{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/livebot-generating-live-video-comments-based","title":"LiveBot: Generating Live Video Comments Based on Visual and Textual Contexts","arxiv_id":"1809.04938","date":"2018-09-13","proceeding":null,"authors":["Shuming Ma","Lei Cui","Damai Dai","Furu Wei","Xu sun"],"abstract":"We introduce the task of automatic live commenting. Live commenting, which is\nalso called `video barrage', is an emerging feature on online video sites that\nallows real-time comments from viewers to fly across the screen like bullets or\nroll at the right side of the screen. The live comments are a mixture of\nopinions for the video and the chit chats with other comments. Automatic live\ncommenting requires AI agents to comprehend the videos and interact with human\nviewers who also make the comments, so it is a good testbed of an AI agent's\nability of dealing with both dynamic vision and language. In this work, we\nconstruct a large-scale live comment dataset with 2,361 videos and 895,929 live\ncomments. Then, we introduce two neural models to generate live comments based\non the visual and textual contexts, which achieve better performance than\nprevious neural baselines such as the sequence-to-sequence model. Finally, we\nprovide a retrieval-based evaluation protocol for automatic live commenting\nwhere the model is asked to sort a set of candidate comments based on the\nlog-likelihood score, and evaluated on metrics such as mean-reciprocal-rank.\nPutting it all together, we demonstrate the first `LiveBot'.","url_abs":"http://arxiv.org/abs/1809.04938v2","url_pdf":"http://arxiv.org/pdf/1809.04938v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"livebot-generating-live-video-comments-based","repo_url":"https://github.com/lancopku/livebot","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"livebot-generating-live-video-comments-based","repo_url":"https://github.com/VP-0822/Video-Keyword-Extractor","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"livebot-generating-live-video-comments-based","repo_url":"https://github.com/fireflyHunter/OpenNMT-Livebot","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[{"slug":"live-comment-dataset","name":"Live Comment Dataset","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.04938","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}