{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:YQTEBTQA7AFIEIRVBW4H4NPHMM","short_pith_number":"pith:YQTEBTQA","canonical_record":{"source":{"id":"2606.00160","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2026-05-29T09:04:27Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"78cd9ec7c817493668dbf5d2459b7e83c92c6a933c18284145735c101066ac02","abstract_canon_sha256":"35970e1747bb7dfbc4dbce08f66697169f395c37de5e390e382f5e9bf429367f"},"schema_version":"1.0"},"canonical_sha256":"c42640ce00f80a8222350db87e35e7630fb63f4f54a28aa071c1ef93dd0da80e","source":{"kind":"arxiv","id":"2606.00160","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.00160","created_at":"2026-06-02T01:03:20Z"},{"alias_kind":"arxiv_version","alias_value":"2606.00160v1","created_at":"2026-06-02T01:03:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00160","created_at":"2026-06-02T01:03:20Z"},{"alias_kind":"pith_short_12","alias_value":"YQTEBTQA7AFI","created_at":"2026-06-02T01:03:20Z"},{"alias_kind":"pith_short_16","alias_value":"YQTEBTQA7AFIEIRV","created_at":"2026-06-02T01:03:20Z"},{"alias_kind":"pith_short_8","alias_value":"YQTEBTQA","created_at":"2026-06-02T01:03:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:YQTEBTQA7AFIEIRVBW4H4NPHMM","target":"record","payload":{"canonical_record":{"source":{"id":"2606.00160","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2026-05-29T09:04:27Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"78cd9ec7c817493668dbf5d2459b7e83c92c6a933c18284145735c101066ac02","abstract_canon_sha256":"35970e1747bb7dfbc4dbce08f66697169f395c37de5e390e382f5e9bf429367f"},"schema_version":"1.0"},"canonical_sha256":"c42640ce00f80a8222350db87e35e7630fb63f4f54a28aa071c1ef93dd0da80e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:20.153702Z","signature_b64":"fqmY5Jh72WkfSzF43duFvynpyFX/YR5t+DmjAh+rg3mJAD03CJ95I2VjAUvVaXTZErsNHx4zMpCC6ZzgI874DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c42640ce00f80a8222350db87e35e7630fb63f4f54a28aa071c1ef93dd0da80e","last_reissued_at":"2026-06-02T01:03:20.153272Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:20.153272Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.00160","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":"2026-06-02T01:03:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5HX/YS/P6jTB+4xU8mRbd5eWKh+IpMs50MwodfIz412jWphlir5w74VP2nKmKYMe3Xkyazdb4VeNpR/TCamwBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T06:18:56.093924Z"},"content_sha256":"7daa27a2cbd07160dc49d4717a630449b44333c61ed77db6e27fb896f23bb862","schema_version":"1.0","event_id":"sha256:7daa27a2cbd07160dc49d4717a630449b44333c61ed77db6e27fb896f23bb862"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:YQTEBTQA7AFIEIRVBW4H4NPHMM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DataShield: Safety-degrading Data Filtering for LLM Benign Instruction Fine-Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CR","authors_text":"Jie Pan, Jinbiao Zhu, Junbo Zhang, Qianli Zhou, Wen Jiang, Xinyang Deng","submitted_at":"2026-05-29T09:04:27Z","abstract_excerpt":"Large language models (LLMs) suffer from degraded safety capabilities even when fine-tuned with benign datasets. However, existing methods for identifying safety-degrading samples in benign datasets suffer from high computational costs and significant noise issues. In this paper, we propose DataShield to efficiently and effectively identify potential safety-degrading samples. Our key intuition is based on the observation that benign fine-tuning increases the overall response compliance of LLMs. DataShield's key technical insight is to quantify each sample's contribution to the model's complian"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00160","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/2606.00160/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":"2026-06-02T01:03:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Hk1sPQDHc7l2r1IE2WiWZ6IkQOpvKB0qeUbf4G2L0q7oID2aotf83mJuWVRv0ugmubucXUlboEdxCPTeoyw1Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T06:18:56.094571Z"},"content_sha256":"4ce0cbd2cce1f4e1cd56cac0f939ef1192eddaba120ccd1b3ce9e5ad7ea89788","schema_version":"1.0","event_id":"sha256:4ce0cbd2cce1f4e1cd56cac0f939ef1192eddaba120ccd1b3ce9e5ad7ea89788"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YQTEBTQA7AFIEIRVBW4H4NPHMM/bundle.json","state_url":"https://pith.science/pith/YQTEBTQA7AFIEIRVBW4H4NPHMM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YQTEBTQA7AFIEIRVBW4H4NPHMM/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":"2026-06-04T06:18:56Z","links":{"resolver":"https://pith.science/pith/YQTEBTQA7AFIEIRVBW4H4NPHMM","bundle":"https://pith.science/pith/YQTEBTQA7AFIEIRVBW4H4NPHMM/bundle.json","state":"https://pith.science/pith/YQTEBTQA7AFIEIRVBW4H4NPHMM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YQTEBTQA7AFIEIRVBW4H4NPHMM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:YQTEBTQA7AFIEIRVBW4H4NPHMM","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":"35970e1747bb7dfbc4dbce08f66697169f395c37de5e390e382f5e9bf429367f","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2026-05-29T09:04:27Z","title_canon_sha256":"78cd9ec7c817493668dbf5d2459b7e83c92c6a933c18284145735c101066ac02"},"schema_version":"1.0","source":{"id":"2606.00160","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.00160","created_at":"2026-06-02T01:03:20Z"},{"alias_kind":"arxiv_version","alias_value":"2606.00160v1","created_at":"2026-06-02T01:03:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00160","created_at":"2026-06-02T01:03:20Z"},{"alias_kind":"pith_short_12","alias_value":"YQTEBTQA7AFI","created_at":"2026-06-02T01:03:20Z"},{"alias_kind":"pith_short_16","alias_value":"YQTEBTQA7AFIEIRV","created_at":"2026-06-02T01:03:20Z"},{"alias_kind":"pith_short_8","alias_value":"YQTEBTQA","created_at":"2026-06-02T01:03:20Z"}],"graph_snapshots":[{"event_id":"sha256:4ce0cbd2cce1f4e1cd56cac0f939ef1192eddaba120ccd1b3ce9e5ad7ea89788","target":"graph","created_at":"2026-06-02T01:03:20Z","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/2606.00160/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) suffer from degraded safety capabilities even when fine-tuned with benign datasets. However, existing methods for identifying safety-degrading samples in benign datasets suffer from high computational costs and significant noise issues. In this paper, we propose DataShield to efficiently and effectively identify potential safety-degrading samples. Our key intuition is based on the observation that benign fine-tuning increases the overall response compliance of LLMs. DataShield's key technical insight is to quantify each sample's contribution to the model's complian","authors_text":"Jie Pan, Jinbiao Zhu, Junbo Zhang, Qianli Zhou, Wen Jiang, Xinyang Deng","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2026-05-29T09:04:27Z","title":"DataShield: Safety-degrading Data Filtering for LLM Benign Instruction Fine-Tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00160","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":"sha256:7daa27a2cbd07160dc49d4717a630449b44333c61ed77db6e27fb896f23bb862","target":"record","created_at":"2026-06-02T01:03:20Z","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":"35970e1747bb7dfbc4dbce08f66697169f395c37de5e390e382f5e9bf429367f","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2026-05-29T09:04:27Z","title_canon_sha256":"78cd9ec7c817493668dbf5d2459b7e83c92c6a933c18284145735c101066ac02"},"schema_version":"1.0","source":{"id":"2606.00160","kind":"arxiv","version":1}},"canonical_sha256":"c42640ce00f80a8222350db87e35e7630fb63f4f54a28aa071c1ef93dd0da80e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c42640ce00f80a8222350db87e35e7630fb63f4f54a28aa071c1ef93dd0da80e","first_computed_at":"2026-06-02T01:03:20.153272Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:03:20.153272Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fqmY5Jh72WkfSzF43duFvynpyFX/YR5t+DmjAh+rg3mJAD03CJ95I2VjAUvVaXTZErsNHx4zMpCC6ZzgI874DA==","signature_status":"signed_v1","signed_at":"2026-06-02T01:03:20.153702Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.00160","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7daa27a2cbd07160dc49d4717a630449b44333c61ed77db6e27fb896f23bb862","sha256:4ce0cbd2cce1f4e1cd56cac0f939ef1192eddaba120ccd1b3ce9e5ad7ea89788"],"state_sha256":"3d75beed08d0af391d21d65d5fabb36fb0e475d4c0653db7530aa9096bed8af2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MeWHKRSHv9P01ugj4YBj1c+GQ3tOMUk9QMqx6BGJopKcIrf9vL8UkcFrieoD3APsq41bdAgmebFUTdA+e1hRBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-04T06:18:56.097162Z","bundle_sha256":"e02a1b7e8b893e702c3027e70c50c68bed8c1d034080c6d4f112b7c84e42ad7e"}}