Facebook has released an open-source framework that allows human tutors to teach AI chatbots how to have more natural conversations.
The company's artificial-intelligence research team, known as FAIR, has released the Python-based ParlAI tool, which gives researchers the ability to train bots over many different datasets at once, including benchmarking datasets like WebQuestions, WikiMovies and bAbl tasks.
Training is split into five tasks: general question answering; sentence completion, where the bot has to fill in the next word of a dialogue; back-and-forth chit-chat; visual dialogue with conversations involving images; and goal oriented dialogue, where a conversation is intended with a specific outcome.
"For the first time, ParlAI provides a place for researchers to collect all the significant dialog tasks in one place," wrote FAIR researchers Jason Weston, Alexander Miller and Will Feng in a blog post.
"Not only does this allow for much easier iteration on any of these tasks individually, but also makes it possible to easily train a bot on all of them (which eventually should lead to better bots) and to evaluate bots across all those skills."
The framework uses Amazon Mechanical Turk, a programme which uses low-paid human operators to do repetitive grunt-work that goes into training AI models, to improve the quality of AI conversations through question-and-answer sessions.
The release comes as part of the company's push into the AI space. The goal is to provide advertisers and brands with the ability to roll out natural-language chatbots on platforms like Facebook Messenger and WhatsApp, allowing customers to make enquiries, bookings and purchases just by talking to an AI helper.
"ParlAI is a platform we hope will bring together the community of researchers working on AI agents that perform dialog," Facebook's researchers said, "and continue pushing the state of the art in dialog research."
-
Cleo attack victim list grows as Hertz confirms customer data stolenNews Hertz has confirmed it suffered a data breach as a result of the Cleo zero-day vulnerability in late 2024, with the car rental giant warning that customer data was stolen.
-
Lateral moves in tech: Why leaders should support employee mobilityIn-depth Encouraging staff to switch roles can have long-term benefits for skills in the tech sector
-
Meta to pay $725 million in Cambridge Analytica lawsuit settlementNews The settlement closes the long-running lawsuit into how Facebook's owner, Meta, handled the Cambridge Analytica scandal
-
Meta's earnings are 'cause for concern' and 2023 looks even bleakerAnalysis Calls for investor faith in metaverse tech only emphasise the worries that its investment strategy won't pay off
-
Microsoft and Meta announce integration deal between Teams and WorkplaceNews Features from both business collaboration platforms will be available to users without having to switch apps
-
Facebook is shutting down its controversial facial recognition systemNews The move will see more than a billion facial templates removed from Facebook's records amid a push for more private applications of the technology
-
'Changing name to Meat': Industry reacts to Facebook's Meta rebrandNews The rebrand attempts to provide a clearer distinction between Facebook and its umbrella company
-
Facebook's Oversight Board demands more transparencyNews Board bashed the social media giant for its preferential treatment of certain high-profile accounts
-
Facebook claims AI managed to reduce hate speech by 50%News The social media platform has hit back at claims the tech it uses to fight hate speech is inadequate
-
Facebook to hire 10,000 workers across the EUNews The high-skilled jobs drive is a “vote of confidence” in the European tech industry
