{"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/large-scale-pretraining-for-visual-dialog-a","title":"Large-scale Pretraining for Visual Dialog: A Simple State-of-the-Art Baseline","arxiv_id":"1912.02379","date":"2019-12-05","proceeding":"ECCV 2020 8","authors":["Vishvak Murahari","Dhruv Batra","Devi Parikh","Abhishek Das"],"abstract":"Prior work in visual dialog has focused on training deep neural models on VisDial in isolation. Instead, we present an approach to leverage pretraining on related vision-language datasets before transferring to visual dialog. We adapt the recently proposed ViLBERT (Lu et al., 2019) model for multi-turn visually-grounded conversations. Our model is pretrained on the Conceptual Captions and Visual Question Answering datasets, and finetuned on VisDial. Our best single model outperforms prior published work (including model ensembles) by more than 1% absolute on NDCG and MRR. Next, we find that additional finetuning using \"dense\" annotations in VisDial leads to even higher NDCG -- more than 10% over our base model -- but hurts MRR -- more than 17% below our base model! This highlights a trade-off between the two primary metrics -- NDCG and MRR -- which we find is due to dense annotations not correlating well with the original ground-truth answers to questions.","url_abs":"https://arxiv.org/abs/1912.02379v2","url_pdf":"https://arxiv.org/pdf/1912.02379v2.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":"large-scale-pretraining-for-visual-dialog-a","repo_url":"https://github.com/vmurahari3/visdial-bert","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"paper_slug":"large-scale-pretraining-for-visual-dialog-a","repo_url":"https://github.com/zihaow123/unimm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"visual-dialogue","task_name":"Visual Dialog"},{"task_slug":"visual-question-answering-1","task_name":"Visual Question Answering"}],"methods":[{"method_slug":"vilbert","method_name":"ViLBERT"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1912.02379","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1912.02379"}},"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/zihaow123/unimm","reach":{"status":"ok","spdx":"GPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/vmurahari3/visdial-bert","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"summary":{"ran_draft_wrong":2,"ran_honours":1,"unverified":3},"by_repo_kind":{"official":{"samples":6,"ran":3,"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":"0f786c407fb1ee4c","entry":"swish","repo":"vmurahari3/visdial-bert","repo_kind":"official","path":"models/vilbert_dialog.py","file_url":"https://github.com/vmurahari3/visdial-bert/blob/HEAD/models/vilbert_dialog.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":2,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"0f786c407fb1ee4c"}},{"code_sha256_prefix":"fdc64f4c72036ae4","entry":"gelu","repo":"vmurahari3/visdial-bert","repo_kind":"official","path":"models/vilbert_dialog.py","file_url":"https://github.com/vmurahari3/visdial-bert/blob/HEAD/models/vilbert_dialog.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"fdc64f4c72036ae4"}},{"code_sha256_prefix":"bba223178ea790b8","entry":"scores_to_ranks","repo":"vmurahari3/visdial-bert","repo_kind":"official","path":"utils/visdial_metrics.py","file_url":"https://github.com/vmurahari3/visdial-bert/blob/HEAD/utils/visdial_metrics.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"bba223178ea790b8"}},{"code_sha256_prefix":"baa5766f4566aafc","entry":"load_tf_weights_in_bert","repo":"vmurahari3/visdial-bert","repo_kind":"official","path":"models/vilbert_dialog.py","file_url":"https://github.com/vmurahari3/visdial-bert/blob/HEAD/models/vilbert_dialog.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"baa5766f4566aafc"}},{"code_sha256_prefix":"21ad1eb3bb7f8c47","entry":"read_command_line","repo":"vmurahari3/visdial-bert","repo_kind":"official","path":"options.py","file_url":"https://github.com/vmurahari3/visdial-bert/blob/HEAD/options.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"21ad1eb3bb7f8c47"}},{"code_sha256_prefix":"5593052e318b6555","entry":"read_options","repo":"vmurahari3/visdial-bert","repo_kind":"official","path":"preprocessing/pre_process_visdial.py","file_url":"https://github.com/vmurahari3/visdial-bert/blob/HEAD/preprocessing/pre_process_visdial.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"5593052e318b6555"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}