{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:C5WJPCKWAARPC5EXUDP7KSR6YM","short_pith_number":"pith:C5WJPCKW","canonical_record":{"source":{"id":"2404.01295","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-01T17:59:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"eb70e90787034bf309e745f454b724b0c156d4e48af8b610abb2aaf9837358c0","abstract_canon_sha256":"87ac016f90203277f5a54b1fb78114a0ef9f767fb2d40c49d9da6bff22206efa"},"schema_version":"1.0"},"canonical_sha256":"176c9789560022f17497a0dff54a3ec33aed5f01761e094d4a4df549a960c17c","source":{"kind":"arxiv","id":"2404.01295","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.01295","created_at":"2026-07-05T08:03:04Z"},{"alias_kind":"arxiv_version","alias_value":"2404.01295v1","created_at":"2026-07-05T08:03:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.01295","created_at":"2026-07-05T08:03:04Z"},{"alias_kind":"pith_short_12","alias_value":"C5WJPCKWAARP","created_at":"2026-07-05T08:03:04Z"},{"alias_kind":"pith_short_16","alias_value":"C5WJPCKWAARPC5EX","created_at":"2026-07-05T08:03:04Z"},{"alias_kind":"pith_short_8","alias_value":"C5WJPCKW","created_at":"2026-07-05T08:03:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:C5WJPCKWAARPC5EXUDP7KSR6YM","target":"record","payload":{"canonical_record":{"source":{"id":"2404.01295","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-01T17:59:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"eb70e90787034bf309e745f454b724b0c156d4e48af8b610abb2aaf9837358c0","abstract_canon_sha256":"87ac016f90203277f5a54b1fb78114a0ef9f767fb2d40c49d9da6bff22206efa"},"schema_version":"1.0"},"canonical_sha256":"176c9789560022f17497a0dff54a3ec33aed5f01761e094d4a4df549a960c17c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:03:04.155648Z","signature_b64":"bSMQtvyh4Dvc+Hqdjt5E4Eh5rHpWquAcnRsRR4E8NmjkzzPG5QYxHZDZuCP+u6lj2cstZgsM0LqNvFgiIuCoAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"176c9789560022f17497a0dff54a3ec33aed5f01761e094d4a4df549a960c17c","last_reissued_at":"2026-07-05T08:03:04.155149Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:03:04.155149Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.01295","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":"2026-07-05T08:03:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VSVTwdErBWAKIuKYtrCfd7UmdCCb6UU7nBMAkBP4+OEf9nhtIh7S7K6u+gRidf/XFGppaoE+jsoHTu60vrMlDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-10T21:10:50.604172Z"},"content_sha256":"7221e57a54433c0e9a56ec3accfab57951917517c60cb48745c7f8759a2c4f9f","schema_version":"1.0","event_id":"sha256:7221e57a54433c0e9a56ec3accfab57951917517c60cb48745c7f8759a2c4f9f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:C5WJPCKWAARPC5EXUDP7KSR6YM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Towards Safety and Helpfulness Balanced Responses via Controllable Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Asli Celikyilmaz, Daniel M. Bikel, Eric Michael Smith, Louis Martin, Soumya Batra, William Yang Wang, Xilun Chen, Yi-Lin Tuan","submitted_at":"2024-04-01T17:59:06Z","abstract_excerpt":"As large language models (LLMs) become easily accessible nowadays, the trade-off between safety and helpfulness can significantly impact user experience. A model that prioritizes safety will cause users to feel less engaged and assisted while prioritizing helpfulness will potentially cause harm. Possible harms include teaching people how to build a bomb, exposing youth to inappropriate content, and hurting users' mental health. In this work, we propose to balance safety and helpfulness in diverse use cases by controlling both attributes in LLM. We explore training-free and fine-tuning methods "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.01295","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/2404.01295/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":"2026-07-05T08:03:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"weLBgO5X5nPZiaSQicyap1DfGWqdjiTmnYKef9WuNFWdv9RGcz5xB8yj6/OzyEI4B5ket6xM70u79hUEvmWABQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-10T21:10:50.604562Z"},"content_sha256":"db9a3b2dce1bb2059ad295f5ee7485f7589386d4a752a6a86ce9c70c185703a6","schema_version":"1.0","event_id":"sha256:db9a3b2dce1bb2059ad295f5ee7485f7589386d4a752a6a86ce9c70c185703a6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/C5WJPCKWAARPC5EXUDP7KSR6YM/bundle.json","state_url":"https://pith.science/pith/C5WJPCKWAARPC5EXUDP7KSR6YM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/C5WJPCKWAARPC5EXUDP7KSR6YM/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":"2026-07-10T21:10:50Z","links":{"resolver":"https://pith.science/pith/C5WJPCKWAARPC5EXUDP7KSR6YM","bundle":"https://pith.science/pith/C5WJPCKWAARPC5EXUDP7KSR6YM/bundle.json","state":"https://pith.science/pith/C5WJPCKWAARPC5EXUDP7KSR6YM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/C5WJPCKWAARPC5EXUDP7KSR6YM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:C5WJPCKWAARPC5EXUDP7KSR6YM","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":"87ac016f90203277f5a54b1fb78114a0ef9f767fb2d40c49d9da6bff22206efa","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-01T17:59:06Z","title_canon_sha256":"eb70e90787034bf309e745f454b724b0c156d4e48af8b610abb2aaf9837358c0"},"schema_version":"1.0","source":{"id":"2404.01295","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.01295","created_at":"2026-07-05T08:03:04Z"},{"alias_kind":"arxiv_version","alias_value":"2404.01295v1","created_at":"2026-07-05T08:03:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.01295","created_at":"2026-07-05T08:03:04Z"},{"alias_kind":"pith_short_12","alias_value":"C5WJPCKWAARP","created_at":"2026-07-05T08:03:04Z"},{"alias_kind":"pith_short_16","alias_value":"C5WJPCKWAARPC5EX","created_at":"2026-07-05T08:03:04Z"},{"alias_kind":"pith_short_8","alias_value":"C5WJPCKW","created_at":"2026-07-05T08:03:04Z"}],"graph_snapshots":[{"event_id":"sha256:db9a3b2dce1bb2059ad295f5ee7485f7589386d4a752a6a86ce9c70c185703a6","target":"graph","created_at":"2026-07-05T08:03:04Z","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/2404.01295/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models (LLMs) become easily accessible nowadays, the trade-off between safety and helpfulness can significantly impact user experience. A model that prioritizes safety will cause users to feel less engaged and assisted while prioritizing helpfulness will potentially cause harm. Possible harms include teaching people how to build a bomb, exposing youth to inappropriate content, and hurting users' mental health. In this work, we propose to balance safety and helpfulness in diverse use cases by controlling both attributes in LLM. We explore training-free and fine-tuning methods ","authors_text":"Asli Celikyilmaz, Daniel M. Bikel, Eric Michael Smith, Louis Martin, Soumya Batra, William Yang Wang, Xilun Chen, Yi-Lin Tuan","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-01T17:59:06Z","title":"Towards Safety and Helpfulness Balanced Responses via Controllable Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.01295","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":"sha256:7221e57a54433c0e9a56ec3accfab57951917517c60cb48745c7f8759a2c4f9f","target":"record","created_at":"2026-07-05T08:03:04Z","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":"87ac016f90203277f5a54b1fb78114a0ef9f767fb2d40c49d9da6bff22206efa","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-01T17:59:06Z","title_canon_sha256":"eb70e90787034bf309e745f454b724b0c156d4e48af8b610abb2aaf9837358c0"},"schema_version":"1.0","source":{"id":"2404.01295","kind":"arxiv","version":1}},"canonical_sha256":"176c9789560022f17497a0dff54a3ec33aed5f01761e094d4a4df549a960c17c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"176c9789560022f17497a0dff54a3ec33aed5f01761e094d4a4df549a960c17c","first_computed_at":"2026-07-05T08:03:04.155149Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:03:04.155149Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bSMQtvyh4Dvc+Hqdjt5E4Eh5rHpWquAcnRsRR4E8NmjkzzPG5QYxHZDZuCP+u6lj2cstZgsM0LqNvFgiIuCoAA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:03:04.155648Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.01295","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7221e57a54433c0e9a56ec3accfab57951917517c60cb48745c7f8759a2c4f9f","sha256:db9a3b2dce1bb2059ad295f5ee7485f7589386d4a752a6a86ce9c70c185703a6"],"state_sha256":"63638c786ba0459f0ec97aab2e03406cdad2c634df8d50b0e941ce7f28de5fa4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gZ/0nQuNOGgtMkN+VyjnCy1qYbuVIr6T2JxeFxMf2WdIOlALVDpFRQwYHtu+muOUUQiEIpqnnSOjmVjzo+cWCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-10T21:10:50.606705Z","bundle_sha256":"49d551b586bf40d68a2a00ac2bd97a4a7711bbe76c0455fd13525bf89e019265"}}