{ "id": "1709.03968", "version": "v1", "published": "2017-09-12T17:41:30.000Z", "updated": "2017-09-12T17:41:30.000Z", "title": "Affective Neural Response Generation", "authors": [ "Nabiha Asghar", "Pascal Poupart", "Jesse Hoey", "Xin Jiang", "Lili Mou" ], "comment": "8 pages", "categories": [ "cs.CL" ], "abstract": "Existing neural conversational models process natural language primarily on a lexico-syntactic level, thereby ignoring one of the most crucial components of human-to-human dialogue: its affective content. We take a step in this direction by proposing three novel ways to incorporate affective/emotional aspects into long short term memory (LSTM) encoder-decoder neural conversation models: (1) affective word embeddings, which are cognitively engineered, (2) affect-based objective functions that augment the standard cross-entropy loss, and (3) affectively diverse beam search for decoding. Experiments show that these techniques improve the open-domain conversational prowess of encoder-decoder networks by enabling them to produce emotionally rich responses that are more interesting and natural.", "revisions": [ { "version": "v1", "updated": "2017-09-12T17:41:30.000Z" } ], "analyses": { "subjects": [ "68T50", "I.2.7" ], "keywords": [ "affective neural response generation", "models process natural language", "process natural language primarily", "neural conversational models process natural" ], "note": { "typesetting": "TeX", "pages": 8, "language": "en", "license": "arXiv", "status": "editable" } } }