1 |
Comparative Error Analysis in Neural and Finite-state Models for Unsupervised Character-level Transduction ...
|
|
|
|
BASE
|
|
Show details
|
|
2 |
Could you give me a hint? Generating inference graphs for defeasible reasoning ...
|
|
|
|
BASE
|
|
Show details
|
|
3 |
Comparative Error Analysis in Neural and Finite-state Models for Unsupervised Character-level Transduction ...
|
|
|
|
BASE
|
|
Show details
|
|
4 |
Could you give me a hint ? Generating inference graphs for defeasible reasoning ...
|
|
|
|
BASE
|
|
Show details
|
|
5 |
Investigating Robustness of Dialog Models to Popular Figurative Language Constructs ...
|
|
|
|
BASE
|
|
Show details
|
|
6 |
Measuring and Improving Consistency in Pretrained Language Models ...
|
|
|
|
BASE
|
|
Show details
|
|
7 |
More Identifiable yet Equally Performant Transformers for Text Classification ...
|
|
|
|
BASE
|
|
Show details
|
|
8 |
Improving Automated Evaluation of Open Domain Dialog via Diverse Reference Augmentation ...
|
|
|
|
BASE
|
|
Show details
|
|
9 |
Style is NOT a single variable: Case Studies for Cross-Stylistic Language Understanding ...
|
|
|
|
BASE
|
|
Show details
|
|
10 |
SelfExplain: A Self-Explaining Architecture for Neural Text Classifiers ...
|
|
|
|
BASE
|
|
Show details
|
|
11 |
Classifying Argumentative Relations Using Logical Mechanisms and Argumentation Schemes ...
|
|
|
|
BASE
|
|
Show details
|
|
12 |
StylePTB: A Compositional Benchmark for Fine-grained Controllable Text Style Transfer ...
|
|
|
|
BASE
|
|
Show details
|
|
13 |
StylePTB: A Compositional Benchmark for Fine-grained Controllable Text Style Transfer ...
|
|
|
|
BASE
|
|
Show details
|
|
14 |
Classifying Argumentative Relations Using Logical Mechanisms and Argumentation Schemes ...
|
|
|
|
BASE
|
|
Show details
|
|
15 |
Extracting Implicitly Asserted Propositions in Argumentation ...
|
|
|
|
BASE
|
|
Show details
|
|
16 |
Probing the Probing Paradigm: Does Probing Accuracy Entail Task Relevance? ...
|
|
|
|
Abstract:
Although neural models have achieved impressive results on several NLP benchmarks, little is understood about the mechanisms they use to perform language tasks. Thus, much recent attention has been devoted to analyzing the sentence representations learned by neural encoders, through the lens of `probing' tasks. However, to what extent was the information encoded in sentence representations, as discovered through a probe, actually used by the model to perform its task? In this work, we examine this probing paradigm through a case study in Natural Language Inference, showing that models can learn to encode linguistic properties even if they are not needed for the task on which the model was trained. We further identify that pretrained word embeddings play a considerable role in encoding these properties rather than the training task itself, highlighting the importance of careful controls when designing probing experiments. Finally, through a set of controlled synthetic tasks, we demonstrate models can encode ... : EACL 2021 ...
|
|
Keyword:
Computation and Language cs.CL; FOS Computer and information sciences
|
|
URL: https://dx.doi.org/10.48550/arxiv.2005.00719 https://arxiv.org/abs/2005.00719
|
|
BASE
|
|
Hide details
|
|
18 |
On the Systematicity of Probing Contextualized Word Representations: The Case of Hypernymy in BERT ...
|
|
|
|
BASE
|
|
Show details
|
|
19 |
On aligning OpenIE extractions with Knowledge Bases: A case study
|
|
|
|
BASE
|
|
Show details
|
|
20 |
Discourse in Multimedia: A Case Study in Extracting Geometry Knowledge from Textbooks
|
|
|
|
In: Computational Linguistics, Vol 45, Iss 4, Pp 627-665 (2020) (2020)
|
|
BASE
|
|
Show details
|
|
|
|