{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:75FGNISRXNK2PAQ3G753BLETMA","short_pith_number":"pith:75FGNISR","canonical_record":{"source":{"id":"2409.10164","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-09-16T10:54:04Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"32bde66e32bb9f556c30385fb76ad7998c23660da78c2c38abbc74a0847667af","abstract_canon_sha256":"4c98347a076e44f4b1fe8c74c69e1e55cb049de99f6cb070fd195efb263bb76f"},"schema_version":"1.0"},"canonical_sha256":"ff4a66a251bb55a7821b37fbb0ac93602a1c1d24ca255faa1159db024224bcef","source":{"kind":"arxiv","id":"2409.10164","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.10164","created_at":"2026-07-05T09:07:36Z"},{"alias_kind":"arxiv_version","alias_value":"2409.10164v1","created_at":"2026-07-05T09:07:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.10164","created_at":"2026-07-05T09:07:36Z"},{"alias_kind":"pith_short_12","alias_value":"75FGNISRXNK2","created_at":"2026-07-05T09:07:36Z"},{"alias_kind":"pith_short_16","alias_value":"75FGNISRXNK2PAQ3","created_at":"2026-07-05T09:07:36Z"},{"alias_kind":"pith_short_8","alias_value":"75FGNISR","created_at":"2026-07-05T09:07:36Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:75FGNISRXNK2PAQ3G753BLETMA","target":"record","payload":{"canonical_record":{"source":{"id":"2409.10164","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-09-16T10:54:04Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"32bde66e32bb9f556c30385fb76ad7998c23660da78c2c38abbc74a0847667af","abstract_canon_sha256":"4c98347a076e44f4b1fe8c74c69e1e55cb049de99f6cb070fd195efb263bb76f"},"schema_version":"1.0"},"canonical_sha256":"ff4a66a251bb55a7821b37fbb0ac93602a1c1d24ca255faa1159db024224bcef","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:07:36.053639Z","signature_b64":"0UQCV52HuL/nWOLmYhRp/rrUN+iDcu+6T+amGq4F6zDBbx+hszbIzt6udUyeiJJFRR9Db8c7vWUi38iRhlMDDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff4a66a251bb55a7821b37fbb0ac93602a1c1d24ca255faa1159db024224bcef","last_reissued_at":"2026-07-05T09:07:36.053170Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:07:36.053170Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2409.10164","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":"2026-07-05T09:07:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RmI1EDoRZkNrLI/K0WAwQn9VpViZGPDWuNT4xzOv4dcPjVIitgkkJM+1dRLiIKSLHnHzUhvKsBshD8uxGOaqDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T09:13:37.999342Z"},"content_sha256":"9c35ac27956bfcd7f03a8f4f187791d2fe25e3364f8a3bd1652861357f0c6eb1","schema_version":"1.0","event_id":"sha256:9c35ac27956bfcd7f03a8f4f187791d2fe25e3364f8a3bd1652861357f0c6eb1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:75FGNISRXNK2PAQ3G753BLETMA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Quantile Regression for Distributional Reward Models in RLHF","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Nicolai Dorka","submitted_at":"2024-09-16T10:54:04Z","abstract_excerpt":"Reinforcement learning from human feedback (RLHF) has become a key method for aligning large language models (LLMs) with human preferences through the use of reward models. However, traditional reward models typically generate point estimates, which oversimplify the diversity and complexity of human values and preferences. In this paper, we introduce Quantile Reward Models (QRMs), a novel approach to reward modeling that learns a distribution over rewards instead of a single scalar value. Our method uses quantile regression to estimate a full, potentially multimodal distribution over preferenc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.10164","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/2409.10164/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":"2026-07-05T09:07:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iVZ8cgV4ZbpkIyzXXr63CbiZYPfSPmdi7cWcqHg9HUFMb1/S13aDuRxvAx4zJWgSvn1Pf3AwcTMKgJuatp2UDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T09:13:37.999727Z"},"content_sha256":"fc40d1bae93926e2c908180a0662a75405376ee0c2da04a2514ae90982a2c6af","schema_version":"1.0","event_id":"sha256:fc40d1bae93926e2c908180a0662a75405376ee0c2da04a2514ae90982a2c6af"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/75FGNISRXNK2PAQ3G753BLETMA/bundle.json","state_url":"https://pith.science/pith/75FGNISRXNK2PAQ3G753BLETMA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/75FGNISRXNK2PAQ3G753BLETMA/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":"2026-08-04T09:13:38Z","links":{"resolver":"https://pith.science/pith/75FGNISRXNK2PAQ3G753BLETMA","bundle":"https://pith.science/pith/75FGNISRXNK2PAQ3G753BLETMA/bundle.json","state":"https://pith.science/pith/75FGNISRXNK2PAQ3G753BLETMA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/75FGNISRXNK2PAQ3G753BLETMA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:75FGNISRXNK2PAQ3G753BLETMA","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":"4c98347a076e44f4b1fe8c74c69e1e55cb049de99f6cb070fd195efb263bb76f","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-09-16T10:54:04Z","title_canon_sha256":"32bde66e32bb9f556c30385fb76ad7998c23660da78c2c38abbc74a0847667af"},"schema_version":"1.0","source":{"id":"2409.10164","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.10164","created_at":"2026-07-05T09:07:36Z"},{"alias_kind":"arxiv_version","alias_value":"2409.10164v1","created_at":"2026-07-05T09:07:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.10164","created_at":"2026-07-05T09:07:36Z"},{"alias_kind":"pith_short_12","alias_value":"75FGNISRXNK2","created_at":"2026-07-05T09:07:36Z"},{"alias_kind":"pith_short_16","alias_value":"75FGNISRXNK2PAQ3","created_at":"2026-07-05T09:07:36Z"},{"alias_kind":"pith_short_8","alias_value":"75FGNISR","created_at":"2026-07-05T09:07:36Z"}],"graph_snapshots":[{"event_id":"sha256:fc40d1bae93926e2c908180a0662a75405376ee0c2da04a2514ae90982a2c6af","target":"graph","created_at":"2026-07-05T09:07:36Z","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/2409.10164/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning from human feedback (RLHF) has become a key method for aligning large language models (LLMs) with human preferences through the use of reward models. However, traditional reward models typically generate point estimates, which oversimplify the diversity and complexity of human values and preferences. In this paper, we introduce Quantile Reward Models (QRMs), a novel approach to reward modeling that learns a distribution over rewards instead of a single scalar value. Our method uses quantile regression to estimate a full, potentially multimodal distribution over preferenc","authors_text":"Nicolai Dorka","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-09-16T10:54:04Z","title":"Quantile Regression for Distributional Reward Models in RLHF"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.10164","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":"sha256:9c35ac27956bfcd7f03a8f4f187791d2fe25e3364f8a3bd1652861357f0c6eb1","target":"record","created_at":"2026-07-05T09:07:36Z","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":"4c98347a076e44f4b1fe8c74c69e1e55cb049de99f6cb070fd195efb263bb76f","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-09-16T10:54:04Z","title_canon_sha256":"32bde66e32bb9f556c30385fb76ad7998c23660da78c2c38abbc74a0847667af"},"schema_version":"1.0","source":{"id":"2409.10164","kind":"arxiv","version":1}},"canonical_sha256":"ff4a66a251bb55a7821b37fbb0ac93602a1c1d24ca255faa1159db024224bcef","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ff4a66a251bb55a7821b37fbb0ac93602a1c1d24ca255faa1159db024224bcef","first_computed_at":"2026-07-05T09:07:36.053170Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:07:36.053170Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0UQCV52HuL/nWOLmYhRp/rrUN+iDcu+6T+amGq4F6zDBbx+hszbIzt6udUyeiJJFRR9Db8c7vWUi38iRhlMDDw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:07:36.053639Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.10164","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9c35ac27956bfcd7f03a8f4f187791d2fe25e3364f8a3bd1652861357f0c6eb1","sha256:fc40d1bae93926e2c908180a0662a75405376ee0c2da04a2514ae90982a2c6af"],"state_sha256":"4bf2536444ea537f07801194a90448399e02395ba3127f0b906aa53e320f81c8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u3ACNoDlAVCDp0QDkG7J0YYuHOK128oQqg1PYTSbi0eYfzhPK3Y4RLstGnwK9/puD6S/1rUN4KlQPZBBBho7AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T09:13:38.002373Z","bundle_sha256":"d5fbf29df8e1adb28c0a5063d5d8db3847fa9021603b79a9640f036a12fe90b8"}}