{"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/unifying-vision-and-language-tasks-via-text","title":"Unifying Vision-and-Language Tasks via Text Generation","arxiv_id":"2102.02779","date":"2021-02-04","proceeding":null,"authors":["Jaemin Cho","Jie Lei","Hao Tan","Mohit Bansal"],"abstract":"Existing methods for vision-and-language learning typically require designing task-specific architectures and objectives for each task. For example, a multi-label answer classifier for visual question answering, a region scorer for referring expression comprehension, and a language decoder for image captioning, etc. To alleviate these hassles, in this work, we propose a unified framework that learns different tasks in a single architecture with the same language modeling objective, i.e., multimodal conditional text generation, where our models learn to generate labels in text based on the visual and textual inputs. On 7 popular vision-and-language benchmarks, including visual question answering, referring expression comprehension, visual commonsense reasoning, most of which have been previously modeled as discriminative tasks, our generative approach (with a single unified architecture) reaches comparable performance to recent task-specific state-of-the-art vision-and-language models. Moreover, our generative approach shows better generalization ability on questions that have rare answers. Also, we show that our framework allows multi-task learning in a single architecture with a single set of parameters, achieving similar performance to separately optimized single-task models. Our code is publicly available at: https://github.com/j-min/VL-T5","url_abs":"https://arxiv.org/abs/2102.02779v2","url_pdf":"https://arxiv.org/pdf/2102.02779v2.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":"unifying-vision-and-language-tasks-via-text","repo_url":"https://github.com/j-min/VL-T5","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"unifying-vision-and-language-tasks-via-text","repo_url":"https://github.com/mitvis/vistext","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[{"task_slug":"conditional-text-generation","task_name":"Conditional Text Generation"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"image-captioning","task_name":"Image Captioning"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"referring-expression","task_name":"Referring Expression"},{"task_slug":"referring-expression-comprehension","task_name":"Referring Expression Comprehension"},{"task_slug":"text-generation","task_name":"Text Generation"},{"task_slug":"visual-commonsense-reasoning","task_name":"Visual Commonsense Reasoning"},{"task_slug":"visual-question-answering-1","task_name":"Visual Question Answering"},{"task_slug":"visual-question-answering","task_name":"Visual Question Answering (VQA)"}],"methods":[{"method_slug":"vl-t5","method_name":"VL-T5"}],"datasets_introduced":[],"methods_introduced":[{"slug":"vl-t5","name":"VL-T5","full_name":"VL-T5"}],"results":[{"leaderboard":"/sota/image-captioning-on-flickr30k-captions-test","task":"Image Captioning","dataset":"Flickr30k Captions test","model":"VL-T5","rank_in_archive_order":7,"of":7,"metrics":{"CIDEr":"2.6","SPICE":"2.0"},"uses_additional_data":false},{"leaderboard":"/sota/image-captioning-on-nocaps-val","task":"Image Captioning","dataset":"nocaps val","model":"VL-T5","rank_in_archive_order":3,"of":3,"metrics":{"CIDEr":"4.4","SPICE":" 5.3"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vcr-q-a-test","task":"Visual Question Answering (VQA)","dataset":"VCR (Q-A) test","model":"VL-T5","rank_in_archive_order":8,"of":11,"metrics":{"Accuracy":"75.3"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vcr-q-ar-test","task":"Visual Question Answering (VQA)","dataset":"VCR (Q-AR) test","model":"VL-T5","rank_in_archive_order":6,"of":7,"metrics":{"Accuracy":"58.9"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-vcr-qa-r-test","task":"Visual Question Answering (VQA)","dataset":"VCR (QA-R) test","model":"VL-T5","rank_in_archive_order":7,"of":8,"metrics":{"Accuracy":"77.8"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2102.02779","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2102.02779"}},"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/mitvis/vistext","reach":{"status":"ok","spdx":"GPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/j-min/VL-T5","reach":null}],"summary":{"ran":2,"unverified":10},"by_repo_kind":{"official":{"samples":12,"ran":2,"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":"b5a6c04b009cca88","entry":"VLSeq2SeqLMOutput","repo":"j-min/VL-T5","repo_kind":"official","path":"VL-T5/src/multitask_model.py","file_url":"https://github.com/j-min/VL-T5/blob/HEAD/VL-T5/src/multitask_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b5a6c04b009cca88"}},{"code_sha256_prefix":"e44fd46c867cf8be","entry":"VisualEmbedding","repo":"j-min/VL-T5","repo_kind":"official","path":"VL-T5/src/multitask_model.py","file_url":"https://github.com/j-min/VL-T5/blob/HEAD/VL-T5/src/multitask_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e44fd46c867cf8be"}},{"code_sha256_prefix":"9ae4551ab4057d4f","entry":"JointEncoder","repo":"j-min/VL-T5","repo_kind":"official","path":"VL-T5/src/multitask_model.py","file_url":"https://github.com/j-min/VL-T5/blob/HEAD/VL-T5/src/multitask_model.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":"9ae4551ab4057d4f"}},{"code_sha256_prefix":"5d69b18e68c1e195","entry":"VLT5","repo":"j-min/VL-T5","repo_kind":"official","path":"VL-T5/src/multitask_model.py","file_url":"https://github.com/j-min/VL-T5/blob/HEAD/VL-T5/src/multitask_model.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":"5d69b18e68c1e195"}},{"code_sha256_prefix":"943c42cc626cc9a5","entry":"VLT5COCOCaption","repo":"j-min/VL-T5","repo_kind":"official","path":"VL-T5/src/multitask_model.py","file_url":"https://github.com/j-min/VL-T5/blob/HEAD/VL-T5/src/multitask_model.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":"943c42cc626cc9a5"}},{"code_sha256_prefix":"c6c8a88fa5a29395","entry":"VLT5GQA","repo":"j-min/VL-T5","repo_kind":"official","path":"VL-T5/src/multitask_model.py","file_url":"https://github.com/j-min/VL-T5/blob/HEAD/VL-T5/src/multitask_model.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":"c6c8a88fa5a29395"}},{"code_sha256_prefix":"8f8e3dcb1533f514","entry":"VLT5MMT","repo":"j-min/VL-T5","repo_kind":"official","path":"VL-T5/src/multitask_model.py","file_url":"https://github.com/j-min/VL-T5/blob/HEAD/VL-T5/src/multitask_model.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":"8f8e3dcb1533f514"}},{"code_sha256_prefix":"615b17a71b757c72","entry":"VLT5MultiTask","repo":"j-min/VL-T5","repo_kind":"official","path":"VL-T5/src/multitask_model.py","file_url":"https://github.com/j-min/VL-T5/blob/HEAD/VL-T5/src/multitask_model.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":"615b17a71b757c72"}},{"code_sha256_prefix":"a991fbc2a12b6068","entry":"VLT5NLVR","repo":"j-min/VL-T5","repo_kind":"official","path":"VL-T5/src/multitask_model.py","file_url":"https://github.com/j-min/VL-T5/blob/HEAD/VL-T5/src/multitask_model.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":"a991fbc2a12b6068"}},{"code_sha256_prefix":"99a4eac0e06e3904","entry":"VLT5RefCOCO","repo":"j-min/VL-T5","repo_kind":"official","path":"VL-T5/src/multitask_model.py","file_url":"https://github.com/j-min/VL-T5/blob/HEAD/VL-T5/src/multitask_model.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":"99a4eac0e06e3904"}},{"code_sha256_prefix":"b4349d2ccfd035b9","entry":"VLT5VCR","repo":"j-min/VL-T5","repo_kind":"official","path":"VL-T5/src/multitask_model.py","file_url":"https://github.com/j-min/VL-T5/blob/HEAD/VL-T5/src/multitask_model.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":"b4349d2ccfd035b9"}},{"code_sha256_prefix":"79580be4d7e75e1a","entry":"VLT5VQA","repo":"j-min/VL-T5","repo_kind":"official","path":"VL-T5/src/multitask_model.py","file_url":"https://github.com/j-min/VL-T5/blob/HEAD/VL-T5/src/multitask_model.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":"79580be4d7e75e1a"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}