{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:CVFSLXIWN6T4DUFAKIP6VGGPMD","short_pith_number":"pith:CVFSLXIW","canonical_record":{"source":{"id":"2509.00391","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-30T07:04:29Z","cross_cats_sorted":["cs.AI","cs.CR","cs.LG"],"title_canon_sha256":"18cb1bb359cffd72f2f8221243fd34f4211ee804dc9b35e54004368852ec1c87","abstract_canon_sha256":"3b7828974bb846826035d97293cb1cd8696afa9bf1924bb05d450ebe11379d1f"},"schema_version":"1.0"},"canonical_sha256":"154b25dd166fa7c1d0a0521fea98cf60cddffe83256d430390c4fc1e2b07706b","source":{"kind":"arxiv","id":"2509.00391","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.00391","created_at":"2026-07-05T12:02:15Z"},{"alias_kind":"arxiv_version","alias_value":"2509.00391v1","created_at":"2026-07-05T12:02:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.00391","created_at":"2026-07-05T12:02:15Z"},{"alias_kind":"pith_short_12","alias_value":"CVFSLXIWN6T4","created_at":"2026-07-05T12:02:15Z"},{"alias_kind":"pith_short_16","alias_value":"CVFSLXIWN6T4DUFA","created_at":"2026-07-05T12:02:15Z"},{"alias_kind":"pith_short_8","alias_value":"CVFSLXIW","created_at":"2026-07-05T12:02:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:CVFSLXIWN6T4DUFAKIP6VGGPMD","target":"record","payload":{"canonical_record":{"source":{"id":"2509.00391","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-30T07:04:29Z","cross_cats_sorted":["cs.AI","cs.CR","cs.LG"],"title_canon_sha256":"18cb1bb359cffd72f2f8221243fd34f4211ee804dc9b35e54004368852ec1c87","abstract_canon_sha256":"3b7828974bb846826035d97293cb1cd8696afa9bf1924bb05d450ebe11379d1f"},"schema_version":"1.0"},"canonical_sha256":"154b25dd166fa7c1d0a0521fea98cf60cddffe83256d430390c4fc1e2b07706b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:02:15.341693Z","signature_b64":"SzfxHMKSQ9Dx8ZERIOtZ/BBd0mdYmrBxK8/ZvjEZN2f369in61OnZzAhjaCUIx/DuHzrSX3BFgrH4K+tEz4QDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"154b25dd166fa7c1d0a0521fea98cf60cddffe83256d430390c4fc1e2b07706b","last_reissued_at":"2026-07-05T12:02:15.341192Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:02:15.341192Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.00391","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T12:02:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4y7IIBGVuD+IYKZi1lPDGX9aSNcAb2QNhuevkAOl0qzNGv7pn0p4YmGaSJUmtFym3FkHZz+wQWZ5DH9/y/oIDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T04:24:20.377765Z"},"content_sha256":"3b5eb00143ca40711f41cabdf01882b9dd22433ee43ff7ad0d123bdcff48a856","schema_version":"1.0","event_id":"sha256:3b5eb00143ca40711f41cabdf01882b9dd22433ee43ff7ad0d123bdcff48a856"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:CVFSLXIWN6T4DUFAKIP6VGGPMD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Resurgence of GCG Adversarial Attacks on Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Huizhen Shu, Peikang Hu, Xuying Li, Yuting Tan, Zhuo Li","submitted_at":"2025-08-30T07:04:29Z","abstract_excerpt":"Gradient-based adversarial prompting, such as the Greedy Coordinate Gradient (GCG) algorithm, has emerged as a powerful method for jailbreaking large language models (LLMs). In this paper, we present a systematic appraisal of GCG and its annealing-augmented variant, T-GCG, across open-source LLMs of varying scales. Using Qwen2.5-0.5B, LLaMA-3.2-1B, and GPT-OSS-20B, we evaluate attack effectiveness on both safety-oriented prompts (AdvBench) and reasoning-intensive coding prompts. Our study reveals three key findings: (1) attack success rates (ASR) decrease with model size, reflecting the increa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.00391","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2509.00391/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T12:02:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JJ9q/ALMfpO/sGTLTOIYipqiCwLK6aCs6yUQ4NZ3JgtxuzwUttUq8f7iXN/OCbPpbjWsBBdNONiqHp6s89/1Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T04:24:20.378262Z"},"content_sha256":"6bef362ecd2534e2209f31f447fe39f45009736cddbc15978852e4b24c1e24d0","schema_version":"1.0","event_id":"sha256:6bef362ecd2534e2209f31f447fe39f45009736cddbc15978852e4b24c1e24d0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CVFSLXIWN6T4DUFAKIP6VGGPMD/bundle.json","state_url":"https://pith.science/pith/CVFSLXIWN6T4DUFAKIP6VGGPMD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CVFSLXIWN6T4DUFAKIP6VGGPMD/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T04:24:20Z","links":{"resolver":"https://pith.science/pith/CVFSLXIWN6T4DUFAKIP6VGGPMD","bundle":"https://pith.science/pith/CVFSLXIWN6T4DUFAKIP6VGGPMD/bundle.json","state":"https://pith.science/pith/CVFSLXIWN6T4DUFAKIP6VGGPMD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CVFSLXIWN6T4DUFAKIP6VGGPMD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:CVFSLXIWN6T4DUFAKIP6VGGPMD","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"3b7828974bb846826035d97293cb1cd8696afa9bf1924bb05d450ebe11379d1f","cross_cats_sorted":["cs.AI","cs.CR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-30T07:04:29Z","title_canon_sha256":"18cb1bb359cffd72f2f8221243fd34f4211ee804dc9b35e54004368852ec1c87"},"schema_version":"1.0","source":{"id":"2509.00391","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.00391","created_at":"2026-07-05T12:02:15Z"},{"alias_kind":"arxiv_version","alias_value":"2509.00391v1","created_at":"2026-07-05T12:02:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.00391","created_at":"2026-07-05T12:02:15Z"},{"alias_kind":"pith_short_12","alias_value":"CVFSLXIWN6T4","created_at":"2026-07-05T12:02:15Z"},{"alias_kind":"pith_short_16","alias_value":"CVFSLXIWN6T4DUFA","created_at":"2026-07-05T12:02:15Z"},{"alias_kind":"pith_short_8","alias_value":"CVFSLXIW","created_at":"2026-07-05T12:02:15Z"}],"graph_snapshots":[{"event_id":"sha256:6bef362ecd2534e2209f31f447fe39f45009736cddbc15978852e4b24c1e24d0","target":"graph","created_at":"2026-07-05T12:02:15Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2509.00391/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Gradient-based adversarial prompting, such as the Greedy Coordinate Gradient (GCG) algorithm, has emerged as a powerful method for jailbreaking large language models (LLMs). In this paper, we present a systematic appraisal of GCG and its annealing-augmented variant, T-GCG, across open-source LLMs of varying scales. Using Qwen2.5-0.5B, LLaMA-3.2-1B, and GPT-OSS-20B, we evaluate attack effectiveness on both safety-oriented prompts (AdvBench) and reasoning-intensive coding prompts. Our study reveals three key findings: (1) attack success rates (ASR) decrease with model size, reflecting the increa","authors_text":"Huizhen Shu, Peikang Hu, Xuying Li, Yuting Tan, Zhuo Li","cross_cats":["cs.AI","cs.CR","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-30T07:04:29Z","title":"The Resurgence of GCG Adversarial Attacks on Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.00391","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3b5eb00143ca40711f41cabdf01882b9dd22433ee43ff7ad0d123bdcff48a856","target":"record","created_at":"2026-07-05T12:02:15Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"3b7828974bb846826035d97293cb1cd8696afa9bf1924bb05d450ebe11379d1f","cross_cats_sorted":["cs.AI","cs.CR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-30T07:04:29Z","title_canon_sha256":"18cb1bb359cffd72f2f8221243fd34f4211ee804dc9b35e54004368852ec1c87"},"schema_version":"1.0","source":{"id":"2509.00391","kind":"arxiv","version":1}},"canonical_sha256":"154b25dd166fa7c1d0a0521fea98cf60cddffe83256d430390c4fc1e2b07706b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"154b25dd166fa7c1d0a0521fea98cf60cddffe83256d430390c4fc1e2b07706b","first_computed_at":"2026-07-05T12:02:15.341192Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:02:15.341192Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SzfxHMKSQ9Dx8ZERIOtZ/BBd0mdYmrBxK8/ZvjEZN2f369in61OnZzAhjaCUIx/DuHzrSX3BFgrH4K+tEz4QDw==","signature_status":"signed_v1","signed_at":"2026-07-05T12:02:15.341693Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.00391","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3b5eb00143ca40711f41cabdf01882b9dd22433ee43ff7ad0d123bdcff48a856","sha256:6bef362ecd2534e2209f31f447fe39f45009736cddbc15978852e4b24c1e24d0"],"state_sha256":"a053ca93542f0616055dcbf655bdf3fa547aa78f494c59c6ef08399ca0be5ccb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GJCUSLocComLKfBCAkYz478pkZbWP+a9wWAJjof8TrlissRrTnygBejKiJTxdbb+mPRyH2UFA41mhSlRLbFnBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T04:24:20.382242Z","bundle_sha256":"0b9281c157ff9d1f9b1d865250ddfa826651b7c17e5ffa73664a4a567965e40d"}}