20 citations · 20 across the 7 of their papers we have counts for
8 papers
LAIT: Efficient Multi-Segment Encoding in Transformers with Layer-Adjustable Interaction
Jeremiah Milbauer, Annie Louis, Mohammad Javad Hosseini +3
Transformer encoders contextualize token representations by attending to all other tokens at each layer, leading to quadratic increase in compute effort with the input length. In p…
Sources of Hallucination by Large Language Models on Inference Tasks
Nick McKenna, Tianyi Li, Liang Cheng +3
Large Language Models (LLMs) are claimed to be capable of Natural Language Inference (NLI), necessary for applied tasks like question answering and summarization. We present a seri…
Resolving Indirect Referring Expressions for Entity Selection
Mohammad Javad Hosseini, Filip Radlinski, Silvia Pareti +1
Recent advances in language modeling have enabled new conversational systems. In particular, it is often desirable for people to make choices among specified options when using suc…
Language Models Are Poor Learners of Directional Inference
Tianyi Li, Mohammad Javad Hosseini, Sabine Weber +1
We examine LMs' competence of directional predicate entailments by supervised fine-tuning with prompts. Our analysis shows that contrary to their apparent success on standard NLI,…
Cross-lingual Inference with A Chinese Entailment Graph
Tianyi Li, Sabine Weber, Mohammad Javad Hosseini +2
Predicate entailment detection is a crucial task for question-answering from text, where previous work has explored unsupervised learning of entailment graphs from typed open relat…
Incorporating Temporal Information in Entailment Graph Mining
Liane Guillou, Sander Bijl de Vroe, Mohammad Javad Hosseini +2
We present a novel method for injecting temporality into entailment graphs to address the problem of spurious entailments, which may arise from similar but temporally distinct even…
Multivalent Entailment Graphs for Question Answering
Nick McKenna, Liane Guillou, Mohammad Javad Hosseini +3
Drawing inferences between open-domain natural language predicates is a necessity for true language understanding. There has been much progress in unsupervised learning of entailme…
Jointly Modeling Hierarchical and Horizontal Features for Relational Triple Extraction
Zhepei Wei, Yantao Jia, Yuan Tian +4
Recent works on relational triple extraction have shown the superiority of jointly extracting entities and relations over the pipelined extraction manner. However, most existing jo…