No abstract

The Situated Interactive Multi-Modal Conversations (SIMMC) 2.0 aims to create virtual shopping assistants that can accept complex multi-modal inputs, i.e. visual appearances of objects and user utterances. It consists of four subtasks, multi-modal disambiguation (MM-Disamb), multi-modal coreference resolution (MM-Coref), multi-modal dialog state tracking (MM-DST), and response retrieval and generation. While many task-oriented dialog systems usually tackle each subtask separately, we propose a jointly learned multi-modal encoderdecoder that incorporates visual inputs and performs all four subtasks at once for efficiency. This approach won the MM-Coref and response retrieval subtasks and was nominated runnerup for the remaining subtasks using a single unified model at the 10th Dialog Systems Technology Challenge (DSTC10), setting a high bar for the novel task of multi-modal task-oriented dialog systems. 1

No abstract

By carefully analyzing South Korean photographer Kim Kichan's works on the ordinary people's lives in the streets of Seoul in the 1970s, this chapter problematizes two photographic ideologies that have informed South Korea's anticolonial minjung nationalism: photography as a truthful record of reality and as a nostalgic emblem of the past. Instead of indexing such notions of linear history, this chapter treats Kim Kichan's works as the interfaces located among diverse forces and elements in the matrix of history that far exceed the nationalist horizon of minjung imaginary. More specifically, Kim Kichan's photographs foreground the three-partite relationship ofhuman-animal-environment in narrow streets, providing a significant axis around which the very notion of the human and its agency are redefined.