We propose CONDOR, a weighted knowledge distillation approach that distills the information from a common-sense KG to a domain-specific KG by intelligently balancing domain relevance and generalization. We use a teacher–student distillation framework where a teacher model trained on a large common-sense knowledge graph guides a student model trained on a smaller domain-specific graph, using a combination of distillation loss and domain-specific triple completion loss to improve downstream task performance. We benchmark CONDOR on offensive speech and mental health domains across classification and generation tasks. Using COMET, pre-trained on common-sense commonsense KGs like ATOMIC and ConceptNet—as the teacher, and domain-specific KGs such as StereoKG (offensive speech) and HealKG (mental health) as student inputs, we evaluate on LatentHatred, SBIC, EDOS, and HOPE datasets. Results show that enriching domain-specific knowledge with distilled commonsense significantly improves downstream performance. Disclaimer. Some examples in the paper are offensive in nature. Reader discretion is advised.

Multi-hop query answering over a Knowledge Graph (KG) involves traversing one or more hops from the start node to answer a query. Path-based and logic-based methods are state-of-the-art for multi-hop question answering. The former is used in link prediction tasks. The latter is for answering complex logical queries. The logical multi-hop querying technique embeds the KG and queries in the same embedding space. The existing work incorporates First Order Logic (FOL) operators, such as conjunction ($\wedge$), disjunction ($\vee$), and negation ($\lnot$), in queries. Though current models have most of the building blocks to execute the FOL queries, they cannot use the dense information of multi-modal entities in the case of Multi-Modal Knowledge Graphs (MMKGs). We propose RConE, an embedding method to capture the multi-modal information needed to answer a query. The model first shortlists candidate (multi-modal) entities containing the answer. It then finds the solution (sub-entities) within those entities. Several existing works tackle path-based question-answering in MMKGs. However, to our knowledge, we are the first to introduce logical constructs in querying MMKGs and to answer queries that involve sub-entities of multi-modal entities as the answer. Extensive evaluation of four publicly available MMKGs indicates that RConE outperforms the current state-of-the-art. The source code and datasets are available athttps://github.com/kracr/rcone-qa-mmkg.