{"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/coarse-to-fine-reasoning-for-visual-question","title":"Coarse-to-Fine Reasoning for Visual Question Answering","arxiv_id":"2110.02526","date":"2021-10-06","proceeding":null,"authors":["Binh X. Nguyen","Tuong Do","Huy Tran","Erman Tjiputra","Quang D. Tran","Anh Nguyen"],"abstract":"Bridging the semantic gap between image and question is an important step to improve the accuracy of the Visual Question Answering (VQA) task. However, most of the existing VQA methods focus on attention mechanisms or visual relations for reasoning the answer, while the features at different semantic levels are not fully utilized. In this paper, we present a new reasoning framework to fill the gap between visual features and semantic clues in the VQA task. Our method first extracts the features and predicates from the image and question. We then propose a new reasoning framework to effectively jointly learn these features and predicates in a coarse-to-fine manner. The intensively experimental results on three large-scale VQA datasets show that our proposed approach achieves superior accuracy comparing with other state-of-the-art methods. Furthermore, our reasoning framework also provides an explainable way to understand the decision of the deep neural network when predicting the answer.","url_abs":"https://arxiv.org/abs/2110.02526v2","url_pdf":"https://arxiv.org/pdf/2110.02526v2.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":"coarse-to-fine-reasoning-for-visual-question","repo_url":"https://github.com/aioz-ai/cfr_vqa","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"coarse-to-fine-reasoning-for-visual-question","repo_url":"https://github.com/aioz-ai/crf_vqa","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"visual-question-answering-1","task_name":"Visual Question Answering"},{"task_slug":"visual-question-answering","task_name":"Visual Question Answering (VQA)"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/visual-question-answering-on-gqa-test-dev","task":"Visual Question Answering (VQA)","dataset":"GQA test-dev","model":"CFR","rank_in_archive_order":1,"of":17,"metrics":{"Accuracy":"72.1"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vqa-v2-test-dev","task":"Visual Question Answering (VQA)","dataset":"VQA v2 test-dev","model":"CFR","rank_in_archive_order":23,"of":56,"metrics":{"Accuracy":"72.5"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-visual7w","task":"Visual Question Answering (VQA)","dataset":"Visual7W","model":"CFR","rank_in_archive_order":3,"of":4,"metrics":{"Percentage correct":"71.9"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2110.02526","atlas_url":"https://app.syntology.ai/?focus=2110.02526","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2110.02526"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/aioz-ai/cfr_vqa","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/aioz-ai/crf_vqa","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":1,"ran_violates":1,"ran_honours":2,"unverified":5},"by_repo_kind":{"official":{"samples":9,"ran":4,"repositories":2}},"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":"aioz-ai/crf_vqa","repo_kind":"official","path":"lxrt/modeling.py","file_url":"https://github.com/aioz-ai/crf_vqa/blob/HEAD/lxrt/modeling.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":2,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0f786c407fb1ee4c"}},{"code_sha256_prefix":"544da5baebb8c5e3","entry":"compute_score_with_logits","repo":"aioz-ai/cfr_vqa","repo_kind":"official","path":"src/FFOE/trainer.py","file_url":"https://github.com/aioz-ai/cfr_vqa/blob/HEAD/src/FFOE/trainer.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"544da5baebb8c5e3"}},{"code_sha256_prefix":"fdc64f4c72036ae4","entry":"gelu","repo":"aioz-ai/crf_vqa","repo_kind":"official","path":"lxrt/modeling.py","file_url":"https://github.com/aioz-ai/crf_vqa/blob/HEAD/lxrt/modeling.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fdc64f4c72036ae4"}},{"code_sha256_prefix":"1df9a5ffd9b38c34","entry":"pil_loader","repo":"aioz-ai/crf_vqa","repo_kind":"official","path":"src/utils.py","file_url":"https://github.com/aioz-ai/crf_vqa/blob/HEAD/src/utils.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1df9a5ffd9b38c34"}},{"code_sha256_prefix":"71aa385eadeb1226","entry":"create_glove_embedding_init","repo":"aioz-ai/crf_vqa","repo_kind":"official","path":"src/dataset.py","file_url":"https://github.com/aioz-ai/crf_vqa/blob/HEAD/src/dataset.py","link_basis":"harvester_set","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":"71aa385eadeb1226"}},{"code_sha256_prefix":"8c1252eacab71602","entry":"load_folder","repo":"aioz-ai/crf_vqa","repo_kind":"official","path":"src/utils.py","file_url":"https://github.com/aioz-ai/crf_vqa/blob/HEAD/src/utils.py","link_basis":"harvester_set","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":"8c1252eacab71602"}},{"code_sha256_prefix":"bf90bd27fe08f3a2","entry":"load_imageid","repo":"aioz-ai/crf_vqa","repo_kind":"official","path":"src/utils.py","file_url":"https://github.com/aioz-ai/crf_vqa/blob/HEAD/src/utils.py","link_basis":"harvester_set","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":"bf90bd27fe08f3a2"}},{"code_sha256_prefix":"7099305544368003","entry":"load_tf_weights_in_bert","repo":"aioz-ai/crf_vqa","repo_kind":"official","path":"lxrt/modeling.py","file_url":"https://github.com/aioz-ai/crf_vqa/blob/HEAD/lxrt/modeling.py","link_basis":"harvester_set","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":"7099305544368003"}},{"code_sha256_prefix":"31b89762375a5c2d","entry":"tfidf_from_questions","repo":"aioz-ai/crf_vqa","repo_kind":"official","path":"src/dataset.py","file_url":"https://github.com/aioz-ai/crf_vqa/blob/HEAD/src/dataset.py","link_basis":"harvester_set","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":"31b89762375a5c2d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}