{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DEP2WS5QZP5J5BCIYIBNQY22QI","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":"db1b51c24e8513b6eea18e5ab024469c52c171a9aaa4c9f6ef6d8ab8640047be","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-04-08T17:11:29Z","title_canon_sha256":"5ed74519ed329a9a0390f8b49180d0f9d5fbe0382632f3847530e8f0024d5ea9"},"schema_version":"1.0","source":{"id":"2604.07303","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.07303","created_at":"2026-05-20T02:05:43Z"},{"alias_kind":"arxiv_version","alias_value":"2604.07303v2","created_at":"2026-05-20T02:05:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.07303","created_at":"2026-05-20T02:05:43Z"},{"alias_kind":"pith_short_12","alias_value":"DEP2WS5QZP5J","created_at":"2026-05-20T02:05:43Z"},{"alias_kind":"pith_short_16","alias_value":"DEP2WS5QZP5J5BCI","created_at":"2026-05-20T02:05:43Z"},{"alias_kind":"pith_short_8","alias_value":"DEP2WS5Q","created_at":"2026-05-20T02:05:43Z"}],"graph_snapshots":[{"event_id":"sha256:a1b834cf951286b30e45588d5c9b7e684bd2985db388a7e2cde890c8dce0040c","target":"graph","created_at":"2026-05-20T02:05:43Z","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":"The neural network-based fish model exhibited the smallest gap across goal-reaching performance and most other metrics, indicating higher behavioral fidelity than conventional rule-based models under this benchmark. More importantly, this separation shows that the proposed evaluation can quantitatively distinguish candidate models under matched closed-loop conditions."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That policies trained in simulation to guide a virtual fish to goals transfer sufficiently well to the physical RoboFish hardware and that the chosen behavioral metrics (goal-reaching, inter-individual distances, wall interactions, alignment) are sufficient to reveal meaningful differences in model fidelity."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A robotic fish uses RL policies to interact with live fish and ranks behavioral models by Wasserstein distance between simulated and real distributions of metrics such as goal-reaching performance and alignment."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A robotic fish distinguishes neural network models of fish behavior as more accurate than rule-based models through closed-loop interactions."}],"snapshot_sha256":"51e42cee4ab6b60fcb89a60aa6624f8ecd76934bb893977311c157577e352d25"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2604.07303/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Understanding and modeling animal behavior is essential for studying collective motion, decision-making, and bio-inspired robotics. Yet, evaluating the accuracy of behavioral models still often relies on offline comparisons to static trajectory statistics. Here we introduce a reinforcement-learning-based framework that uses a biomimetic robotic fish (RoboFish) to evaluate computational models of live fish behavior through closed-loop interaction. We trained policies in simulation using four distinct fish models-a simple constant-follow baseline, two rule-based models, and a biologically ground","authors_text":"Andreas Gerken, David Bierbach, Jens Krause, Mathis Hocke, Tim Landgraf","cross_cats":[],"headline":"A robotic fish distinguishes neural network models of fish behavior as more accurate than rule-based models through closed-loop interactions.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-04-08T17:11:29Z","title":"Robots that learn to evaluate models of collective behavior"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.07303","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-10T17:47:07.594410Z","id":"dbc9bb1e-a55d-4aad-963a-48d82131ed1a","model_set":{"reader":"grok-4.3"},"one_line_summary":"A robotic fish uses RL policies to interact with live fish and ranks behavioral models by Wasserstein distance between simulated and real distributions of metrics such as goal-reaching performance and alignment.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A robotic fish distinguishes neural network models of fish behavior as more accurate than rule-based models through closed-loop interactions.","strongest_claim":"The neural network-based fish model exhibited the smallest gap across goal-reaching performance and most other metrics, indicating higher behavioral fidelity than conventional rule-based models under this benchmark. More importantly, this separation shows that the proposed evaluation can quantitatively distinguish candidate models under matched closed-loop conditions.","weakest_assumption":"That policies trained in simulation to guide a virtual fish to goals transfer sufficiently well to the physical RoboFish hardware and that the chosen behavioral metrics (goal-reaching, inter-individual distances, wall interactions, alignment) are sufficient to reveal meaningful differences in model fidelity."}},"verdict_id":"dbc9bb1e-a55d-4aad-963a-48d82131ed1a"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:42fe2d568c5b7eb04c7e6bf6b0e8221c1ab39d706bd8206c2012925437fd2647","target":"record","created_at":"2026-05-20T02:05:43Z","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":"db1b51c24e8513b6eea18e5ab024469c52c171a9aaa4c9f6ef6d8ab8640047be","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-04-08T17:11:29Z","title_canon_sha256":"5ed74519ed329a9a0390f8b49180d0f9d5fbe0382632f3847530e8f0024d5ea9"},"schema_version":"1.0","source":{"id":"2604.07303","kind":"arxiv","version":2}},"canonical_sha256":"191fab4bb0cbfa9e8448c202d8635a823bfee45d5626aae23849ae4cf18880a7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"191fab4bb0cbfa9e8448c202d8635a823bfee45d5626aae23849ae4cf18880a7","first_computed_at":"2026-05-20T02:05:43.090292Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T02:05:43.090292Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LdtBoRqlbLhyoc2UYzQI0z0Fb+VZdr+qsH1+7bo4uskNxLap6rtkxHsYwFPHKnS/eIl/HRa3SNslImgaK/fZBg==","signature_status":"signed_v1","signed_at":"2026-05-20T02:05:43.091182Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.07303","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:42fe2d568c5b7eb04c7e6bf6b0e8221c1ab39d706bd8206c2012925437fd2647","sha256:a1b834cf951286b30e45588d5c9b7e684bd2985db388a7e2cde890c8dce0040c"],"state_sha256":"85bdffcb50bda7908cc92d116a893c30ad365f9d26bae2661eaa9fa7b78d4517"}