اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدMethods for the Design and Evaluation of HCI+NLP Systems
75
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
HCI and NLP traditionally focus on different evaluation methods. While HCI involves a small number of people directly and deeply, NLP traditionally relies on standardized benchmark evaluations that involve a larger number of people indirectly. We present five methodological proposals at the intersection of HCI and NLP and situate them in the context of ML-based NLP models. Our goal is to foster interdisciplinary collaboration and progress in both fields by emphasizing what the fields can learn from each other.
قيم البحث
اقرأ أيضاً
Open-domain human-computer conversation has been attracting increasing attention over the past few years. However, there does not exist a standard automatic evaluation metric for open-domain dialog systems; researchers usually resort to human annotat
ion for model evaluation, which is time- and labor-intensive. In this paper, we propose RUBER, a Referenced metric and Unreferenced metric Blended Evaluation Routine, which evaluates a reply by taking into consideration both a groundtruth reply and a query (previous user-issued utterance). Our metric is learnable, but its training does not require labels of human satisfaction. Hence, RUBER is flexible and extensible to different datasets and languages. Experiments on both retrieval and generative dialog systems show that RUBER has a high correlation with human annotation.
Code-switching is the use of more than one language in the same conversation or utterance. Recently, multilingual contextual embedding models, trained on multiple monolingual corpora, have shown promising results on cross-lingual and multilingual tas
ks. We present an evaluation benchmark, GLUECoS, for code-switched languages, that spans several NLP tasks in English-Hindi and English-Spanish. Specifically, our evaluation benchmark includes Language Identification from text, POS tagging, Named Entity Recognition, Sentiment Analysis, Question Answering and a new task for code-switching, Natural Language Inference. We present results on all these tasks using cross-lingual word embedding models and multilingual models. In addition, we fine-tune multilingual models on artificially generated code-switched data. Although multilingual models perform significantly better than cross-lingual models, our results show that in most tasks, across both language pairs, multilingual models fine-tuned on code-switched data perform best, showing that multilingual models can be further optimized for code-switching tasks.
Widespread adoption of deep models has motivated a pressing need for approaches to interpret network outputs and to facilitate model debugging. Instance attribution methods constitute one means of accomplishing these goals by retrieving training inst
ances that (may have) led to a particular prediction. Influence functions (IF; Koh and Liang 2017) provide machinery for doing this by quantifying the effect that perturbing individual train instances would have on a specific test prediction. However, even approximating the IF is computationally expensive, to the degree that may be prohibitive in many cases. Might simpler approaches (e.g., retrieving train examples most similar to a given test point) perform comparably? In this work, we evaluate the degree to which different potential instance attribution agree with respect to the importance of training samples. We find that simple retrieval methods yield training instances that differ from those identified via gradient-based methods (such as IFs), but that nonetheless exhibit desirable characteristics similar to more complex attribution methods. Code for all methods and experiments in this paper is available at: https://github.com/successar/instance_attributions_NLP.
Transformers are the dominant architecture in NLP, but their training and fine-tuning is still very challenging. In this paper, we present the design and implementation of a visual analytic framework for assisting researchers in such process, by prov
iding them with valuable insights about the models intrinsic properties and behaviours. Our framework offers an intuitive overview that allows the user to explore different facets of the model (e.g., hidden states, attention) through interactive visualization, and allows a suite of built-in algorithms that compute the importance of model components and different parts of the input sequence. Case studies and feedback from a user focus group indicate that the framework is useful, and suggest several improvements.
Faceted navigation is a very useful component in todays search engines. It is especially useful when user has an exploratory information need or prefer certain attribute values than others. Existing work has tried to optimize faceted systems in many
aspects, but little work has been done on optimizing numerical facet ranges (e.g., price ranges of product). In this paper, we introduce for the first time the research problem on numerical facet range partition and formally frame it as an optimization problem. To enable quantitative evaluation of a partition algorithm, we propose an evaluation metric to be applied to search engine logs. We further propose two range partition algorithms that computationally optimize the defined metric. Experimental results on a two-month search log from a major e-Commerce engine show that our proposed method can significantly outperform baseline.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
