{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:V5CNDYJ2D2FWRKD2RANGEWXLYO","short_pith_number":"pith:V5CNDYJ2","schema_version":"1.0","canonical_sha256":"af44d1e13a1e8b68a87a881a625aebc38261a1159db5adc8c26966381adc7698","source":{"kind":"arxiv","id":"2306.04707","version":1},"attestation_state":"computed","paper":{"title":"Improving Open Language Models by Learning from Organic Interactions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Da Ju, Eric Michael Smith, Jason Weston, Jing Xu, Joshua Lane, Kurt Shuster, Megan Ung, Mojtaba Komeili, Morteza Behrooz, Rashel Moritz, Sainbayar Sukhbaatar, William Ngan, Y-Lan Boureau","submitted_at":"2023-06-07T18:19:46Z","abstract_excerpt":"We present BlenderBot 3x, an update on the conversational model BlenderBot 3, which is now trained using organic conversation and feedback data from participating users of the system in order to improve both its skills and safety. We are publicly releasing the participating de-identified interaction data for use by the research community, in order to spur further progress. Training models with organic data is challenging because interactions with people \"in the wild\" include both high quality conversations and feedback, as well as adversarial and toxic behavior. We study techniques that enable"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2306.04707","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-07T18:19:46Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2df986a38426784e590550fc0c1ed5a7fde93e6a633033305262c17a4281e058","abstract_canon_sha256":"c887a05964407b8ee23f04082cfc264f495e49591423480ba8327a923ebb1578"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:18:40.835529Z","signature_b64":"Z4+cAV2Lz6jcJ7E/x96c3DexDpdugeiDSndc4wKSbbAlb+p5GUZKhxfLzxZAboFe8SaJZjNZltI7A+YxT8YXCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"af44d1e13a1e8b68a87a881a625aebc38261a1159db5adc8c26966381adc7698","last_reissued_at":"2026-07-05T06:18:40.834967Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:18:40.834967Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Open Language Models by Learning from Organic Interactions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Da Ju, Eric Michael Smith, Jason Weston, Jing Xu, Joshua Lane, Kurt Shuster, Megan Ung, Mojtaba Komeili, Morteza Behrooz, Rashel Moritz, Sainbayar Sukhbaatar, William Ngan, Y-Lan Boureau","submitted_at":"2023-06-07T18:19:46Z","abstract_excerpt":"We present BlenderBot 3x, an update on the conversational model BlenderBot 3, which is now trained using organic conversation and feedback data from participating users of the system in order to improve both its skills and safety. We are publicly releasing the participating de-identified interaction data for use by the research community, in order to spur further progress. Training models with organic data is challenging because interactions with people \"in the wild\" include both high quality conversations and feedback, as well as adversarial and toxic behavior. We study techniques that enable"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.04707","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/2306.04707/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2306.04707","created_at":"2026-07-05T06:18:40.835024+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.04707v1","created_at":"2026-07-05T06:18:40.835024+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.04707","created_at":"2026-07-05T06:18:40.835024+00:00"},{"alias_kind":"pith_short_12","alias_value":"V5CNDYJ2D2FW","created_at":"2026-07-05T06:18:40.835024+00:00"},{"alias_kind":"pith_short_16","alias_value":"V5CNDYJ2D2FWRKD2","created_at":"2026-07-05T06:18:40.835024+00:00"},{"alias_kind":"pith_short_8","alias_value":"V5CNDYJ2","created_at":"2026-07-05T06:18:40.835024+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2309.11495","citing_title":"Chain-of-Verification Reduces Hallucination in Large Language Models","ref_index":188,"is_internal_anchor":false},{"citing_arxiv_id":"2405.01470","citing_title":"WildChat: 1M ChatGPT Interaction Logs in the Wild","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/V5CNDYJ2D2FWRKD2RANGEWXLYO","json":"https://pith.science/pith/V5CNDYJ2D2FWRKD2RANGEWXLYO.json","graph_json":"https://pith.science/api/pith-number/V5CNDYJ2D2FWRKD2RANGEWXLYO/graph.json","events_json":"https://pith.science/api/pith-number/V5CNDYJ2D2FWRKD2RANGEWXLYO/events.json","paper":"https://pith.science/paper/V5CNDYJ2"},"agent_actions":{"view_html":"https://pith.science/pith/V5CNDYJ2D2FWRKD2RANGEWXLYO","download_json":"https://pith.science/pith/V5CNDYJ2D2FWRKD2RANGEWXLYO.json","view_paper":"https://pith.science/paper/V5CNDYJ2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.04707&json=true","fetch_graph":"https://pith.science/api/pith-number/V5CNDYJ2D2FWRKD2RANGEWXLYO/graph.json","fetch_events":"https://pith.science/api/pith-number/V5CNDYJ2D2FWRKD2RANGEWXLYO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V5CNDYJ2D2FWRKD2RANGEWXLYO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V5CNDYJ2D2FWRKD2RANGEWXLYO/action/storage_attestation","attest_author":"https://pith.science/pith/V5CNDYJ2D2FWRKD2RANGEWXLYO/action/author_attestation","sign_citation":"https://pith.science/pith/V5CNDYJ2D2FWRKD2RANGEWXLYO/action/citation_signature","submit_replication":"https://pith.science/pith/V5CNDYJ2D2FWRKD2RANGEWXLYO/action/replication_record"}},"created_at":"2026-07-05T06:18:40.835024+00:00","updated_at":"2026-07-05T06:18:40.835024+00:00"}