{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:OXBJW7AQMOJTMUDH4GQALRKLRW","short_pith_number":"pith:OXBJW7AQ","canonical_record":{"source":{"id":"2502.05291","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-07T19:50:02Z","cross_cats_sorted":[],"title_canon_sha256":"b013774685a05ac30cf92412ecb7079950c992462c7688c5dd45b21abcabe98a","abstract_canon_sha256":"afccfd967496872a2f70a23a4cdb93437774d0c2449b26cf74416be80c472cb5"},"schema_version":"1.0"},"canonical_sha256":"75c29b7c106393365067e1a005c54b8d859c8e3a89a7e02f22610a4dc1452c1c","source":{"kind":"arxiv","id":"2502.05291","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.05291","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"arxiv_version","alias_value":"2502.05291v2","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.05291","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"pith_short_12","alias_value":"OXBJW7AQMOJT","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"pith_short_16","alias_value":"OXBJW7AQMOJTMUDH","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"pith_short_8","alias_value":"OXBJW7AQ","created_at":"2026-07-05T10:51:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:OXBJW7AQMOJTMUDH4GQALRKLRW","target":"record","payload":{"canonical_record":{"source":{"id":"2502.05291","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-07T19:50:02Z","cross_cats_sorted":[],"title_canon_sha256":"b013774685a05ac30cf92412ecb7079950c992462c7688c5dd45b21abcabe98a","abstract_canon_sha256":"afccfd967496872a2f70a23a4cdb93437774d0c2449b26cf74416be80c472cb5"},"schema_version":"1.0"},"canonical_sha256":"75c29b7c106393365067e1a005c54b8d859c8e3a89a7e02f22610a4dc1452c1c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:51:32.421899Z","signature_b64":"uMclb3/XhCRGM8APP0LJkh8WNQ2GFz8zvBbLb++pW4lbv9JbOWXTDix/9FBzSvmrJzCs+7FzsUp5asz7VSZqAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"75c29b7c106393365067e1a005c54b8d859c8e3a89a7e02f22610a4dc1452c1c","last_reissued_at":"2026-07-05T10:51:32.421484Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:51:32.421484Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.05291","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:51:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ib25Dagx2+K8KPywINcJDTM6Y3P384U4chWFER8l692MpU7JiwdDq/wofId+GFDv+uc5z7HY8OkB8sUslAWmBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T09:12:35.468494Z"},"content_sha256":"3ae4146175d9b342a693b6179537a56f641ebe833adc459153d684301aabe765","schema_version":"1.0","event_id":"sha256:3ae4146175d9b342a693b6179537a56f641ebe833adc459153d684301aabe765"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:OXBJW7AQMOJTMUDH4GQALRKLRW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Can LLMs Rank the Harmfulness of Smaller LLMs? We are Not There Yet","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Berk Atil, Rebecca J. Passonneau, Sarkar Snigdha Sarathi Das, Vipul Gupta","submitted_at":"2025-02-07T19:50:02Z","abstract_excerpt":"Large language models (LLMs) have become ubiquitous, thus it is important to understand their risks and limitations. Smaller LLMs can be deployed where compute resources are constrained, such as edge devices, but with different propensity to generate harmful output. Mitigation of LLM harm typically depends on annotating the harmfulness of LLM output, which is expensive to collect from humans. This work studies two questions: How do smaller LLMs rank regarding generation of harmful content? How well can larger LLMs annotate harmfulness? We prompt three small LLMs to elicit harmful content of va"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.05291","kind":"arxiv","version":2},"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/2502.05291/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":"2026-07-05T10:51:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L2KwHahcD0lrVzdu/Db9j+bk9qesCFXoRnvBZdPUYW/j7fybqjEJqLtw20ZHlLdi8hN5e7yWQ80Z1w977WECBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T09:12:35.469334Z"},"content_sha256":"ff0338efff254ad32e2bd3edcdab038e70d21d09eae3ab32b1357ee610b5c710","schema_version":"1.0","event_id":"sha256:ff0338efff254ad32e2bd3edcdab038e70d21d09eae3ab32b1357ee610b5c710"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OXBJW7AQMOJTMUDH4GQALRKLRW/bundle.json","state_url":"https://pith.science/pith/OXBJW7AQMOJTMUDH4GQALRKLRW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OXBJW7AQMOJTMUDH4GQALRKLRW/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":"2026-08-09T09:12:35Z","links":{"resolver":"https://pith.science/pith/OXBJW7AQMOJTMUDH4GQALRKLRW","bundle":"https://pith.science/pith/OXBJW7AQMOJTMUDH4GQALRKLRW/bundle.json","state":"https://pith.science/pith/OXBJW7AQMOJTMUDH4GQALRKLRW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OXBJW7AQMOJTMUDH4GQALRKLRW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OXBJW7AQMOJTMUDH4GQALRKLRW","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":"afccfd967496872a2f70a23a4cdb93437774d0c2449b26cf74416be80c472cb5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-07T19:50:02Z","title_canon_sha256":"b013774685a05ac30cf92412ecb7079950c992462c7688c5dd45b21abcabe98a"},"schema_version":"1.0","source":{"id":"2502.05291","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.05291","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"arxiv_version","alias_value":"2502.05291v2","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.05291","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"pith_short_12","alias_value":"OXBJW7AQMOJT","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"pith_short_16","alias_value":"OXBJW7AQMOJTMUDH","created_at":"2026-07-05T10:51:32Z"},{"alias_kind":"pith_short_8","alias_value":"OXBJW7AQ","created_at":"2026-07-05T10:51:32Z"}],"graph_snapshots":[{"event_id":"sha256:ff0338efff254ad32e2bd3edcdab038e70d21d09eae3ab32b1357ee610b5c710","target":"graph","created_at":"2026-07-05T10:51:32Z","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/2502.05291/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have become ubiquitous, thus it is important to understand their risks and limitations. Smaller LLMs can be deployed where compute resources are constrained, such as edge devices, but with different propensity to generate harmful output. Mitigation of LLM harm typically depends on annotating the harmfulness of LLM output, which is expensive to collect from humans. This work studies two questions: How do smaller LLMs rank regarding generation of harmful content? How well can larger LLMs annotate harmfulness? We prompt three small LLMs to elicit harmful content of va","authors_text":"Berk Atil, Rebecca J. Passonneau, Sarkar Snigdha Sarathi Das, Vipul Gupta","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-07T19:50:02Z","title":"Can LLMs Rank the Harmfulness of Smaller LLMs? We are Not There Yet"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.05291","kind":"arxiv","version":2},"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":"sha256:3ae4146175d9b342a693b6179537a56f641ebe833adc459153d684301aabe765","target":"record","created_at":"2026-07-05T10:51:32Z","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":"afccfd967496872a2f70a23a4cdb93437774d0c2449b26cf74416be80c472cb5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-07T19:50:02Z","title_canon_sha256":"b013774685a05ac30cf92412ecb7079950c992462c7688c5dd45b21abcabe98a"},"schema_version":"1.0","source":{"id":"2502.05291","kind":"arxiv","version":2}},"canonical_sha256":"75c29b7c106393365067e1a005c54b8d859c8e3a89a7e02f22610a4dc1452c1c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"75c29b7c106393365067e1a005c54b8d859c8e3a89a7e02f22610a4dc1452c1c","first_computed_at":"2026-07-05T10:51:32.421484Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:51:32.421484Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uMclb3/XhCRGM8APP0LJkh8WNQ2GFz8zvBbLb++pW4lbv9JbOWXTDix/9FBzSvmrJzCs+7FzsUp5asz7VSZqAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:51:32.421899Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.05291","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3ae4146175d9b342a693b6179537a56f641ebe833adc459153d684301aabe765","sha256:ff0338efff254ad32e2bd3edcdab038e70d21d09eae3ab32b1357ee610b5c710"],"state_sha256":"c0cd6ac1864c69f6e165e878805ee99bc340b8cae345222d84cabb0024fdee0b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"itS602Z1t3n31ZrnjMdTIirCBt5C/XW9+gUygjV0l5Ur5sy56bmBT+CgGnapye1NUbsaE71hUA2V8SBPnKYbAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T09:12:35.484604Z","bundle_sha256":"021de540e3df190d86c76c20e64be9ad7211c1fe820884a1554511e4b10123b9"}}