{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:UJD6C5LKM6FET57BR5S427227D","short_pith_number":"pith:UJD6C5LK","canonical_record":{"source":{"id":"2605.05863","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-07T08:32:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"385fe671c4fa7b2102bd48a27a22b37a07480575b6974d319ac81587ec48f3c2","abstract_canon_sha256":"c9929fbe9928ffdf20365251fc61cf2891f678711bc7705ee68bedcdf8189391"},"schema_version":"1.0"},"canonical_sha256":"a247e1756a678a49f7e18f65cd7f5af8e8ff466fe8f75a884968938cd2b826da","source":{"kind":"arxiv","id":"2605.05863","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.05863","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"arxiv_version","alias_value":"2605.05863v2","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.05863","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"pith_short_12","alias_value":"UJD6C5LKM6FE","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"pith_short_16","alias_value":"UJD6C5LKM6FET57B","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"pith_short_8","alias_value":"UJD6C5LK","created_at":"2026-05-21T01:05:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:UJD6C5LKM6FET57BR5S427227D","target":"record","payload":{"canonical_record":{"source":{"id":"2605.05863","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-07T08:32:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"385fe671c4fa7b2102bd48a27a22b37a07480575b6974d319ac81587ec48f3c2","abstract_canon_sha256":"c9929fbe9928ffdf20365251fc61cf2891f678711bc7705ee68bedcdf8189391"},"schema_version":"1.0"},"canonical_sha256":"a247e1756a678a49f7e18f65cd7f5af8e8ff466fe8f75a884968938cd2b826da","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:05:20.300587Z","signature_b64":"JH0r353m5FZtVyMpsR4mhtDk87FLbw0AaipGs9t+0EozVJk8zFfrtetWjHzXZPIlpNDfP2VwX3wGGwnNi/7tCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a247e1756a678a49f7e18f65cd7f5af8e8ff466fe8f75a884968938cd2b826da","last_reissued_at":"2026-05-21T01:05:20.299992Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:05:20.299992Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.05863","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":"2026-05-21T01:05:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2AU2a7M6YP9SAS78bFfmMycBqjq9DRVo5pvjY92ZN6eKL/uwwpuNPT4r9a+R6SikyHlLh54LGeMGAZGPkwVrDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T04:50:43.691086Z"},"content_sha256":"3fbca816c16eeb921b6e735750193a43fd36c8c7f5b26a00d99a4239148c9ba9","schema_version":"1.0","event_id":"sha256:3fbca816c16eeb921b6e735750193a43fd36c8c7f5b26a00d99a4239148c9ba9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:UJD6C5LKM6FET57BR5S427227D","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SOPE: Stabilizing Off-Policy Evaluation for Online RL with Prior Data","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"SOPE uses an actor-aligned OPE signal on held-out data to automatically stop offline phases in online RL with prior data.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Alessandro Sestini, Andrew D. Bagdanov, Carlo Romeo, Girolamo Macaluso","submitted_at":"2026-05-07T08:32:09Z","abstract_excerpt":"Incorporating prior data into online reinforcement learning accelerates training but typically forces a difficult trade-off between high computational costs and long, multi-stage training pipelines. While fixed-length stabilization phases are significantly more computationally efficient than static update schedules, they require task-dependent manual tuning, risking either the waste of prior knowledge or severe overfitting. To address this, we propose SOPE, an algorithm that uses an actor-aligned Off-Policy Policy Evaluation (OPE) signal as an automated early-stopping mechanism to dynamically "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Evaluated on 25 continuous control tasks from the Minari benchmark suite, SOPE improves baseline performance by up to 45.6% while reducing the required TFLOPs by up to 22x, thus balancing the tradeoff between sample and computational efficiency.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the actor-aligned OPE signal evaluated on a held-out validation split under the current policy's action distribution accurately detects the saturation point of out-of-distribution benefits without either stopping too early (wasting prior knowledge) or too late (causing overfitting).","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SOPE uses an actor-aligned OPE signal on a held-out validation split to dynamically stop offline stabilization phases in online RL, improving performance up to 45.6% and cutting TFLOPs up to 22x on 25 Minari tasks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"SOPE uses an actor-aligned OPE signal on held-out data to automatically stop offline phases in online RL with prior data.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"9d9f3a58267f272a031f4a8ec3df5db7e734dda18b4f217eba9b3e2c990741e3"},"source":{"id":"2605.05863","kind":"arxiv","version":2},"verdict":{"id":"cbbfe96c-6794-43dd-8ec4-ae751b735dc1","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T14:50:37.993640Z","strongest_claim":"Evaluated on 25 continuous control tasks from the Minari benchmark suite, SOPE improves baseline performance by up to 45.6% while reducing the required TFLOPs by up to 22x, thus balancing the tradeoff between sample and computational efficiency.","one_line_summary":"SOPE uses an actor-aligned OPE signal on a held-out validation split to dynamically stop offline stabilization phases in online RL, improving performance up to 45.6% and cutting TFLOPs up to 22x on 25 Minari tasks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the actor-aligned OPE signal evaluated on a held-out validation split under the current policy's action distribution accurately detects the saturation point of out-of-distribution benefits without either stopping too early (wasting prior knowledge) or too late (causing overfitting).","pith_extraction_headline":"SOPE uses an actor-aligned OPE signal on held-out data to automatically stop offline phases in online RL with prior data."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.05863/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T13:42:04.568842Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-20T08:41:37.757500Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T19:31:19.547150Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T13:10:19.048833Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"b18319b5272c7d5c096da65596b096e7694a12c061693846f15a8f159d8db496"},"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":"cbbfe96c-6794-43dd-8ec4-ae751b735dc1"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-21T01:05:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sB3CbgRb1YXuGjkQV2xkbwd7Obcb9bi1Rx1UFHX9FzKSPbB7rN39tmsaBtko208OL2SmyjNUm/0CtX9zNoCAAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T04:50:43.691626Z"},"content_sha256":"606055a58db9727e34519d39433251b4aa69b241d50bd6ddd494e02cd38315fb","schema_version":"1.0","event_id":"sha256:606055a58db9727e34519d39433251b4aa69b241d50bd6ddd494e02cd38315fb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UJD6C5LKM6FET57BR5S427227D/bundle.json","state_url":"https://pith.science/pith/UJD6C5LKM6FET57BR5S427227D/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UJD6C5LKM6FET57BR5S427227D/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":"2026-05-28T04:50:43Z","links":{"resolver":"https://pith.science/pith/UJD6C5LKM6FET57BR5S427227D","bundle":"https://pith.science/pith/UJD6C5LKM6FET57BR5S427227D/bundle.json","state":"https://pith.science/pith/UJD6C5LKM6FET57BR5S427227D/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UJD6C5LKM6FET57BR5S427227D/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:UJD6C5LKM6FET57BR5S427227D","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":"c9929fbe9928ffdf20365251fc61cf2891f678711bc7705ee68bedcdf8189391","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-07T08:32:09Z","title_canon_sha256":"385fe671c4fa7b2102bd48a27a22b37a07480575b6974d319ac81587ec48f3c2"},"schema_version":"1.0","source":{"id":"2605.05863","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.05863","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"arxiv_version","alias_value":"2605.05863v2","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.05863","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"pith_short_12","alias_value":"UJD6C5LKM6FE","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"pith_short_16","alias_value":"UJD6C5LKM6FET57B","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"pith_short_8","alias_value":"UJD6C5LK","created_at":"2026-05-21T01:05:20Z"}],"graph_snapshots":[{"event_id":"sha256:606055a58db9727e34519d39433251b4aa69b241d50bd6ddd494e02cd38315fb","target":"graph","created_at":"2026-05-21T01:05:20Z","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":"Evaluated on 25 continuous control tasks from the Minari benchmark suite, SOPE improves baseline performance by up to 45.6% while reducing the required TFLOPs by up to 22x, thus balancing the tradeoff between sample and computational efficiency."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the actor-aligned OPE signal evaluated on a held-out validation split under the current policy's action distribution accurately detects the saturation point of out-of-distribution benefits without either stopping too early (wasting prior knowledge) or too late (causing overfitting)."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"SOPE uses an actor-aligned OPE signal on a held-out validation split to dynamically stop offline stabilization phases in online RL, improving performance up to 45.6% and cutting TFLOPs up to 22x on 25 Minari tasks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"SOPE uses an actor-aligned OPE signal on held-out data to automatically stop offline phases in online RL with prior data."}],"snapshot_sha256":"9d9f3a58267f272a031f4a8ec3df5db7e734dda18b4f217eba9b3e2c990741e3"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T13:42:04.568842Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T08:41:37.757500Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T19:31:19.547150Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T13:10:19.048833Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.05863/integrity.json","findings":[],"snapshot_sha256":"b18319b5272c7d5c096da65596b096e7694a12c061693846f15a8f159d8db496","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Incorporating prior data into online reinforcement learning accelerates training but typically forces a difficult trade-off between high computational costs and long, multi-stage training pipelines. While fixed-length stabilization phases are significantly more computationally efficient than static update schedules, they require task-dependent manual tuning, risking either the waste of prior knowledge or severe overfitting. To address this, we propose SOPE, an algorithm that uses an actor-aligned Off-Policy Policy Evaluation (OPE) signal as an automated early-stopping mechanism to dynamically ","authors_text":"Alessandro Sestini, Andrew D. Bagdanov, Carlo Romeo, Girolamo Macaluso","cross_cats":["cs.AI"],"headline":"SOPE uses an actor-aligned OPE signal on held-out data to automatically stop offline phases in online RL with prior data.","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-07T08:32:09Z","title":"SOPE: Stabilizing Off-Policy Evaluation for Online RL with Prior Data"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.05863","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T14:50:37.993640Z","id":"cbbfe96c-6794-43dd-8ec4-ae751b735dc1","model_set":{"reader":"grok-4.3"},"one_line_summary":"SOPE uses an actor-aligned OPE signal on a held-out validation split to dynamically stop offline stabilization phases in online RL, improving performance up to 45.6% and cutting TFLOPs up to 22x on 25 Minari tasks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"SOPE uses an actor-aligned OPE signal on held-out data to automatically stop offline phases in online RL with prior data.","strongest_claim":"Evaluated on 25 continuous control tasks from the Minari benchmark suite, SOPE improves baseline performance by up to 45.6% while reducing the required TFLOPs by up to 22x, thus balancing the tradeoff between sample and computational efficiency.","weakest_assumption":"That the actor-aligned OPE signal evaluated on a held-out validation split under the current policy's action distribution accurately detects the saturation point of out-of-distribution benefits without either stopping too early (wasting prior knowledge) or too late (causing overfitting)."}},"verdict_id":"cbbfe96c-6794-43dd-8ec4-ae751b735dc1"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3fbca816c16eeb921b6e735750193a43fd36c8c7f5b26a00d99a4239148c9ba9","target":"record","created_at":"2026-05-21T01:05:20Z","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":"c9929fbe9928ffdf20365251fc61cf2891f678711bc7705ee68bedcdf8189391","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-07T08:32:09Z","title_canon_sha256":"385fe671c4fa7b2102bd48a27a22b37a07480575b6974d319ac81587ec48f3c2"},"schema_version":"1.0","source":{"id":"2605.05863","kind":"arxiv","version":2}},"canonical_sha256":"a247e1756a678a49f7e18f65cd7f5af8e8ff466fe8f75a884968938cd2b826da","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a247e1756a678a49f7e18f65cd7f5af8e8ff466fe8f75a884968938cd2b826da","first_computed_at":"2026-05-21T01:05:20.299992Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-21T01:05:20.299992Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"JH0r353m5FZtVyMpsR4mhtDk87FLbw0AaipGs9t+0EozVJk8zFfrtetWjHzXZPIlpNDfP2VwX3wGGwnNi/7tCw==","signature_status":"signed_v1","signed_at":"2026-05-21T01:05:20.300587Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.05863","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3fbca816c16eeb921b6e735750193a43fd36c8c7f5b26a00d99a4239148c9ba9","sha256:606055a58db9727e34519d39433251b4aa69b241d50bd6ddd494e02cd38315fb"],"state_sha256":"6f591471097352ba9cb20bd2a7f80761ca493851b3a147d48915d689756acab9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5I8Fx6KhVh6m0xvBpVTtgYJYMweaqR38YXDnZHZDk5X4Zdm3tHQPKO6PAXqYtA+HrsD3Exl/QrK2GgmZooGABw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-28T04:50:43.694357Z","bundle_sha256":"7ec7d43e38f213c762f06dbfac9a9e3a31d156d7f76a8cfd85eb242ac9a23b82"}}