{"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/circumventing-concept-erasure-methods-for","title":"Circumventing Concept Erasure Methods For Text-to-Image Generative Models","arxiv_id":"2308.01508","date":"2023-08-03","proceeding":null,"authors":["Minh Pham","Kelly O. Marshall","Niv Cohen","Govind Mittal","Chinmay Hegde"],"abstract":"Text-to-image generative models can produce photo-realistic images for an extremely broad range of concepts, and their usage has proliferated widely among the general public. On the flip side, these models have numerous drawbacks, including their potential to generate images featuring sexually explicit content, mirror artistic styles without permission, or even hallucinate (or deepfake) the likenesses of celebrities. Consequently, various methods have been proposed in order to \"erase\" sensitive concepts from text-to-image models. In this work, we examine five recently proposed concept erasure methods, and show that targeted concepts are not fully excised from any of these methods. Specifically, we leverage the existence of special learned word embeddings that can retrieve \"erased\" concepts from the sanitized models with no alterations to their weights. Our results highlight the brittleness of post hoc concept erasure methods, and call into question their use in the algorithmic toolkit for AI safety.","url_abs":"https://arxiv.org/abs/2308.01508v2","url_pdf":"https://arxiv.org/pdf/2308.01508v2.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":"circumventing-concept-erasure-methods-for","repo_url":"https://github.com/nyu-dice-lab/circumventing-concept-erasure","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"face-swapping","task_name":"Face Swapping"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"}],"methods":[{"method_slug":"flip","method_name":"FLIP"},{"method_slug":"hoc","method_name":"HOC"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2308.01508","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2308.01508"}},"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/nyu-dice-lab/circumventing-concept-erasure","reach":{"status":"ok"}}],"summary":{"ran":1,"unverified":1},"by_repo_kind":{"official":{"samples":2,"ran":1,"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":2,"samples":[{"code_sha256_prefix":"023a563c50d5db53","entry":"save_image","repo":"nyu-dice-lab/circumventing-concept-erasure","repo_kind":"official","path":"ac/utils.py","file_url":"https://github.com/nyu-dice-lab/circumventing-concept-erasure/blob/HEAD/ac/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"023a563c50d5db53"}},{"code_sha256_prefix":"dd9576b7def71e1e","entry":"get_random_prompt","repo":"nyu-dice-lab/circumventing-concept-erasure","repo_kind":"official","path":"ac/utils.py","file_url":"https://github.com/nyu-dice-lab/circumventing-concept-erasure/blob/HEAD/ac/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":"dd9576b7def71e1e"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}