Haptik introduces HINT3 to benchmark performance of Dialogue Agents

Intent Detection is a vital part of the Natural Language Understanding (NLU) pipeline of  Task-oriented dialogue systems. Recent advances in NLP have enabled systems that perform quite well on existing intent detection benchmarking datasets like HWU64, CLINC150, BANKING77 as shown in Larson et al., 2019, Casanueva et al., 2020. However, most existing datasets for intent detection are generated using crowdsourcing services. This difference in dataset preparation methodology leads to assumptions about training data which are no longer valid in the real world. In the real world, definition of intent often varies across users, tasks and domains.

Read More

Open-Sourcing The Nlu ‘Swiss Army Knife’ For Conversational AI

We recently open-sourced a repository called Multi-Task learning which can help with multiple conversational AI tasks. In this blog, we have explained why such an architecture can change the way we build conversational AI systems and the thought process behind building it. 

Read More

The Haptik Open Source Challenge

As we’ve said in the past, we at Haptik are highly grateful for the open source community for the role it has played in getting us where we are, and are always looking to contribute back with work such as Pacemaker. However, while we’ll continue shipping code, we really would like others to get involved as well.

Read More