and lay summaries of all published evidence pertaining to different clinical topics. We then propose a new metric based on likelihood scores from a masked language model pretrained on scientific texts. We show that this automated measure better differentiates between technical and lay summaries than existing heuristics. We introduce and evaluate baseline encoder-decoder Transformer models for simplification and propose a novel augmentation to these in which we explicitly penalize the decoder for producing "jargon" terms; we find that this yields improvements over baselines in terms of readability.

Paragraph-level Simplification of Medical Texts

How can we effectively inform content selection in Transformer-based abstractive summarization models?

Attention Head Masking for Inference Time Content Selection in Abstractive Summarization

Petroni et al. (2019) demonstrated that it is possible to retrieve world facts from a pre-trained language model by expressing them as cloze-style prompts and interpret the model's prediction accuracy as a lower bound on the amount of factual information it encodes. Subsequent work has attempted to tighten the estimate by searching for better prompts, using a disjoint set of facts as training data. In this work, we make two complementary contributions to better understand these factual probing techniques. First, we propose OptiPrompt, a novel and efficient method which directly optimizes in continuous embedding space. We find this simple method is able to predict an additional 6.4% of facts in the LAMA benchmark. Second, we raise a more important question: Can we really interpret these probing results as a lower bound? Is it possible that these prompt-search methods learn from the training data too? We find, somewhat surprisingly, that the training data used by these methods contains certain regularities of the underlying fact distribution, and all the existing prompt methods, including ours, are able to exploit them for better fact prediction. We conduct a set of control experiments to disentangle "learning" from "learning to recall", providing a more detailed picture of what different prompts can reveal about pre-trained language models.

Factual Probing Is [MASK]: Learning vs. Learning to Recall

Saliency methods are widely used to interpret neural network predictions, but different variants of saliency methods often disagree even on the interpretations of the same prediction made by the same model. In these cases, how do we identify when are these interpretations trustworthy enough to be used in analyses? To address this question, we conduct a comprehensive and quantitative evaluation of saliency methods on a fundamental category of NLP models: neural language models. We evaluate the quality of prediction interpretations from two perspectives that each represents a desirable property of these interpretations: plausibility and faithfulness. Our evaluation is conducted on four different datasets constructed from the existing human annotation of syntactic and semantic agreements, on both sentence-level and document-level. Through our evaluation, we identified various ways saliency methods could yield interpretations of low quality. We recommend that future work deploying such methods to neural language models should carefully validate their interpretations before drawing insights.

Evaluating Saliency Methods for Neural Language Models

Adversarial examples expose the vulnerabilities of natural language processing (NLP) models, and can be used to evaluate and improve their robustness. Existing techniques of generating such examples are typically driven by local heuristic rules that are agnostic to the context, often resulting in unnatural and ungrammatical outputs. This paper presents CLARE, a ContextuaLized AdversaRial Example generation model that produces fluent and grammatical outputs through a mask-then-infill procedure. CLARE builds on a pre-trained masked language model and modifies the inputs in a context-aware manner. We propose three contextualized perturbations, Replace, Insert and Merge, that allow for generating outputs of varied lengths. CLARE can flexibly combine these perturbations and apply them at any position in the inputs, and is thus able to attack the victim model more effectively with fewer edits. Extensive experiments and human evaluation demonstrate that CLARE outperforms the baselines in terms of attack success rate, textual similarity, fluency and grammaticality.

Contextualized Perturbation for Textual Adversarial Attack

Understanding how the linguistic structure is encoded in contextualized embedding could help explain their impressive performance across NLP. Existing approaches for probing them usually call for training classifiers and use the accuracy, mutual information, or complexity as a proxy for the representation's goodness. In this work, we argue that doing so can be unreliable because different representations may need different classifiers. We develop a heuristic, DirectProbe, that directly studies the geometry of a representation by building upon the notion of a version space for a task. Experiments with several linguistic tasks and contextualized embeddings show that, even without training classifiers, DirectProbe can shine lights on how an embedding space represents labels and also anticipate the classifier performance for the representation.

DirectProbe: Studying Representations without Classifiers

Transfer learning that adapts a model trained on data-rich sources to low-resource targets has been widely applied in natural language processing (NLP). However, when training a transfer model over multiple sources, not every source is equally useful for the target. To better transfer a model, it is essential to understand the values of the sources. In this paper, we develop SEAL-Shap, an efficient source valuation framework for quantifying the usefulness of the sources (e.g., domains/languages) in transfer learning based on the Shapley value method. Experiments and comprehensive analyses on both cross-domain and cross-lingual transfers demonstrate that our framework is not only effective in choosing useful transfer sources but also the source values match the intuitive source-target similarity.

Too Much in Common: Shifting of Embeddings in Transformer Language Models and its Implications

Downloads

Next from NAACL 2021

Hurdles to Progress in Long-form Question Answering

Similar lecture

QMSum: A New Benchmark for Query-based Multi-domain Meeting Summarization

Too Much in Common: Shifting of Embeddings in Transformer Language Models and its Implications

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from NAACL 2021

Hurdles to Progress in Long-form Question Answering

Similar lecture

QMSum: A New Benchmark for Query-based Multi-domain Meeting Summarization

Downloads