{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:CXKT2WJBWUQKD73OXIAF2RAG4T","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":"e1dfde1e930e245983de2ebb3aca1e88f2fab98e1478d36e091a3bba3ac4192d","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-01T05:39:36Z","title_canon_sha256":"13f2a7abb05498d37aeaefcc9a938b0b0d36096e1fbb3add25cd9876143f7f08"},"schema_version":"1.0","source":{"id":"2404.00934","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.00934","created_at":"2026-07-05T08:03:51Z"},{"alias_kind":"arxiv_version","alias_value":"2404.00934v2","created_at":"2026-07-05T08:03:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.00934","created_at":"2026-07-05T08:03:51Z"},{"alias_kind":"pith_short_12","alias_value":"CXKT2WJBWUQK","created_at":"2026-07-05T08:03:51Z"},{"alias_kind":"pith_short_16","alias_value":"CXKT2WJBWUQKD73O","created_at":"2026-07-05T08:03:51Z"},{"alias_kind":"pith_short_8","alias_value":"CXKT2WJB","created_at":"2026-07-05T08:03:51Z"}],"graph_snapshots":[{"event_id":"sha256:d289f201d21258104dd1ca25802c06bfc52380adc36d093107fd6971bbb263cc","target":"graph","created_at":"2026-07-05T08:03:51Z","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/2404.00934/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"ChatGLM is a free-to-use AI service powered by the ChatGLM family of large language models (LLMs). In this paper, we present the ChatGLM-RLHF pipeline -- a reinforcement learning from human feedback (RLHF) system -- designed to enhance ChatGLM's alignment with human preferences. ChatGLM-RLHF encompasses three major components: the collection of human preference data, the training of the reward model, and the optimization of policies. Throughout the process of integrating ChatGLM-RLHF into production, we encountered and addressed several unprecedented challenges. We introduce the strategies to ","authors_text":"Aohan Zeng, Hongning Wang, Jie Tang, Minlie Huang, Qinkai Zheng, Xiaohan Zhang, Xiao Liu, Yilin Niu, Yuxiao Dong, Zhengxiao Du, Zhenyu Hou","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-01T05:39:36Z","title":"ChatGLM-RLHF: Practices of Aligning Large Language Models with Human Feedback"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.00934","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:a97100e9e2a9b4c86048511934031b89cc7553b96a5e8b4830e78a8f9ce2458a","target":"record","created_at":"2026-07-05T08:03:51Z","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":"e1dfde1e930e245983de2ebb3aca1e88f2fab98e1478d36e091a3bba3ac4192d","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-01T05:39:36Z","title_canon_sha256":"13f2a7abb05498d37aeaefcc9a938b0b0d36096e1fbb3add25cd9876143f7f08"},"schema_version":"1.0","source":{"id":"2404.00934","kind":"arxiv","version":2}},"canonical_sha256":"15d53d5921b520a1ff6eba005d4406e4d077f2cd85356432ed1dbe236f630404","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"15d53d5921b520a1ff6eba005d4406e4d077f2cd85356432ed1dbe236f630404","first_computed_at":"2026-07-05T08:03:51.482784Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:03:51.482784Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Pd5m+9poAbIpwOBeQEvaVZBwSiwmoMjDUVYYr8hilr/VCSfyISxmW/IOF88aVbcu6xcNr0DgLJfiXmZjkUqrAA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:03:51.483222Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.00934","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a97100e9e2a9b4c86048511934031b89cc7553b96a5e8b4830e78a8f9ce2458a","sha256:d289f201d21258104dd1ca25802c06bfc52380adc36d093107fd6971bbb263cc"],"state_sha256":"e5f231f5650c7e5b9c09ce09b17d9ba9efcc990d7d8f222fa4d3127d21adc5c4"}