Multi-View Domain Adapted Sentence Embeddings for Low-Resource Unsupervised Duplicate Question Detection

Abstract

We address the problem of Duplicate Question Detection (DQD) in low-resource domainspecific Community Question Answering forums. Our multi-view framework MV-DASE combines an ensemble of sentence encoders via Generalized Canonical Correlation Analysis, using unlabeled data only. In our experiments, the ensemble includes generic and domain-specific averaged word embeddings, domain-finetuned BERT and the Universal Sentence Encoder. We evaluate MV-DASE on the CQADupStack corpus and on additional low-resource Stack Exchange forums. Combining the strengths of different encoders, we significantly outperform BM25, all singleview systems as well as a recent supervised domain-adversarial DQD method.

MultiCQA: Zero-Shot Transfer of Self-Supervised Text Matching Models on a Massive Scale

“…Current models seemingly match similar keywords or phrases of the questions and answers, often without truly understanding them in context. (Rücklé et al, 2019b), ‡ is the MICRON model (Han et al, 2019), is the BERT model in (Ma et al, 2019), and is MV-DASE (Poerner and Schütze, 2019). Table 5: A mistake of MultiCQA RBa-lg (zero-shot transfer) on AskUbuntu.…”

“…Shah et al (2018) use adversarial domain adaptation for duplicate question detection. Poerner and Schütze (2019) adapt the combination of different sentence embeddings to individual target domains. Rücklé et al (2019b) use weakly supervised training, self-supervised training methods, and question generation.…”

“…In contrast to other text matching tasks in NLP, they compare texts of different lengths-e.g., answers can be long explanations or descriptions-and often deal with expert domains. This makes it difficult to transfer models across domains (Shah et al, 2018), and to apply common approaches such as universal sentence embeddings without further domain or task adaptations (Poerner and Schütze, 2019).…”

“…Reasons are that (1) there exist a large number of domains, and (2) in-domain training data is often scarce. Previous work proposed domain adaptation techniques (Poerner and Schütze, 2019;Shah et al, 2018), training with unlabeled data (Rücklé et al, 2019b), and shallow architectures (Rücklé et al, 2019a). However, these approaches result in entirely separate models that are specialized to individual target domains.…”

Learning from Unlabelled Data for Clinical Semantic Textual Similarity

“…Semantic textual similarity (STS) measures the degree of semantic equivalence between two text snippets, based on a graded numerical value, with applications including question answering (Yadav et al, 2020), duplicate detection (Poerner and Schütze, 2019), and entity linking (Zhou et al, 2020).…”