{"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/amplegcg-plus-a-strong-generative-model-of","title":"AmpleGCG-Plus: A Strong Generative Model of Adversarial Suffixes to Jailbreak LLMs with Higher Success Rates in Fewer Attempts","arxiv_id":"2410.22143","date":"2024-10-29","proceeding":null,"authors":["Vishal Kumar","Zeyi Liao","Jaylen Jones","Huan Sun"],"abstract":"Although large language models (LLMs) are typically aligned, they remain vulnerable to jailbreaking through either carefully crafted prompts in natural language or, interestingly, gibberish adversarial suffixes. However, gibberish tokens have received relatively less attention despite their success in attacking aligned LLMs. Recent work, AmpleGCG~\\citep{liao2024amplegcg}, demonstrates that a generative model can quickly produce numerous customizable gibberish adversarial suffixes for any harmful query, exposing a range of alignment gaps in out-of-distribution (OOD) language spaces. To bring more attention to this area, we introduce AmpleGCG-Plus, an enhanced version that achieves better performance in fewer attempts. Through a series of exploratory experiments, we identify several training strategies to improve the learning of gibberish suffixes. Our results, verified under a strict evaluation setting, show that it outperforms AmpleGCG on both open-weight and closed-source models, achieving increases in attack success rate (ASR) of up to 17\\% in the white-box setting against Llama-2-7B-chat, and more than tripling ASR in the black-box setting against GPT-4. Notably, AmpleGCG-Plus jailbreaks the newer GPT-4o series of models at similar rates to GPT-4, and, uncovers vulnerabilities against the recently proposed circuit breakers defense. We publicly release AmpleGCG-Plus along with our collected training datasets.","url_abs":"https://arxiv.org/abs/2410.22143v1","url_pdf":"https://arxiv.org/pdf/2410.22143v1.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":"amplegcg-plus-a-strong-generative-model-of","repo_url":"https://github.com/osu-nlp-group/amplegcg","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-4","method_name":"GPT-4"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2410.22143","atlas_url":"https://app.syntology.ai/?focus=2410.22143","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.22143"}},"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/osu-nlp-group/amplegcg","reach":{"status":"ok","spdx":"NOASSERTION"}}],"summary":{"unverified":7},"by_repo_kind":{"listed":{"samples":7,"ran":0,"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":7,"samples":[{"code_sha256_prefix":"4bca82dd290d2fe8","entry":"attack_collate_fn","repo":"osu-nlp-group/amplegcg","repo_kind":"listed","path":"add_reward.py","file_url":"https://github.com/osu-nlp-group/amplegcg/blob/HEAD/add_reward.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"4bca82dd290d2fe8"}},{"code_sha256_prefix":"5ac829d1f5eef4b5","entry":"cal_loss_avg","repo":"osu-nlp-group/amplegcg","repo_kind":"listed","path":"lm_components.py","file_url":"https://github.com/osu-nlp-group/amplegcg/blob/HEAD/lm_components.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"5ac829d1f5eef4b5"}},{"code_sha256_prefix":"fd7e04be620ed414","entry":"check_torch_dtype","repo":"osu-nlp-group/amplegcg","repo_kind":"listed","path":"lm_components.py","file_url":"https://github.com/osu-nlp-group/amplegcg/blob/HEAD/lm_components.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"fd7e04be620ed414"}},{"code_sha256_prefix":"7e8884a0be4b5571","entry":"deter_if_harm","repo":"osu-nlp-group/amplegcg","repo_kind":"listed","path":"utility.py","file_url":"https://github.com/osu-nlp-group/amplegcg/blob/HEAD/utility.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"7e8884a0be4b5571"}},{"code_sha256_prefix":"3270be5954fd31b4","entry":"deter_if_harm_new","repo":"osu-nlp-group/amplegcg","repo_kind":"listed","path":"utility.py","file_url":"https://github.com/osu-nlp-group/amplegcg/blob/HEAD/utility.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"3270be5954fd31b4"}},{"code_sha256_prefix":"caed583620387c3a","entry":"repeat_texts_l","repo":"osu-nlp-group/amplegcg","repo_kind":"listed","path":"evaluate_augmentedGCG.py","file_url":"https://github.com/osu-nlp-group/amplegcg/blob/HEAD/evaluate_augmentedGCG.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"caed583620387c3a"}},{"code_sha256_prefix":"7004723aa7319a84","entry":"unique_random_concat_combinations","repo":"osu-nlp-group/amplegcg","repo_kind":"listed","path":"evaluate_augmentedGCG.py","file_url":"https://github.com/osu-nlp-group/amplegcg/blob/HEAD/evaluate_augmentedGCG.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"7004723aa7319a84"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}