A Span Selection Model for Semantic Role Labeling

Ouchi, Hiroki; Shindo, Haruo; Matsumoto, Yūji

doi:10.18653/v1/d18-1191

Cited by 85 publications

(76 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For syntactic chunking, we use a variant of the Reconciled Span Parser (Joshi et al, 2018). For SRL, we use the span selection model (BiLSTM-Span model) (Ouchi et al, 2018). Each model is trained on a source domain training set and was evaluated on a target domain test set 6 .…”

Section: Methodsmentioning

confidence: 99%

Transductive Learning of Neural Language Models for Syntactic and Semantic Analysis

Ouchi

Suzuki

Inui

2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

Self Cite

View full text Add to dashboard Cite

In transductive learning, an unlabeled test set is used for model training. While this setting deviates from the common assumption of a completely unseen test set, it is applicable in many real-world scenarios, where the texts to be processed are known in advance. However, despite its practical advantages, transductive learning is underexplored in natural language processing. Here, we conduct an empirical study of transductive learning for neural models and demonstrate its utility in syntactic and semantic tasks. Specifically, we fine-tune language models (LMs) on an unlabeled test set to obtain test-set-specific word representations. Through extensive experiments, we demonstrate that despite its simplicity, transductive LM fine-tuning consistently improves state-ofthe-art neural models in both in-domain and out-of-domain settings.

show abstract

Section: Methodsmentioning

confidence: 99%

Transductive Learning of Neural Language Models for Syntactic and Semantic Analysis

Ouchi

Suzuki

Inui

2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

Self Cite

View full text Add to dashboard Cite

show abstract

“…Semantic Role Labeling. Span-based SRL only exists on English data (Zhou and Xu, 2015;Strubell et al, 2018;Ouchi et al, 2018). Dependency-based SRL models such as Cai et al, 2018;Li et al, 2019) are the state-of-the-art for English.…”

Section: Related Workmentioning

confidence: 99%

Translate and Label! An Encoder-Decoder Approach for Cross-lingual Semantic Role Labeling

Daza

Frank

2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

View full text Add to dashboard Cite

We propose a Cross-lingual Encoder-Decoder model that simultaneously translates and generates sentences with Semantic Role Labeling annotations in a resource-poor target language. Unlike annotation projection techniques, our model does not need parallel data during inference time. Our approach can be applied in monolingual, multilingual and cross-lingual settings and is able to produce dependencybased and span-based SRL annotations. We benchmark the labeling performance of our model in different monolingual and multilingual settings using well-known SRL datasets. We then train our model in a cross-lingual setting to generate new SRL labeled data. Finally, we measure the effectiveness of our method by using the generated data to augment the training basis for resource-poor languages and perform manual evaluation to show that it produces high-quality sentences and assigns accurate semantic role annotations. Our proposed architecture offers a flexible method for leveraging SRL data in multiple languages.

show abstract

“…Instead, suppose we had a function g(y, L x ) → R + that measures a loss between output y and a grammar L x such that g(y, L x ) = 0 if and only if there are no grammatical errors in y. That is, g(y, L x ) = 0 for the feasible region 1 Since our submission, the previous SOTA (Peters et al 2018) in SRL on which we apply our technique has been advanced by 1.7 F1 points (Ouchi, Shindo, and Matsumoto 2018). However, this is a training time improvement which is orthogonal to our work.…”

Section: Problem Definition and Motivationmentioning

confidence: 99%

Gradient-Based Inference for Networks with Output Constraints

Lee

Mehta

Wick

et al. 2019

AAAI

View full text Add to dashboard Cite

Practitioners apply neural networks to increasingly complex problems in natural language processing, such as syntactic parsing and semantic role labeling that have rich output structures. Many such structured-prediction problems require deterministic constraints on the output values; for example, in sequence-to-sequence syntactic parsing, we require that the sequential outputs encode valid trees. While hidden units might capture such properties, the network is not always able to learn such constraints from the training data alone, and practitioners must then resort to post-processing. In this paper, we present an inference method for neural networks that enforces deterministic constraints on outputs without performing rule-based post-processing or expensive discrete search. Instead, in the spirit of gradient-based training, we enforce constraints with gradient-based inference (GBI): for each input at test-time, we nudge continuous model weights until the network's unconstrained inference procedure generates an output that satisfies the constraints. We study the efficacy of GBI on three tasks with hard constraints: semantic role labeling, syntactic parsing, and sequence transduction. In each case, the algorithm not only satisfies constraints, but improves accuracy, even when the underlying network is stateof-the-art.

show abstract

A Span Selection Model for Semantic Role Labeling

Cited by 85 publications

References 39 publications

Transductive Learning of Neural Language Models for Syntactic and Semantic Analysis

Transductive Learning of Neural Language Models for Syntactic and Semantic Analysis

Translate and Label! An Encoder-Decoder Approach for Cross-lingual Semantic Role Labeling

Gradient-Based Inference for Networks with Output Constraints

Contact Info

Product

Resources

About