This paper describes the results of a study designed to assess human expert ratings of educational concept features for use in automatic core concept extraction systems. Digital library resources provided the content base for human experts to annotate automatically extracted concepts on seven dimensions: coreness, local importance, topic, content, phrasing, structure, and function. The annotated concepts were used as training data to build a machine learning classifier as part of a tool used to predict the core concepts in the document. These predictions were compared with the experts' judgment of concept coreness.

Meaning Representations (AMRs) are broad-coverage sentence-level semantic graphs. Existing approaches to generating text from AMR have focused on training sequenceto-sequence or graph-to-sequence models on AMR annotated data only. In this paper, we propose an alternative approach that combines a strong pre-trained language model with cycle consistency-based re-scoring. Despite the simplicity of the approach, our experimental results show these models outperform all previous techniques on the English LDC2017T10 dataset, including the recent use of transformer architectures. In addition to the standard evaluation metrics, we provide human evaluation experiments that further substantiate the strength of our approach.

We present DR.DECR (Dense Retrieval with Distillation-Enhanced Cross-Lingual Representation), a new cross-lingual information retrieval (CLIR) system trained using multi-stage knowledge distillation (KD). The teacher of DR.DECR relies on a highly effective but computationally expensive two-stage inference process consisting of query translation and monolingual IR, while the student, DR.DECR, executes a single CLIR step. We teach DR.DECR powerful multilingual representations as well as CLIR by optimizing two corresponding KD objectives. Learning useful representations of non-English text from an English-only retriever is accomplished through a cross-lingual token alignment algorithm that relies on the representation capabilities of the underlying multilingual encoders. In both in-domain and zero-shot out-of-domain evaluation, DR.DECR demonstrates far superior accuracy over direct fine-tuning with labeled CLIR data. It is also the best single-model retriever on the XOR-TyDi benchmark at the time of this writing.

Recent work has shown that commonly available machine reading comprehension (MRC) datasets can be used to train high-performance neural information retrieval (IR) systems. However, the evaluation of neural IR has so far been limited to standard supervised learning settings, where they have outperformed traditional term matching baselines. We conduct in-domain and out-of-domain evaluations of neural IR, and seek to improve its robustness across different scenarios, including zeroshot settings. We show that synthetic training examples generated using a sequence-tosequence generator can be effective towards this goal: in our experiments, pre-training with synthetic examples improves retrieval performance in both in-domain and out-of-domain evaluation on five different test sets.

No abstract