Paper-Conference

Rubrik’s Cube: Testing a New Rubric for Evaluating Explanations on the CUBE dataset
The performance and usability of Large-Language Models (LLMs) are driving their use in explanation generation tasks. However, despite …
Diana Galvan-Sosa, Gabrielle Gaudeau, Pride Kavumba, Yunmeng Li, Hongyi Gu, Zheng Yuan, Keisuke Sakaguchi, Paula Buttery
PDF Cite DOI URL Code and dataset
Prompting for explanations improves Adversarial NLI. Is this true? Yes it is true because it weakens superficial cues
Explanation prompts ask language models to not only assign a particular label to a giveninput, such as true, entailment, or …
Pride Kavumba, Ana Brassard, Benjamin Heinzerling, Kentaro Inui
PDF Cite URL
因果的 プロンプトによる NLI の敵対的ロバスト性の強化
Pride Kavumba, Ana Brassard, Benjamin Heinzerling, 坂口慶祐, 乾健太郎
PDF Cite
COPA-SSE: Semi-structured Explanations for Commonsense Reasoning
We present Semi-Structured Explanations for COPA (COPA-SSE), a new crowdsourced dataset of 9,747 semi-structured, English common sense …
Ana Brassard, Benjamin Heinzerling, Pride Kavumba, Kentaro Inui
PDF Cite URL Dataset
Are Prompt-based Models Clueless?
Finetuning large pre-trained language models with a task-specific head has advanced the state-of-the-art on many natural language …
Pride Kavumba, Ryo Takahashi, Yusuke Oda
PDF Cite DOI URL
プロンプトモデルは表面的手がかりを 利用するか
Pride Kavumba, 高橋諒, 小田悠介
PDF Cite
Learning to Learn to be Right for the Right Reasons
Improving model generalization on held-out data is one of the core objectives in common- sense reasoning. Recent work has shown that …
Pride Kavumba, Benjamin Heinzerling, Ana Brassard, Kentaro Inui
PDF Cite DOI URL
None the wiser? Adding “None’’Mitigates Superficial Cues in Multiple-Choice Benchmarks
Pride Kavumba, Ana Brassard, Benjamin Heinzerling, Naoya Inoue, Kentaro Inui
PDF Cite
Improving Evidence Detection by Leveraging Warrants
Recognizing the implicit link between a claim and a piece of evidence (i.e. warrant) is the key to improving the performance of …
Keshav Singh, Paul Reisert, Naoya Inoue, Pride Kavumba, Kentaro Inui
PDF Cite DOI URL
When Choosing Plausible Alternatives, Clever Hans can be Clever
Pretrained language models, such as BERT and RoBERTa, have shown large improvements in the commonsense reasoning benchmark COPA. …
Pride Kavumba, Naoya Inoue, Benjamin Heinzerling, Keshav Singh, Paul Reisert, Kentaro Inui
PDF Cite DOI URL Dataset Hugging Face dataset