2021 is shaping up to be a great year for developers, but not so much for companies hiring them, with 61% of HR professionals believing this will be their biggest challenge in the year ahead. Recruiters expect a shortage of qualified programmers in 2021. fizkes / iStock Hiring managers worry that they’ll struggle to recruit […]
Pioneering scientists like Rothermel dealt with this intractable problem by ignoring it. Instead, they searched for factors, such as wind speed and slope, that could help them predict a fire’s next move in real time. Looking back, Finney says, it’s a miracle that Rothermel’s equations work for wildfires at all. There’s the sheer difference in […]
From smart toilets and disinfecting robots to transparent OLED displays and sleep tech, CES 2021 was a showcase for the latest innovations in consumer and enterprise technology. CES 2021 is a wrap. And although this year’s all-digital event was a significantly different experience from past shows, there was plenty of innovative tech on display. TechRepublic’s […]
Researchers at Auburn University in Alabama and Adobe Research discovered the flaw when they tried to get an NLP system to generate explanations for its behavior, such as why it claimed different sentences meant the same thing. When they tested their approach, they realized that shuffling words in a sentence made no difference to the explanations. “This is a general problem to all NLP models,” says Anh Nguyen at Auburn University, who led the work.
The team looked at several state-of-the-art NLP systems based on BERT (a language model developed by Google that underpins many of the latest systems, including GPT-3). All of these systems score better than humans on GLUE (General Language Understanding Evaluation), a standard set of tasks designed to test language comprehension, such as spotting paraphrases, judging if a sentence expresses positive or negative sentiments, and verbal reasoning.
Man bites dog: They found that these systems couldn’t tell when words in a sentence were jumbled up, even when the new order changed the meaning. For example, the systems correctly spotted that the sentences “Does marijuana cause cancer?” and “How can smoking marijuana give you lung cancer?” were paraphrases. But they were even more certain that “You smoking cancer how marijuana lung can give?” and “Lung can give marijuana smoking how you cancer?” meant the same thing too. The systems also decided that sentences with opposite meanings—such as “Does marijuana cause cancer?” and “Does cancer cause marijuana?”—were asking the same question.
The only task where word order mattered was one in which the models had to check the grammatical structure of a sentence. Otherwise, between 75% and 90% of the tested systems’ answers did not change when the words were shuffled.
What’s going on? The models appear to pick up on a few key words in a sentence, whatever order they come in. They do not understand language as we do, and GLUE—a very popular benchmark—does not measure true language use. In many cases, the task a model is trained on does not force it to care about word order or syntax in general. In other words, GLUE teaches NLP models to jump through hoops.
Many researchers have started to use a harder set of tests called SuperGLUE, but Nguyen suspects it will have similar problems.
Does it matter? It depends on the application. On one hand, an AI that still understands when you make a typo or say something garbled, as another human could, would be useful. But in general, word order is crucial when unpicking a sentence’s meaning.
fix it How to? The good news is that it might not be too hard to fix. The researchers found that forcing a model to focus on word order, by training it to do a task where word order mattered (such as spotting grammatical errors), also made the model perform better on other tasks. This suggests that tweaking the tasks that models are trained to do will make them better overall.