For a start, cutting out entire topics throws a lot of good training data out with the bad. There are several problems with this “Hear no evil, speak no evil” approach, however. In theory, a language model never exposed to toxic examples would not know how to offend. Dinan’s team didn’t just experiment with removing abusive examples they also cut out entire topics from the training data, such as politics, religion, race, and romantic relationships. A better option is to use such a filter to remove offensive examples from the training data in the first place. The bolt-on filter would also require extra computing power to run. If that filter was removed, the offensive bot would be exposed again. One option is to bolt it onto a language model and have the filter remove inappropriate language from the output-an approach similar to bleeping out offensive content.īut this would require language models to have such a filter attached all the time. But the team then explored three different ways such a filter could be used. This is a basic first step for many AI-powered hate-speech filters. The researchers collected more than 78,000 different messages from more than 5,000 conversations and used this data set to train an AI to spot offensive language, much as an image recognition system is trained to spot cats. To do this, the participants used profanity (such as “Holy fuck he’s ugly!”) or asked inappropriate questions (such as “Women should stay in the home. Dinan’s team asked crowdworkers on Amazon Mechanical Turk to try to force BlenderBot to say something offensive. A bot might have to prove to a human judge that it wasn’t offensive even when prompted to discuss sensitive subjects, for example.īut to stop a language model from generating offensive text, you first need to be able to spot it.Įmily Dinan and her colleagues at Facebook AI Research presented a paper at the workshop that looked at ways to remove offensive output from BlenderBot, a chatbot built on Facebook’s language model Blender, which was trained on Reddit. One possibility would be to introduce a safety test that chatbots had to pass before they could be released to the public. Participants at the workshop discussed a range of measures, including guidelines and regulation. “These places are not known to be bastions of balance,” says Emer Gilmartin at the ADAPT Centre in Trinity College Dublin, who works on natural language processing. Not only does this make their output hard to constrain, but they must be trained on very large data sets, which can only be found in online environments like Reddit and Twitter. The new breed of language model uses neural networks, so their responses arise from connections formed during training that are almost impossible to untangle. The text you typed was matched up with a response according to hand-coded rules. Yet until recently, most chatbots used rule-based AI. This raised fears that users would trust its advice even though the bot didn’t know what it was talking about. ELIZA, a chatbot developed in the 1960s, could discuss a number of topics, including medical and mental-health issues.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |