{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:MM7HFCVLDZPEVIPBB2UPMHBFT6","short_pith_number":"pith:MM7HFCVL","canonical_record":{"source":{"id":"2605.02405","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-04T09:49:44Z","cross_cats_sorted":[],"title_canon_sha256":"ea82347402b15b363bc05a0a8da92d11b9ffaf00997f6f10f3eb4185d9cf9088","abstract_canon_sha256":"61f68c8b92a808c60fcfe68e1aefcd140bef09ca51d4cfe7c5dcb3d1aaf5ca7f"},"schema_version":"1.0"},"canonical_sha256":"633e728aab1e5e4aa1e10ea8f61c259f84fcbbb0915bfa0c2ee996290f2e5db6","source":{"kind":"arxiv","id":"2605.02405","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.02405","created_at":"2026-06-26T01:15:53Z"},{"alias_kind":"arxiv_version","alias_value":"2605.02405v2","created_at":"2026-06-26T01:15:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.02405","created_at":"2026-06-26T01:15:53Z"},{"alias_kind":"pith_short_12","alias_value":"MM7HFCVLDZPE","created_at":"2026-06-26T01:15:53Z"},{"alias_kind":"pith_short_16","alias_value":"MM7HFCVLDZPEVIPB","created_at":"2026-06-26T01:15:53Z"},{"alias_kind":"pith_short_8","alias_value":"MM7HFCVL","created_at":"2026-06-26T01:15:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:MM7HFCVLDZPEVIPBB2UPMHBFT6","target":"record","payload":{"canonical_record":{"source":{"id":"2605.02405","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-04T09:49:44Z","cross_cats_sorted":[],"title_canon_sha256":"ea82347402b15b363bc05a0a8da92d11b9ffaf00997f6f10f3eb4185d9cf9088","abstract_canon_sha256":"61f68c8b92a808c60fcfe68e1aefcd140bef09ca51d4cfe7c5dcb3d1aaf5ca7f"},"schema_version":"1.0"},"canonical_sha256":"633e728aab1e5e4aa1e10ea8f61c259f84fcbbb0915bfa0c2ee996290f2e5db6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-26T01:15:53.016618Z","signature_b64":"fmHWlu1PQFfpRH1whegBvtLYIj58LKhyihOfYMUToXDs3jdkSO4zd0RLk/meK10QpTAaD0lMJjpGltMXlRliCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"633e728aab1e5e4aa1e10ea8f61c259f84fcbbb0915bfa0c2ee996290f2e5db6","last_reissued_at":"2026-06-26T01:15:53.016172Z","signature_status":"signed_v1","first_computed_at":"2026-06-26T01:15:53.016172Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.02405","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-26T01:15:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HK/jojVFaUW5gHNYJWMy5m3tAtjw0g4chmbkdWr7HKuND+sNo5WNuuWOBY5RyWqDTlqe4vhsiMTqzgzugB7MCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-08T19:06:56.956027Z"},"content_sha256":"5189ad57c67e073c30cd1f73bf2247459d0ede6fc3bc878804766188e22614e0","schema_version":"1.0","event_id":"sha256:5189ad57c67e073c30cd1f73bf2247459d0ede6fc3bc878804766188e22614e0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:MM7HFCVLDZPEVIPBB2UPMHBFT6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Closed-Loop CO2 Storage Control With History-Based Reinforcement Learning and Latent Model-Based Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"History-conditioned reinforcement learning recovers nearly all privileged-state performance for CO2 storage control with only well-level data.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Sofianos Panagiotis Fotias, Vassilis Gaganis","submitted_at":"2026-05-04T09:49:44Z","abstract_excerpt":"Closed-loop management of geological CO2 storage requires control policies that adapt to uncertain reservoir behavior while relying on observations that are realistically available during operation. This work formulates CO2 injection and brine-production control as a partially observable sequential decision problem and studies deployable deep reinforcement-learning controllers trained with high-fidelity reservoir simulation. We first compare privileged-state, well-only, history-conditioned, masking-curriculum, and asymmetric teacher-student model-free policies in order to quantify the value of"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"history-conditioned policies recover nearly all of the privileged-state performance while using only deployable well-level information, and that latent model-based retuning outperforms direct model-free retuning under the same scenario-specific real-simulator budget in the abnormal operating cases.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That high-fidelity reservoir simulations used for training and adaptation accurately represent real-world reservoir behavior and that the latent dynamics model captures the necessary changes under failures, leakage, and connectivity shifts.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"History-conditioned RL policies recover nearly all privileged-state performance with deployable well data, and latent model-based retuning outperforms direct model-free retuning under abnormal reservoir conditions.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"History-conditioned reinforcement learning recovers nearly all privileged-state performance for CO2 storage control with only well-level data.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"5f59b1d0fa05b68bed5b15d5deb9bc6e033e2c73a7edeb1bc2a4ad159968ae6e"},"source":{"id":"2605.02405","kind":"arxiv","version":2},"verdict":{"id":"5900c3ee-e1ca-4a57-8938-b34e507afc48","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-09T16:11:36.032983Z","strongest_claim":"history-conditioned policies recover nearly all of the privileged-state performance while using only deployable well-level information, and that latent model-based retuning outperforms direct model-free retuning under the same scenario-specific real-simulator budget in the abnormal operating cases.","one_line_summary":"History-conditioned RL policies recover nearly all privileged-state performance with deployable well data, and latent model-based retuning outperforms direct model-free retuning under abnormal reservoir conditions.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That high-fidelity reservoir simulations used for training and adaptation accurately represent real-world reservoir behavior and that the latent dynamics model captures the necessary changes under failures, leakage, and connectivity shifts.","pith_extraction_headline":"History-conditioned reinforcement learning recovers nearly all privileged-state performance for CO2 storage control with only well-level data."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.02405/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T15:41:06.406293Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-20T03:31:22.284792Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T16:23:22.308870Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"c792f11afa112b5c22560baa0780faf2b891e8ec36845be4431790acb5be901c"},"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":"5900c3ee-e1ca-4a57-8938-b34e507afc48"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-26T01:15:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"s1sxaSXkobRo0eKCGaEQ//Fx4g80Y8BpBs5ztR3LgQ1k3ydDiYTlADF0JRookV/JvQ6ITVBIvH18etRiVgSYAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-08T19:06:56.956536Z"},"content_sha256":"ebd982f1fb6110786b63531a76528d044b392d599392f2da66da82cc64a1143f","schema_version":"1.0","event_id":"sha256:ebd982f1fb6110786b63531a76528d044b392d599392f2da66da82cc64a1143f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MM7HFCVLDZPEVIPBB2UPMHBFT6/bundle.json","state_url":"https://pith.science/pith/MM7HFCVLDZPEVIPBB2UPMHBFT6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MM7HFCVLDZPEVIPBB2UPMHBFT6/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":"2026-07-08T19:06:56Z","links":{"resolver":"https://pith.science/pith/MM7HFCVLDZPEVIPBB2UPMHBFT6","bundle":"https://pith.science/pith/MM7HFCVLDZPEVIPBB2UPMHBFT6/bundle.json","state":"https://pith.science/pith/MM7HFCVLDZPEVIPBB2UPMHBFT6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MM7HFCVLDZPEVIPBB2UPMHBFT6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:MM7HFCVLDZPEVIPBB2UPMHBFT6","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":"61f68c8b92a808c60fcfe68e1aefcd140bef09ca51d4cfe7c5dcb3d1aaf5ca7f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-04T09:49:44Z","title_canon_sha256":"ea82347402b15b363bc05a0a8da92d11b9ffaf00997f6f10f3eb4185d9cf9088"},"schema_version":"1.0","source":{"id":"2605.02405","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.02405","created_at":"2026-06-26T01:15:53Z"},{"alias_kind":"arxiv_version","alias_value":"2605.02405v2","created_at":"2026-06-26T01:15:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.02405","created_at":"2026-06-26T01:15:53Z"},{"alias_kind":"pith_short_12","alias_value":"MM7HFCVLDZPE","created_at":"2026-06-26T01:15:53Z"},{"alias_kind":"pith_short_16","alias_value":"MM7HFCVLDZPEVIPB","created_at":"2026-06-26T01:15:53Z"},{"alias_kind":"pith_short_8","alias_value":"MM7HFCVL","created_at":"2026-06-26T01:15:53Z"}],"graph_snapshots":[{"event_id":"sha256:ebd982f1fb6110786b63531a76528d044b392d599392f2da66da82cc64a1143f","target":"graph","created_at":"2026-06-26T01:15:53Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"history-conditioned policies recover nearly all of the privileged-state performance while using only deployable well-level information, and that latent model-based retuning outperforms direct model-free retuning under the same scenario-specific real-simulator budget in the abnormal operating cases."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That high-fidelity reservoir simulations used for training and adaptation accurately represent real-world reservoir behavior and that the latent dynamics model captures the necessary changes under failures, leakage, and connectivity shifts."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"History-conditioned RL policies recover nearly all privileged-state performance with deployable well data, and latent model-based retuning outperforms direct model-free retuning under abnormal reservoir conditions."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"History-conditioned reinforcement learning recovers nearly all privileged-state performance for CO2 storage control with only well-level data."}],"snapshot_sha256":"5f59b1d0fa05b68bed5b15d5deb9bc6e033e2c73a7edeb1bc2a4ad159968ae6e"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T15:41:06.406293Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-20T03:31:22.284792Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T16:23:22.308870Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.02405/integrity.json","findings":[],"snapshot_sha256":"c792f11afa112b5c22560baa0780faf2b891e8ec36845be4431790acb5be901c","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Closed-loop management of geological CO2 storage requires control policies that adapt to uncertain reservoir behavior while relying on observations that are realistically available during operation. This work formulates CO2 injection and brine-production control as a partially observable sequential decision problem and studies deployable deep reinforcement-learning controllers trained with high-fidelity reservoir simulation. We first compare privileged-state, well-only, history-conditioned, masking-curriculum, and asymmetric teacher-student model-free policies in order to quantify the value of","authors_text":"Sofianos Panagiotis Fotias, Vassilis Gaganis","cross_cats":[],"headline":"History-conditioned reinforcement learning recovers nearly all privileged-state performance for CO2 storage control with only well-level data.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-04T09:49:44Z","title":"Closed-Loop CO2 Storage Control With History-Based Reinforcement Learning and Latent Model-Based Adaptation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.02405","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-09T16:11:36.032983Z","id":"5900c3ee-e1ca-4a57-8938-b34e507afc48","model_set":{"reader":"grok-4.3"},"one_line_summary":"History-conditioned RL policies recover nearly all privileged-state performance with deployable well data, and latent model-based retuning outperforms direct model-free retuning under abnormal reservoir conditions.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"History-conditioned reinforcement learning recovers nearly all privileged-state performance for CO2 storage control with only well-level data.","strongest_claim":"history-conditioned policies recover nearly all of the privileged-state performance while using only deployable well-level information, and that latent model-based retuning outperforms direct model-free retuning under the same scenario-specific real-simulator budget in the abnormal operating cases.","weakest_assumption":"That high-fidelity reservoir simulations used for training and adaptation accurately represent real-world reservoir behavior and that the latent dynamics model captures the necessary changes under failures, leakage, and connectivity shifts."}},"verdict_id":"5900c3ee-e1ca-4a57-8938-b34e507afc48"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:5189ad57c67e073c30cd1f73bf2247459d0ede6fc3bc878804766188e22614e0","target":"record","created_at":"2026-06-26T01:15:53Z","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":"61f68c8b92a808c60fcfe68e1aefcd140bef09ca51d4cfe7c5dcb3d1aaf5ca7f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-04T09:49:44Z","title_canon_sha256":"ea82347402b15b363bc05a0a8da92d11b9ffaf00997f6f10f3eb4185d9cf9088"},"schema_version":"1.0","source":{"id":"2605.02405","kind":"arxiv","version":2}},"canonical_sha256":"633e728aab1e5e4aa1e10ea8f61c259f84fcbbb0915bfa0c2ee996290f2e5db6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"633e728aab1e5e4aa1e10ea8f61c259f84fcbbb0915bfa0c2ee996290f2e5db6","first_computed_at":"2026-06-26T01:15:53.016172Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-26T01:15:53.016172Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fmHWlu1PQFfpRH1whegBvtLYIj58LKhyihOfYMUToXDs3jdkSO4zd0RLk/meK10QpTAaD0lMJjpGltMXlRliCQ==","signature_status":"signed_v1","signed_at":"2026-06-26T01:15:53.016618Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.02405","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5189ad57c67e073c30cd1f73bf2247459d0ede6fc3bc878804766188e22614e0","sha256:ebd982f1fb6110786b63531a76528d044b392d599392f2da66da82cc64a1143f"],"state_sha256":"c77b75ec4a99aefacd64048476aa4ceac693ee9ee65df9df53d6358d1acd6bdc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uI9t404/6jJPmjHbXEV7Zrhbk5p0Qaj6UpK5Nty/39KUyFWeKXCY7AjJfgVfUTxqez/HNVSVQX1wRzl0a1l6Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-08T19:06:56.959978Z","bundle_sha256":"7a8244a999566ce1748af76c41589080264ee053271fc081087431ee7eb7d29d"}}