Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userInfluence Tuning: Demoting Spurious Correlations via Instance Attribution and Instance-Driven Updates
التأثير على ضبط: إزالة الارتباطات الزائفة عبر الإسناد المثيل وتحديثات مدفوعة بالمثيل
210
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
من بين القيود الأكثر أهمية في نماذج NLP التعليمية العميقة هي عدم قابلية الترجمة الشفوية، واعتمادها على الارتباطات الزائفة.اقترح العمل السابق مناهج مختلفة لتفسير نماذج الصندوق الأسود للكشف عن الارتباطات الزائفة، ولكن تم استخدام البحث في المقام الأول في سيناريوهات تفاعل الكمبيوتر البشري.لا يزال لا يزال غير متوقع سواء أم لا يمكن استخدام تفسيرات النمطية أو كيفية استخدامها تلقائيا "ميزات مربكة".في هذا العمل، نقترح التأثير على الضبط --- الإجراء الذي يرفع تفسيرات نموذجية لتحديث معلمات النموذج نحو تفسير معقول (بدلا من التفسير الذي يعتمد على الأنماط الزائفة في البيانات) بالإضافة إلى تعلم التنبؤ بالملصقات المهمة.نظرا لأنه في برنامج إعداد خاضع للرقابة، يمكن أن يساعد التأثير على الضبط في إزالة النموذج من الأنماط الزائفة في البيانات، مما يتفوق بشكل كبير على أساليب خط الأساس التي تستخدم التدريب الخصم.
Among the most critical limitations of deep learning NLP models are their lack of interpretability, and their reliance on spurious correlations. Prior work proposed various approaches to interpreting the black-box models to unveil the spurious correlations, but the research was primarily used in human-computer interaction scenarios. It still remains underexplored whether or how such model interpretations can be used to automatically unlearn'' confounding features. In this work, we propose influence tuning---a procedure that leverages model interpretations to update the model parameters towards a plausible interpretation (rather than an interpretation that relies on spurious patterns in the data) in addition to learning to predict the task labels. We show that in a controlled setup, influence tuning can help deconfounding the model from spurious patterns in data, significantly outperforming baseline methods that use adversarial training.
References used
https://aclanthology.org/
rate research
Read More
Recent work in cross-topic argument mining attempts to learn models that generalise across topics rather than merely relying on within-topic spurious correlations. We examine the effectiveness of this approach by analysing the output of single-task a
nd multi-task models for cross-topic argument mining, through a combination of linear approximations of their decision boundaries, manual feature grouping, challenge examples, and ablations across the input vocabulary. Surprisingly, we show that cross-topic models still rely mostly on spurious correlations and only generalise within closely related topics, e.g., a model trained only on closed-class words and a few common open-class words outperforms a state-of-the-art cross-topic model on distant target topics.
Abstract Interpretable rationales for model predictions are crucial in practical applications. We develop neural models that possess an interpretable inference process for dependency parsing. Our models adopt instance-based inference, where dependenc
y edges are extracted and labeled by comparing them to edges in a training set. The training edges are explicitly used for the predictions; thus, it is easy to grasp the contribution of each edge to the predictions. Our experiments show that our instance-based models achieve competitive accuracy with standard neural models and have the reasonable plausibility of instance-based explanations.
Existing works on information extraction (IE) have mainly solved the four main tasks separately (entity mention recognition, relation extraction, event trigger detection, and argument extraction), thus failing to benefit from inter-dependencies betwe
en tasks. This paper presents a novel deep learning model to simultaneously solve the four tasks of IE in a single model (called FourIE). Compared to few prior work on jointly performing four IE tasks, FourIE features two novel contributions to capture inter-dependencies between tasks. First, at the representation level, we introduce an interaction graph between instances of the four tasks that is used to enrich the prediction representation for one instance with those from related instances of other tasks. Second, at the label level, we propose a dependency graph for the information types in the four IE tasks that captures the connections between the types expressed in an input sentence. A new regularization mechanism is introduced to enforce the consistency between the golden and predicted type dependency graphs to improve representation learning. We show that the proposed model achieves the state-of-the-art performance for joint IE on both monolingual and multilingual learning settings with three different languages.
Data-to-text (D2T) generation in the biomedical domain is a promising - yet mostly unexplored - field of research. Here, we apply neural models for D2T generation to a real-world dataset consisting of package leaflets of European medicines. We show t
hat fine-tuned transformers are able to generate realistic, multi-sentence text from data in the biomedical domain, yet have important limitations. We also release a new dataset (BioLeaflets) for benchmarking D2T generation models in the biomedical domain.
A private learning scheme TextHide was recently proposed to protect the private text data during the training phase via so-called instance encoding. We propose a novel reconstruction attack to break TextHide by recovering the private training data, a
nd thus unveil the privacy risks of instance encoding. We have experimentally validated the effectiveness of the reconstruction attack with two commonly-used datasets for sentence classification. Our attack would advance the development of privacy preserving machine learning in the context of natural language processing.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
