VOGUE: Answer Verbalization Through Multi-Task Learning

In recent years, there have been significant developments in Question Answering over Knowledge Graphs (KGQA). Despite all the notable advancements, current KGQA systems only focus on answer generation techniques and not verbalization. However, real-world scenarios (e.g., voice assistants such as Alexa, Siri, etc.), users prefer verbalized answers instead of a generated response. This paper addresses task verbalization for (complex) question answering knowledge graphs. this context, we propose multi-task-based framework: VOGUE (Verbalization thrOuGh mUlti-task lEarning). The framework attempts to generate using hybrid approach through multi-task learning paradigm. Our can results based questions queries inputs concurrently. comprises four modules that are trained simultaneously learning. We evaluate our existing datasets verbalization, it outperforms baselines both BLEU METEOR scores.