{"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/pace-unified-multi-modal-dialogue-pre","title":"PaCE: Unified Multi-modal Dialogue Pre-training with Progressive and Compositional Experts","arxiv_id":"2305.14839","date":"2023-05-24","proceeding":null,"authors":["Yunshui Li","Binyuan Hui","Zhichao Yin","Min Yang","Fei Huang","Yongbin Li"],"abstract":"Perceiving multi-modal information and fulfilling dialogues with humans is a long-term goal of artificial intelligence. Pre-training is commonly regarded as an effective approach for multi-modal dialogue. However, due to the limited availability of multi-modal dialogue data, there is still scarce research on multi-modal dialogue pre-training. Yet another intriguing challenge emerges from the encompassing nature of multi-modal dialogue, which involves various modalities and tasks. Moreover, new forms of tasks may arise at unpredictable points in the future. Hence, it is essential for designed multi-modal dialogue models to possess sufficient flexibility to adapt to such scenarios. This paper proposes \\textbf{PaCE}, a unified, structured, compositional multi-modal dialogue pre-training framework. It utilizes a combination of several fundamental experts to accommodate multiple dialogue-related tasks and can be pre-trained using limited dialogue and extensive non-dialogue multi-modal data. Furthermore, we propose a progressive training method where old experts from the past can assist new experts, facilitating the expansion of their capabilities. Experimental results demonstrate that PaCE achieves state-of-the-art results on eight multi-modal dialog benchmarks.","url_abs":"https://arxiv.org/abs/2305.14839v2","url_pdf":"https://arxiv.org/pdf/2305.14839v2.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":"pace-unified-multi-modal-dialogue-pre","repo_url":"https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/pace","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"dialogue-state-tracking","task_name":"Dialogue State Tracking"},{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"multimodal-intent-recognition","task_name":"Multimodal Intent Recognition"},{"task_slug":"response-generation","task_name":"Response Generation"},{"task_slug":"text-retrieval","task_name":"Text Retrieval"},{"task_slug":"visual-dialogue","task_name":"Visual Dialog"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/dialogue-state-tracking-on-mmconv","task":"Dialogue State Tracking","dataset":"MMConv","model":"PaCE","rank_in_archive_order":1,"of":2,"metrics":{"Categorical Accuracy":"92.2","Non-Categorical Accuracy":"43.4","Overall":"39.2"},"uses_additional_data":false},{"leaderboard":"/sota/dialogue-state-tracking-on-simmc2-0","task":"Dialogue State Tracking","dataset":"SIMMC2.0","model":"PaCE","rank_in_archive_order":1,"of":5,"metrics":{"Act F1":"97.1","Slot F1":"87.0"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-photochat","task":"Image Retrieval","dataset":"PhotoChat","model":"PaCE","rank_in_archive_order":1,"of":5,"metrics":{"R1":"15.2","R@10":"49.6","R@5":"36.7","Sum(R@1,5,10)":"101.5"},"uses_additional_data":false},{"leaderboard":"/sota/multimodal-intent-recognition-on-mmdialog","task":"Multimodal Intent Recognition","dataset":"MMDialog","model":"PaCE","rank_in_archive_order":1,"of":4,"metrics":{"F1":"77.6"},"uses_additional_data":false},{"leaderboard":"/sota/multimodal-intent-recognition-on-photochat","task":"Multimodal Intent Recognition","dataset":"PhotoChat","model":"PaCE","rank_in_archive_order":1,"of":6,"metrics":{"F1":"63.8","Precision":"63.3","Recall":"68"},"uses_additional_data":false},{"leaderboard":"/sota/response-generation-on-mmconv","task":"Response Generation","dataset":"MMConv","model":"PaCE","rank_in_archive_order":1,"of":2,"metrics":{"BLEU":"22","Comb.":"44.7","Inform":"34.5","Success":"13.9"},"uses_additional_data":false},{"leaderboard":"/sota/response-generation-on-simmc2-0","task":"Response Generation","dataset":"SIMMC2.0","model":"PaCE","rank_in_archive_order":1,"of":5,"metrics":{"BLEU":"34.1"},"uses_additional_data":false},{"leaderboard":"/sota/text-retrieval-on-image-chat","task":"Text Retrieval","dataset":"Image-Chat","model":"PaCE","rank_in_archive_order":1,"of":3,"metrics":{"R@1":"51.9","R@5":"76.8","Sum(R@1,5)":"128.7"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2305.14839","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2305.14839"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/pace","reach":null}],"summary":{"unverified":1},"by_repo_kind":{"official":{"samples":1,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"94634ba582cc19b4","entry":"TransformerSS","repo":"AlibabaResearch/DAMO-ConvAI","repo_kind":"official","path":"pace/pace/modules/pace_module.py","file_url":"https://github.com/AlibabaResearch/DAMO-ConvAI/blob/HEAD/pace/pace/modules/pace_module.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"94634ba582cc19b4"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}