Phan

Neutral and Contradiction Biases in Premise-Hypothesis sentencesPhoto by Possessed Photography on UnsplashMotivationNatural language inference (NLI) is a widely-studied natural language processing task to determine the entailment relation between two statements: a premise and a hypothesis.The Stanford NLI corpus (version 1.0) is a collection of 570k human-written English sentence pairs manually labeled for training NLP models on NLI task. Although there are state-of- the-art NLP models that can achieve high accuracy on…

For three years, Dan Howell didn’t post anything on YouTube, and for his 6 million subscribers, this was a very big deal. As one half of the longtime vlogger duo Dan and Phil, he was known for writing and performing sketch comedy, internet culture commentary, and occasionally more serious vlogs about mental illness. A typical Dan and Phil endeavor, be it a book or a world tour or a series, might be called something like “The Super Amazing Project” or “Interactive Introverts,” nodding to early memes associated with…