{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:PJHHKZKXTWPQR6FBHYNLQ5JNEK","short_pith_number":"pith:PJHHKZKX","canonical_record":{"source":{"id":"2508.14027","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-19T17:37:35Z","cross_cats_sorted":[],"title_canon_sha256":"0e67df2fd31d35a40a44eab6d3452535d188c2d08c51b8bc73a055c24391c2d4","abstract_canon_sha256":"c06e7f19ced4a2f873fb4c6f54c4e1f48fdecf67dac0bb332bf1b16cfc32dd64"},"schema_version":"1.0"},"canonical_sha256":"7a4e7565579d9f08f8a13e1ab8752d229343f1261f6f38877f2309226e21cf0a","source":{"kind":"arxiv","id":"2508.14027","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.14027","created_at":"2026-07-05T11:56:11Z"},{"alias_kind":"arxiv_version","alias_value":"2508.14027v1","created_at":"2026-07-05T11:56:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.14027","created_at":"2026-07-05T11:56:11Z"},{"alias_kind":"pith_short_12","alias_value":"PJHHKZKXTWPQ","created_at":"2026-07-05T11:56:11Z"},{"alias_kind":"pith_short_16","alias_value":"PJHHKZKXTWPQR6FB","created_at":"2026-07-05T11:56:11Z"},{"alias_kind":"pith_short_8","alias_value":"PJHHKZKX","created_at":"2026-07-05T11:56:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:PJHHKZKXTWPQR6FBHYNLQ5JNEK","target":"record","payload":{"canonical_record":{"source":{"id":"2508.14027","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-19T17:37:35Z","cross_cats_sorted":[],"title_canon_sha256":"0e67df2fd31d35a40a44eab6d3452535d188c2d08c51b8bc73a055c24391c2d4","abstract_canon_sha256":"c06e7f19ced4a2f873fb4c6f54c4e1f48fdecf67dac0bb332bf1b16cfc32dd64"},"schema_version":"1.0"},"canonical_sha256":"7a4e7565579d9f08f8a13e1ab8752d229343f1261f6f38877f2309226e21cf0a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:56:11.127263Z","signature_b64":"NheCDSkLv/hX/op4SnXDIeS1riRis8/3819Go8+IM48mFZ3YPNdekaMNBBlnr2ttNjg4i9tGHoBFyCbKD3IzAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a4e7565579d9f08f8a13e1ab8752d229343f1261f6f38877f2309226e21cf0a","last_reissued_at":"2026-07-05T11:56:11.126853Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:56:11.126853Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.14027","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":"2026-07-05T11:56:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9YU1PnTSVSGRRnIoJdJXcfCXnpKYbN76JKy2VMXAbXjfV0aB47hPSu04pcZSdY0COI2xwV/V2VbwELgUHI0RCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T01:50:13.052944Z"},"content_sha256":"b3d4abc30a3e04aee7626d9f6fb47575febbd43b3662ba8dea836716c259f612","schema_version":"1.0","event_id":"sha256:b3d4abc30a3e04aee7626d9f6fb47575febbd43b3662ba8dea836716c259f612"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:PJHHKZKXTWPQR6FBHYNLQ5JNEK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learning from Preferences and Mixed Demonstrations in General Settings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Carl Henrik Ek, Jason R Brown, Robert D Mullins","submitted_at":"2025-08-19T17:37:35Z","abstract_excerpt":"Reinforcement learning is a general method for learning in sequential settings, but it can often be difficult to specify a good reward function when the task is complex. In these cases, preference feedback or expert demonstrations can be used instead. However, existing approaches utilising both together are often ad-hoc, rely on domain-specific properties, or won't scale. We develop a new framing for learning from human data, \\emph{reward-rational partial orderings over observations}, designed to be flexible and scalable. Based on this we introduce a practical algorithm, LEOPARD: Learning Esti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.14027","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/2508.14027/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":"2026-07-05T11:56:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DOeKqQmPK5XUHsGL2/K8KZA1gBBS5Em+pxoDTZ/NPGl4u6o8ZiV9/pZHmewp3S19t6DZqSTIhcgn9R9EQF6ACw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T01:50:13.053463Z"},"content_sha256":"46d00b9823b96b5770807cee3a907b308bcda05a154eae18aad3c743c149e01a","schema_version":"1.0","event_id":"sha256:46d00b9823b96b5770807cee3a907b308bcda05a154eae18aad3c743c149e01a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PJHHKZKXTWPQR6FBHYNLQ5JNEK/bundle.json","state_url":"https://pith.science/pith/PJHHKZKXTWPQR6FBHYNLQ5JNEK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PJHHKZKXTWPQR6FBHYNLQ5JNEK/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":"2026-08-20T01:50:13Z","links":{"resolver":"https://pith.science/pith/PJHHKZKXTWPQR6FBHYNLQ5JNEK","bundle":"https://pith.science/pith/PJHHKZKXTWPQR6FBHYNLQ5JNEK/bundle.json","state":"https://pith.science/pith/PJHHKZKXTWPQR6FBHYNLQ5JNEK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PJHHKZKXTWPQR6FBHYNLQ5JNEK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PJHHKZKXTWPQR6FBHYNLQ5JNEK","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":"c06e7f19ced4a2f873fb4c6f54c4e1f48fdecf67dac0bb332bf1b16cfc32dd64","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-19T17:37:35Z","title_canon_sha256":"0e67df2fd31d35a40a44eab6d3452535d188c2d08c51b8bc73a055c24391c2d4"},"schema_version":"1.0","source":{"id":"2508.14027","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.14027","created_at":"2026-07-05T11:56:11Z"},{"alias_kind":"arxiv_version","alias_value":"2508.14027v1","created_at":"2026-07-05T11:56:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.14027","created_at":"2026-07-05T11:56:11Z"},{"alias_kind":"pith_short_12","alias_value":"PJHHKZKXTWPQ","created_at":"2026-07-05T11:56:11Z"},{"alias_kind":"pith_short_16","alias_value":"PJHHKZKXTWPQR6FB","created_at":"2026-07-05T11:56:11Z"},{"alias_kind":"pith_short_8","alias_value":"PJHHKZKX","created_at":"2026-07-05T11:56:11Z"}],"graph_snapshots":[{"event_id":"sha256:46d00b9823b96b5770807cee3a907b308bcda05a154eae18aad3c743c149e01a","target":"graph","created_at":"2026-07-05T11:56:11Z","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/2508.14027/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning is a general method for learning in sequential settings, but it can often be difficult to specify a good reward function when the task is complex. In these cases, preference feedback or expert demonstrations can be used instead. However, existing approaches utilising both together are often ad-hoc, rely on domain-specific properties, or won't scale. We develop a new framing for learning from human data, \\emph{reward-rational partial orderings over observations}, designed to be flexible and scalable. Based on this we introduce a practical algorithm, LEOPARD: Learning Esti","authors_text":"Carl Henrik Ek, Jason R Brown, Robert D Mullins","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-19T17:37:35Z","title":"Learning from Preferences and Mixed Demonstrations in General Settings"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.14027","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:b3d4abc30a3e04aee7626d9f6fb47575febbd43b3662ba8dea836716c259f612","target":"record","created_at":"2026-07-05T11:56:11Z","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":"c06e7f19ced4a2f873fb4c6f54c4e1f48fdecf67dac0bb332bf1b16cfc32dd64","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-19T17:37:35Z","title_canon_sha256":"0e67df2fd31d35a40a44eab6d3452535d188c2d08c51b8bc73a055c24391c2d4"},"schema_version":"1.0","source":{"id":"2508.14027","kind":"arxiv","version":1}},"canonical_sha256":"7a4e7565579d9f08f8a13e1ab8752d229343f1261f6f38877f2309226e21cf0a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7a4e7565579d9f08f8a13e1ab8752d229343f1261f6f38877f2309226e21cf0a","first_computed_at":"2026-07-05T11:56:11.126853Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:56:11.126853Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NheCDSkLv/hX/op4SnXDIeS1riRis8/3819Go8+IM48mFZ3YPNdekaMNBBlnr2ttNjg4i9tGHoBFyCbKD3IzAw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:56:11.127263Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.14027","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b3d4abc30a3e04aee7626d9f6fb47575febbd43b3662ba8dea836716c259f612","sha256:46d00b9823b96b5770807cee3a907b308bcda05a154eae18aad3c743c149e01a"],"state_sha256":"99c5a8dd1d072d29dc708b83d887b7aa46bdfee069c188a53714896515fb27dd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UdbOy4HfPeWUjzRz3G+dEQJ3wuZkyJx8L55a7961e1gTX3yCvw269+xo+TUKh7BS5IDfeqxDgs5zFk0CFDRTDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T01:50:13.057579Z","bundle_sha256":"8e4e3c80d2e9fbdfe9601e5f7d6d07aae7e95d8b32779fe90a687794d2045ece"}}