Self-training is one promising solution for it, which struggles to construct a set of high-quality pseudo training instances for the target domain.
Thanks to the strong representation learning capability of deep learning, especially pre-training techniques with language model loss, dependency parsing has achieved great performance boost in the in-domain scenario with abundant labeled training data for target domains.
Opinion Role Labeling (ORL), aiming to identify the key roles of opinion, has received increasing interest.
Pre-trained language models (PLMs) have shown great potentials in natural language processing (NLP) including rhetorical structure theory (RST) discourse parsing. Current PLMs are obtained by sentence-level pre-training, which is different from the basic processing unit, i. e. element discourse unit (EDU). To this end, we propose a second-stage EDU-level pre-training approach in this work, which presents two novel tasks to learn effective EDU representations continually based on well pre-trained language models. Concretely, the two tasks are (1) next EDU prediction (NEP) and (2) discourse marker prediction (DMP). We take a state-of-the-art transition-based neural parser as baseline, and adopt it with a light bi-gram EDU modification to effectively explore the EDU-level pre-trained EDU representation. Experimental results on a benckmark dataset show that our method is highly effective, leading a 2. 1-point improvement in F1-score. All codes and pre-trained models will be released publicly to facilitate future studies.
Recent advances of multilingual word representations weaken the input divergences across languages, making cross-lingual transfer similar to the monolingual cross-domain and semi-supervised settings.
Text ranking is a critical task in various information retrieval applications, and the recent success of Large Language Models (LLMs) in natural language processing has sparked interest in their application to text ranking.
Our experimental results demonstrate that self-training for constituency parsing, equipped with an LLM, outperforms traditional methods regardless of the LLM's performance.
As such cases span from English to other natural or programming languages, from retrieval to classification and beyond, it is desirable to build a unified embedding model rather than dedicated ones for each scenario.
To address this problem, we propose a novel Grounded Entity-Landmark Adaptive (GELA) pre-training paradigm for VLN tasks.
A scene-event mapping mechanism is first designed to bridge the gap between the underlying scene structure and the high-level event semantic structure, resulting in an overall hierarchical scene-event (termed ICE) graph structure.
We present GTE, a general-purpose text embedding model trained with multi-stage contrastive learning.
Structured Natural Language Processing (XNLP) is an important subset of NLP that entails understanding the underlying semantic or syntactic structure of texts, which serves as a foundational component for many downstream applications.
In this paper, we conduct a holistic exploration of the Universal Decompositional Semantic (UDS) Parsing.
Dialogue-level dependency parsing has received insufficient attention, especially for Chinese.
Latest efforts on cross-lingual relation extraction (XRE) aggressively leverage the language-consistent structural features from the universal dependency (UD) resource, while they may largely suffer from biased transfer (e. g., either target-biased or source-biased) due to the inevitable linguistic disparity between languages.
In this work, we investigate a more realistic unsupervised multimodal machine translation (UMMT) setup, inference-time image-free UMMT, where the model is trained with source-text image pairs, and tested with only source-text inputs.
With an external 3D scene extractor, we obtain the 3D objects and scene features for input images, based on which we construct a target object-centered 3D spatial scene graph (Go3D-S2G), such that we model the spatial semantics of target objects within the holistic 3D scenes.
In this study, we propose to enhance the ABSA robustness by systematically rethinking the bottlenecks from all possible angles, including model, data, and training.
Universally modeling all typical information extraction tasks (UIE) with one generative language model (GLM) has revealed great potential by the latest study, where various IE predictions are unified into a linearized hierarchical expression under a GLM.
Zero-shot information extraction (IE) aims to build IE systems from the unannotated text.
Preliminary experiments on En-Zh and En-Ja news domain corpora demonstrate that monolingual data can significantly improve translation quality (e. g., +3. 15 BLEU on En-Zh).
We apply BABERT for feature induction of Chinese sequence labeling tasks.
Ranked #1 on Chinese Word Segmentation on MSRA
Accordingly, we manually annotate a dataset to facilitate the investigation of the newly-introduced task and build several benchmark encoder-decoder models by using VL-BART and VL-T5 as backbones.
In this work, we investigate the integration of a latent graph for CSRL.
Successful Machine Learning based Named Entity Recognition models could fail on texts from some special domains, for instance, Chinese addresses and e-commerce titles, where requires adequate background knowledge.
Prior research has mainly resorted to heuristic rule-based constraints to reduce the noise for specific self-augmentation methods individually.
Recent works of opinion expression identification (OEI) rely heavily on the quality and scale of the manually-constructed training corpus, which could be extremely difficult to satisfy.
Named Entity Recognition (NER) from speech is among Spoken Language Understanding (SLU) tasks, aiming to extract semantic information from the speech signal.
So far, named entity recognition (NER) has been involved with three major types, including flat, overlapped (aka.
Ranked #2 on Chinese Named Entity Recognition on OntoNotes 4
Unified opinion role labeling (ORL) aims to detect all possible opinion structures of 'opinion-holder-target' in one shot, given a text.
Ranked #1 on Fine-Grained Opinion Analysis on MPQA (F1 (Opinion) metric)
The three subtasks are closely related while previous studies model them individually, which ignores their intern connections and meanwhile induces error propagation problem.
Second, we perform relation classification to judge whether a given pair of entity fragments to be overlapping or succession.
Crowdsourcing is regarded as one prospective solution for effective supervised learning, aiming to build large-scale annotated training data by crowd workers.
It performs the two subtasks of SRL: predicate identification and argument role labeling, jointly.
Ranked #1 on Semantic Role Labeling on CoNLL-2009
Emotion detection in conversations (EDC) is to detect the emotion for each utterance in conversations that have multiple speakers.
Ranked #17 on Emotion Recognition in Conversation on EmoryNLP
In this paper, we fill the gap of cross-lingual SRL by proposing an end-to-end SRL model that incorporates a variety of universal features and transfer methods.
This article briefly reviews the representative models of constituent parsing and dependency parsing, and also dependency graph parsing with rich semantics.
Discourse representation tree structure (DRTS) parsing is a novel semantic parsing task which has been concerned most recently.
Many efforts of research are devoted to semantic role labeling (SRL) which is crucial for natural language understanding.
In this paper, we present Chinese lexical fusion recognition, a new task which could be regarded as one kind of coreference recognition.
Conversational emotion recognition (CER) has attracted increasing interests in the natural language processing (NLP) community.
Ranked #20 on Emotion Recognition in Conversation on EmoryNLP
Semantic role labeling (SRL), also known as shallow semantic parsing, is an important yet challenging task in NLP.
Opinion role labeling (ORL) is an important task for fine-grained opinion mining, which identifies important opinion arguments such as holder and target for a given opinion trigger.
Ranked #1 on Fine-Grained Opinion Analysis on MPQA (using extra training data)
Syntax has been demonstrated highly effective in neural machine translation (NMT).
Ranked #8 on Machine Translation on IWSLT2015 English-Vietnamese
To quickly obtain new labeled data, we can choose crowdsourcing as an alternative way at lower cost in a short time.
First, we encode the context of entities on a dependency tree as sentence-level entity embedding based on tree-GRU.
In this paper, we model the problem of disfluency detection using a transition-based framework, which incrementally constructs and labels the disfluency chunk of input sentences using a new transition system without syntax information.
Neural networks have shown promising results for relation extraction.
Ranked #1 on Relation Extraction on ACE 2005 (Sentence Encoder metric)
Our results on standard benchmarks show that state-of-the-art neural models can give accuracies comparable to the best discrete models in the literature for most tasks and combing discrete and neural features unanimously yield better results.
We evaluate our models on two datasets for recognizing regular and irreg- ular biomedical entities.
While part-of-speech (POS) tagging and dependency parsing are observed to be closely related, existing work on joint modeling with manually crafted feature templates suffers from the feature sparsity and incompleteness problems.
We investigate the use of neural network for tweet sarcasm detection, and compare the effects of the continuous automatic features with discrete manual features.
This model divides a sentence or text segment into five parts, namely two target entities and their three contexts.