{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:OHYXGPJFUKTKT3KQPKIGWHHFC3","short_pith_number":"pith:OHYXGPJF","canonical_record":{"source":{"id":"2508.10031","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2025-08-09T02:37:59Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"bb30c3c9ab1cbf27d4ed44be2f25ae5b1ed893fb461f6a92c90471af1f43ec7a","abstract_canon_sha256":"eed69a5d8ba643c09c3b5dcd373728cc85984bfb793e06bea15fb2fdcdaa6abd"},"schema_version":"1.0"},"canonical_sha256":"71f1733d25a2a6a9ed507a906b1ce516fbe4496328bb958aa121d2741916cf7c","source":{"kind":"arxiv","id":"2508.10031","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.10031","created_at":"2026-07-05T11:53:29Z"},{"alias_kind":"arxiv_version","alias_value":"2508.10031v1","created_at":"2026-07-05T11:53:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.10031","created_at":"2026-07-05T11:53:29Z"},{"alias_kind":"pith_short_12","alias_value":"OHYXGPJFUKTK","created_at":"2026-07-05T11:53:29Z"},{"alias_kind":"pith_short_16","alias_value":"OHYXGPJFUKTKT3KQ","created_at":"2026-07-05T11:53:29Z"},{"alias_kind":"pith_short_8","alias_value":"OHYXGPJF","created_at":"2026-07-05T11:53:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:OHYXGPJFUKTKT3KQPKIGWHHFC3","target":"record","payload":{"canonical_record":{"source":{"id":"2508.10031","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2025-08-09T02:37:59Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"bb30c3c9ab1cbf27d4ed44be2f25ae5b1ed893fb461f6a92c90471af1f43ec7a","abstract_canon_sha256":"eed69a5d8ba643c09c3b5dcd373728cc85984bfb793e06bea15fb2fdcdaa6abd"},"schema_version":"1.0"},"canonical_sha256":"71f1733d25a2a6a9ed507a906b1ce516fbe4496328bb958aa121d2741916cf7c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:29.534847Z","signature_b64":"lRm3uz0eVV65POG4miltn6ML95Wcb7KY/cVAgfgI9ApP2q/hOAjEGN5udRpjb+O+mvZuzBnsql/QgxUOdSqxCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"71f1733d25a2a6a9ed507a906b1ce516fbe4496328bb958aa121d2741916cf7c","last_reissued_at":"2026-07-05T11:53:29.534335Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:29.534335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.10031","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":"2026-07-05T11:53:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2C0GchSVlx5D6No6f1cEPz6HdXn15fMnA1MlLvZXGHZQoc20zp0ElbeEj7ZcNiw+TFFQypdObb+gxaAyJQVcDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:36:17.081821Z"},"content_sha256":"b1ba244fbe8412c68f93ce0703d2bba4122138220814db3e974331c0dc22d9b0","schema_version":"1.0","event_id":"sha256:b1ba244fbe8412c68f93ce0703d2bba4122138220814db3e974331c0dc22d9b0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:OHYXGPJFUKTKT3KQPKIGWHHFC3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Context Misleads LLMs: The Role of Context Filtering in Maintaining Safe Alignment of LLMs","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CR","authors_text":"Ian G. Harris, Jinhwa Kim","submitted_at":"2025-08-09T02:37:59Z","abstract_excerpt":"While Large Language Models (LLMs) have shown significant advancements in performance, various jailbreak attacks have posed growing safety and ethical risks. Malicious users often exploit adversarial context to deceive LLMs, prompting them to generate responses to harmful queries. In this study, we propose a new defense mechanism called Context Filtering model, an input pre-processing method designed to filter out untrustworthy and unreliable context while identifying the primary prompts containing the real user intent to uncover concealed malicious intent. Given that enhancing the safety of L"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.10031","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/2508.10031/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":"2026-07-05T11:53:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rBvlxCjDpmdo0+Yl6J/c1K2grKSJHR+xJC2ZBfQUu+bCE+AnckwThwVWdMtH6l3U40noWqjwWLQsjM9xlyZzDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:36:17.082494Z"},"content_sha256":"eeebf50b1cbb5951a07316f004b384b241e074af1f32a8eb183cd2adbc097691","schema_version":"1.0","event_id":"sha256:eeebf50b1cbb5951a07316f004b384b241e074af1f32a8eb183cd2adbc097691"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OHYXGPJFUKTKT3KQPKIGWHHFC3/bundle.json","state_url":"https://pith.science/pith/OHYXGPJFUKTKT3KQPKIGWHHFC3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OHYXGPJFUKTKT3KQPKIGWHHFC3/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":"2026-07-07T06:36:17Z","links":{"resolver":"https://pith.science/pith/OHYXGPJFUKTKT3KQPKIGWHHFC3","bundle":"https://pith.science/pith/OHYXGPJFUKTKT3KQPKIGWHHFC3/bundle.json","state":"https://pith.science/pith/OHYXGPJFUKTKT3KQPKIGWHHFC3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OHYXGPJFUKTKT3KQPKIGWHHFC3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OHYXGPJFUKTKT3KQPKIGWHHFC3","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":"eed69a5d8ba643c09c3b5dcd373728cc85984bfb793e06bea15fb2fdcdaa6abd","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2025-08-09T02:37:59Z","title_canon_sha256":"bb30c3c9ab1cbf27d4ed44be2f25ae5b1ed893fb461f6a92c90471af1f43ec7a"},"schema_version":"1.0","source":{"id":"2508.10031","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.10031","created_at":"2026-07-05T11:53:29Z"},{"alias_kind":"arxiv_version","alias_value":"2508.10031v1","created_at":"2026-07-05T11:53:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.10031","created_at":"2026-07-05T11:53:29Z"},{"alias_kind":"pith_short_12","alias_value":"OHYXGPJFUKTK","created_at":"2026-07-05T11:53:29Z"},{"alias_kind":"pith_short_16","alias_value":"OHYXGPJFUKTKT3KQ","created_at":"2026-07-05T11:53:29Z"},{"alias_kind":"pith_short_8","alias_value":"OHYXGPJF","created_at":"2026-07-05T11:53:29Z"}],"graph_snapshots":[{"event_id":"sha256:eeebf50b1cbb5951a07316f004b384b241e074af1f32a8eb183cd2adbc097691","target":"graph","created_at":"2026-07-05T11:53:29Z","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/2508.10031/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While Large Language Models (LLMs) have shown significant advancements in performance, various jailbreak attacks have posed growing safety and ethical risks. Malicious users often exploit adversarial context to deceive LLMs, prompting them to generate responses to harmful queries. In this study, we propose a new defense mechanism called Context Filtering model, an input pre-processing method designed to filter out untrustworthy and unreliable context while identifying the primary prompts containing the real user intent to uncover concealed malicious intent. Given that enhancing the safety of L","authors_text":"Ian G. Harris, Jinhwa Kim","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2025-08-09T02:37:59Z","title":"Context Misleads LLMs: The Role of Context Filtering in Maintaining Safe Alignment of LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.10031","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:b1ba244fbe8412c68f93ce0703d2bba4122138220814db3e974331c0dc22d9b0","target":"record","created_at":"2026-07-05T11:53:29Z","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":"eed69a5d8ba643c09c3b5dcd373728cc85984bfb793e06bea15fb2fdcdaa6abd","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2025-08-09T02:37:59Z","title_canon_sha256":"bb30c3c9ab1cbf27d4ed44be2f25ae5b1ed893fb461f6a92c90471af1f43ec7a"},"schema_version":"1.0","source":{"id":"2508.10031","kind":"arxiv","version":1}},"canonical_sha256":"71f1733d25a2a6a9ed507a906b1ce516fbe4496328bb958aa121d2741916cf7c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"71f1733d25a2a6a9ed507a906b1ce516fbe4496328bb958aa121d2741916cf7c","first_computed_at":"2026-07-05T11:53:29.534335Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:53:29.534335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lRm3uz0eVV65POG4miltn6ML95Wcb7KY/cVAgfgI9ApP2q/hOAjEGN5udRpjb+O+mvZuzBnsql/QgxUOdSqxCg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:53:29.534847Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.10031","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b1ba244fbe8412c68f93ce0703d2bba4122138220814db3e974331c0dc22d9b0","sha256:eeebf50b1cbb5951a07316f004b384b241e074af1f32a8eb183cd2adbc097691"],"state_sha256":"27de4b96404e3ff522a59d84b6db6ce1b69a052c9ceb15c69bb98add6ed913b5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8coQtTvmMe2vnR69rXHBYy5ExlU8ZMVZewn4K71upCKLaIcVp+OscKsjHcBaBnm6fmHvlbqXJVlXKsCVbo46Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T06:36:17.086053Z","bundle_sha256":"ef900ad6720f9682d4374e2afeac492d8a92c94f06698951cf786a97494b011b"}}