TrICy: Trigger-guided Data-to-text Generation with Intent aware Attention-Copy

<PRE_TAG>Data-to-text (D2T)</POST_TAG> generation is a crucial task in many natural language understanding (NLU) applications and forms the foundation of task-oriented dialog systems. In the context of conversational AI solutions that can work directly with local data on the user's device, architectures utilizing large pre-trained language models (PLMs) are impractical for on-device deployment due to a high memory footprint. To this end, we propose <PRE_TAG>TrICy</POST_TAG>, a novel lightweight framework for an enhanced D2T task that generates text sequences based on the intent in context and may further be guided by user-provided triggers. We leverage an attention-copy mechanism to predict out-of-vocabulary (OOV) words accurately. Performance analyses on <PRE_TAG>E2E NLG dataset</POST_TAG> (<PRE_TAG>BLEU</POST_TAG>: 66.43%, <PRE_TAG>ROUGE-L</POST_TAG>: 70.14%), <PRE_TAG>WebNLG dataset</POST_TAG> (<PRE_TAG>BLEU</POST_TAG>: Seen 64.08%, Unseen 52.35%), and our Custom dataset related to text messaging applications, showcase our architecture's effectiveness. Moreover, we show that by leveraging an optional trigger input, data-to-text generation quality increases significantly and achieves the new SOTA score of 69.29% <PRE_TAG>BLEU</POST_TAG> for E2E NLG. Furthermore, our analyses show that <PRE_TAG>TrICy</POST_TAG> achieves at least 24% and 3% improvement in <PRE_TAG>BLEU</POST_TAG> and METEOR respectively over LLMs like GPT-3, ChatGPT, and Llama 2. We also demonstrate that in some scenarios, performance improvement due to triggers is observed even when they are absent in training.