{"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/soft-prompts-go-hard-steering-visual-language","title":"Self-interpreting Adversarial Images","arxiv_id":"2407.08970","date":"2024-07-12","proceeding":null,"authors":["Tingwei Zhang","Collin Zhang","John X. Morris","Eugene Bagdasarian","Vitaly Shmatikov"],"abstract":"We introduce a new type of indirect, cross-modal injection attacks against visual language models that enable creation of self-interpreting images. These images contain hidden \"meta-instructions\" that control how models answer users' questions about the image and steer models' outputs to express an adversary-chosen style, sentiment, or point of view. Self-interpreting images act as soft prompts, conditioning the model to satisfy the adversary's (meta-)objective while still producing answers based on the image's visual content. Meta-instructions are thus a stronger form of prompt injection. Adversarial images look natural and the model's answers are coherent and plausible, yet they also follow the adversary-chosen interpretation, e.g., political spin, or even objectives that are not achievable with explicit text instructions. We evaluate the efficacy of self-interpreting images for a variety of models, interpretations, and user prompts. We describe how these attacks could cause harm by enabling creation of self-interpreting content that carries spam, misinformation, or spin. Finally, we discuss defenses.","url_abs":"https://arxiv.org/abs/2407.08970v4","url_pdf":"https://arxiv.org/pdf/2407.08970v4.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":"soft-prompts-go-hard-steering-visual-language","repo_url":"https://github.com/tingwei-zhang/soft-prompts-go-hard","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"misinformation","task_name":"Misinformation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2407.08970","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2407.08970"}},"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/tingwei-zhang/soft-prompts-go-hard","reach":null}],"summary":{"ran_draft_wrong":3},"by_repo_kind":{"official":{"samples":3,"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":"3ce7639c17130092","entry":"get_prompt_template","repo":"tingwei-zhang/soft-prompts-go-hard","repo_kind":"official","path":"instructblip_inference.py","file_url":"https://github.com/tingwei-zhang/soft-prompts-go-hard/blob/HEAD/instructblip_inference.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3ce7639c17130092"}},{"code_sha256_prefix":"01650840799a0eef","entry":"read_csv_data","repo":"tingwei-zhang/soft-prompts-go-hard","repo_kind":"official","path":"instructblip_inference.py","file_url":"https://github.com/tingwei-zhang/soft-prompts-go-hard/blob/HEAD/instructblip_inference.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"01650840799a0eef"}},{"code_sha256_prefix":"11e862b24a0f6df0","entry":"read_csv_data","repo":"tingwei-zhang/soft-prompts-go-hard","repo_kind":"official","path":"instructblip_visual_attack.py","file_url":"https://github.com/tingwei-zhang/soft-prompts-go-hard/blob/HEAD/instructblip_visual_attack.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"11e862b24a0f6df0"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}