{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:N3K3BDIEGEX2C7MFBOZL3G2WHN","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":"bb3c172d3b226d9c19435e42539de97835efdc34c3ee87ef09baa4840e3b5d4e","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-04-14T12:16:56Z","title_canon_sha256":"142d5ac54bb02a1c6b6be8ec8c83df63fd0e7cd2f523964f178729e67a20a1af"},"schema_version":"1.0","source":{"id":"2604.12645","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.12645","created_at":"2026-06-04T01:08:50Z"},{"alias_kind":"arxiv_version","alias_value":"2604.12645v2","created_at":"2026-06-04T01:08:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.12645","created_at":"2026-06-04T01:08:50Z"},{"alias_kind":"pith_short_12","alias_value":"N3K3BDIEGEX2","created_at":"2026-06-04T01:08:50Z"},{"alias_kind":"pith_short_16","alias_value":"N3K3BDIEGEX2C7MF","created_at":"2026-06-04T01:08:50Z"},{"alias_kind":"pith_short_8","alias_value":"N3K3BDIE","created_at":"2026-06-04T01:08:50Z"}],"graph_snapshots":[{"event_id":"sha256:a78aa21cefcb05192b5da6f7c429946cfa75d5840e6250feaddeb834a6c201cc","target":"graph","created_at":"2026-06-04T01:08:50Z","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":"We train a single context-dependent policy that is able to solve multiple related monitoring tasks in a simulated reef environment in HoloOcean."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That performance gains observed in the HoloOcean simulator will translate to real-world underwater environments with highly uncertain and non-stationary dynamics."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A context-dependent multi-task RL policy is trained and evaluated in HoloOcean simulation to solve multiple reef monitoring tasks with claimed improvements in sample efficiency, zero-shot generalization, and robustness to water currents."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Contextual multi-task reinforcement learning trains one policy to handle multiple reef monitoring tasks."}],"snapshot_sha256":"d0fcdf051c9c3faa3c6730792595ead42e55e923da379ff3e2d053957aedbf4f"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2604.12645/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Although autonomous underwater vehicles promise the capability of marine ecosystem monitoring, their deployment is fundamentally limited by the difficulty of controlling vehicles under highly uncertain and non-stationary underwater dynamics. To address these challenges, we employ a data-driven reinforcement learning approach to compensate for unknown dynamics and task variations. Traditional single-task reinforcement learning has a tendency to overfit the training environment, thus, limit the long-term usefulness of the learnt policy. Hence, we propose to use a contextual multi-task reinforcem","authors_text":"Frank Kirchner, Mariela De Lucas Alvarez, Melvin Laux, Rebecca Adam, Rina Alo, S\\\"oren T\\\"opper, Yi-Ling Liu","cross_cats":["cs.AI"],"headline":"Contextual multi-task reinforcement learning trains one policy to handle multiple reef monitoring tasks.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-04-14T12:16:56Z","title":"Contextual Multi-Task Reinforcement Learning for Autonomous Reef Monitoring"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.12645","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-10T15:08:54.040768Z","id":"7cc8bb5f-551b-4202-a9da-37d9946a83cb","model_set":{"reader":"grok-4.3"},"one_line_summary":"A context-dependent multi-task RL policy is trained and evaluated in HoloOcean simulation to solve multiple reef monitoring tasks with claimed improvements in sample efficiency, zero-shot generalization, and robustness to water currents.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Contextual multi-task reinforcement learning trains one policy to handle multiple reef monitoring tasks.","strongest_claim":"We train a single context-dependent policy that is able to solve multiple related monitoring tasks in a simulated reef environment in HoloOcean.","weakest_assumption":"That performance gains observed in the HoloOcean simulator will translate to real-world underwater environments with highly uncertain and non-stationary dynamics."}},"verdict_id":"7cc8bb5f-551b-4202-a9da-37d9946a83cb"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:497b962ce596288a5adde7696482fd2c4a04dfe0424a2c1fdd1ac72de846eba3","target":"record","created_at":"2026-06-04T01:08:50Z","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":"bb3c172d3b226d9c19435e42539de97835efdc34c3ee87ef09baa4840e3b5d4e","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-04-14T12:16:56Z","title_canon_sha256":"142d5ac54bb02a1c6b6be8ec8c83df63fd0e7cd2f523964f178729e67a20a1af"},"schema_version":"1.0","source":{"id":"2604.12645","kind":"arxiv","version":2}},"canonical_sha256":"6ed5b08d04312fa17d850bb2bd9b563b6d1fbe4cc18669d34b0da23f0d68d137","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6ed5b08d04312fa17d850bb2bd9b563b6d1fbe4cc18669d34b0da23f0d68d137","first_computed_at":"2026-06-04T01:08:50.037051Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-04T01:08:50.037051Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fRJdSXmbwptTF2RvY8JnJ3RYFMygWpjvDhWTWTDDuaQHyTYszTjMKMo/HZYc2zLJzeCwNNKsiL8zI07AxIKoBA==","signature_status":"signed_v1","signed_at":"2026-06-04T01:08:50.037737Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.12645","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:497b962ce596288a5adde7696482fd2c4a04dfe0424a2c1fdd1ac72de846eba3","sha256:a78aa21cefcb05192b5da6f7c429946cfa75d5840e6250feaddeb834a6c201cc"],"state_sha256":"b5ac4c1f7dfabac8ace636925e563d6725c3c2e9706f845eb840f4944bcb7765"}