{"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/finding-visual-task-vectors","title":"Finding Visual Task Vectors","arxiv_id":"2404.05729","date":"2024-04-08","proceeding":null,"authors":["Alberto Hojel","Yutong Bai","Trevor Darrell","Amir Globerson","Amir Bar"],"abstract":"Visual Prompting is a technique for teaching models to perform a visual task via in-context examples, without any additional training. In this work, we analyze the activations of MAE-VQGAN, a recent Visual Prompting model, and find task vectors, activations that encode task-specific information. Equipped with this insight, we demonstrate that it is possible to identify the task vectors and use them to guide the network towards performing different tasks without providing any input-output examples. To find task vectors, we compute the average intermediate activations per task and use the REINFORCE algorithm to search for the subset of task vectors. The resulting task vectors guide the model towards performing a task better than the original model without the need for input-output examples.","url_abs":"https://arxiv.org/abs/2404.05729v2","url_pdf":"https://arxiv.org/pdf/2404.05729v2.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":"finding-visual-task-vectors","repo_url":"https://github.com/alhojel/visual_task_vectors","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"visual-prompting","task_name":"Visual Prompting"}],"methods":[{"method_slug":"reinforce","method_name":"REINFORCE"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2404.05729","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2404.05729"}},"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/alhojel/visual_task_vectors","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":1,"ran_fixture":1,"unverified":9},"by_repo_kind":{"official":{"samples":11,"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":11,"samples":[{"code_sha256_prefix":"221b2d116fdf1032","entry":"get_obj_from_str","repo":"alhojel/visual_task_vectors","repo_kind":"official","path":"vqgan.py","file_url":"https://github.com/alhojel/visual_task_vectors/blob/HEAD/vqgan.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"221b2d116fdf1032"}},{"code_sha256_prefix":"cb49209c125de1b4","entry":"get_timestep_embedding","repo":"alhojel/visual_task_vectors","repo_kind":"official","path":"vqgan.py","file_url":"https://github.com/alhojel/visual_task_vectors/blob/HEAD/vqgan.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"cb49209c125de1b4"}},{"code_sha256_prefix":"764136a5eff74434","entry":"calculate_metric","repo":"alhojel/visual_task_vectors","repo_kind":"official","path":"evaluate/evaluate_colorization.py","file_url":"https://github.com/alhojel/visual_task_vectors/blob/HEAD/evaluate/evaluate_colorization.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"764136a5eff74434"}},{"code_sha256_prefix":"69532fa60013e9f4","entry":"calculate_metric","repo":"alhojel/visual_task_vectors","repo_kind":"official","path":"evaluate/evaluate_reasoning.py","file_url":"https://github.com/alhojel/visual_task_vectors/blob/HEAD/evaluate/evaluate_reasoning.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"69532fa60013e9f4"}},{"code_sha256_prefix":"c68f71948cfa9d7c","entry":"get_demo_predictions","repo":"alhojel/visual_task_vectors","repo_kind":"official","path":"viz_utils.py","file_url":"https://github.com/alhojel/visual_task_vectors/blob/HEAD/viz_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c68f71948cfa9d7c"}},{"code_sha256_prefix":"8c8d20b40760059c","entry":"get_demo_predictions_with_mask","repo":"alhojel/visual_task_vectors","repo_kind":"official","path":"viz_utils.py","file_url":"https://github.com/alhojel/visual_task_vectors/blob/HEAD/viz_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8c8d20b40760059c"}},{"code_sha256_prefix":"f0ac78b4e05be194","entry":"instantiate_from_config","repo":"alhojel/visual_task_vectors","repo_kind":"official","path":"vqgan.py","file_url":"https://github.com/alhojel/visual_task_vectors/blob/HEAD/vqgan.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f0ac78b4e05be194"}},{"code_sha256_prefix":"f82a5154b6295773","entry":"iou","repo":"alhojel/visual_task_vectors","repo_kind":"official","path":"evaluate/collect_attention_heads.py","file_url":"https://github.com/alhojel/visual_task_vectors/blob/HEAD/evaluate/collect_attention_heads.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f82a5154b6295773"}},{"code_sha256_prefix":"ce2513917f2cc4b5","entry":"is_square","repo":"alhojel/visual_task_vectors","repo_kind":"official","path":"evaluate/evaluate_reasoning.py","file_url":"https://github.com/alhojel/visual_task_vectors/blob/HEAD/evaluate/evaluate_reasoning.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ce2513917f2cc4b5"}},{"code_sha256_prefix":"abe146eee6ac945e","entry":"mse","repo":"alhojel/visual_task_vectors","repo_kind":"official","path":"evaluate/collect_attention_heads.py","file_url":"https://github.com/alhojel/visual_task_vectors/blob/HEAD/evaluate/collect_attention_heads.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"abe146eee6ac945e"}},{"code_sha256_prefix":"ebd3af8bde9488f1","entry":"reverse_trans","repo":"alhojel/visual_task_vectors","repo_kind":"official","path":"tta.py","file_url":"https://github.com/alhojel/visual_task_vectors/blob/HEAD/tta.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ebd3af8bde9488f1"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}