{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KE3EBX6O2PC3ROU3C5YS4HXSI2","short_pith_number":"pith:KE3EBX6O","schema_version":"1.0","canonical_sha256":"513640dfced3c5b8ba9b17712e1ef2468e7a5bfae4fe08988cd1a759d78e63c2","source":{"kind":"arxiv","id":"2606.26392","version":1},"attestation_state":"computed","paper":{"title":"MPC-Injection: Biasing Off-Policy Locomotion RL Toward Controller-Induced Behavior Basins","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Brian Plancher (Dartmouth College), Roy Xing (Dartmouth College), Seyoung Ree (Harvard University)","submitted_at":"2026-06-24T21:23:22Z","abstract_excerpt":"Reinforcement learning (RL) for locomotion frequently converges to locally optimal but undeployable behaviors, such as vibrating limbs or scooting on the torso, that maximize return without producing a usable gait. We present MPC-Injection, a low-overhead method that steers RL toward a designer-preferred gait by inserting transitions into the replay buffer from a model predictive controller solving the same Markov decision process. Unlike reward shaping, MPC-Injection does not require redesigning the task reward, and unlike adversarial imitation learning, it adds no discriminator, no kinematic"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.26392","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-06-24T21:23:22Z","cross_cats_sorted":[],"title_canon_sha256":"8357aa4f80a9e363ea7ab3211bd87e2af21a4bdc452e0b6ac9104d9e40825886","abstract_canon_sha256":"5d8cd0c8b458353fade9d82f93efa523899bc4311a09cae079deadd5ac32e304"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-26T00:15:39.736165Z","signature_b64":"Iz0kqbkyOthPvQ+eIa1MGH1NrC4K4nMIoTrQuyPmhw5oNMnGwxVYQN89L3lPlpa9PCpWClHVPNQDcqw0zWW+Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"513640dfced3c5b8ba9b17712e1ef2468e7a5bfae4fe08988cd1a759d78e63c2","last_reissued_at":"2026-06-26T00:15:39.735619Z","signature_status":"signed_v1","first_computed_at":"2026-06-26T00:15:39.735619Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MPC-Injection: Biasing Off-Policy Locomotion RL Toward Controller-Induced Behavior Basins","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Brian Plancher (Dartmouth College), Roy Xing (Dartmouth College), Seyoung Ree (Harvard University)","submitted_at":"2026-06-24T21:23:22Z","abstract_excerpt":"Reinforcement learning (RL) for locomotion frequently converges to locally optimal but undeployable behaviors, such as vibrating limbs or scooting on the torso, that maximize return without producing a usable gait. We present MPC-Injection, a low-overhead method that steers RL toward a designer-preferred gait by inserting transitions into the replay buffer from a model predictive controller solving the same Markov decision process. Unlike reward shaping, MPC-Injection does not require redesigning the task reward, and unlike adversarial imitation learning, it adds no discriminator, no kinematic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26392","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/2606.26392/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.26392","created_at":"2026-06-26T00:15:39.735702+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.26392v1","created_at":"2026-06-26T00:15:39.735702+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26392","created_at":"2026-06-26T00:15:39.735702+00:00"},{"alias_kind":"pith_short_12","alias_value":"KE3EBX6O2PC3","created_at":"2026-06-26T00:15:39.735702+00:00"},{"alias_kind":"pith_short_16","alias_value":"KE3EBX6O2PC3ROU3","created_at":"2026-06-26T00:15:39.735702+00:00"},{"alias_kind":"pith_short_8","alias_value":"KE3EBX6O","created_at":"2026-06-26T00:15:39.735702+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KE3EBX6O2PC3ROU3C5YS4HXSI2","json":"https://pith.science/pith/KE3EBX6O2PC3ROU3C5YS4HXSI2.json","graph_json":"https://pith.science/api/pith-number/KE3EBX6O2PC3ROU3C5YS4HXSI2/graph.json","events_json":"https://pith.science/api/pith-number/KE3EBX6O2PC3ROU3C5YS4HXSI2/events.json","paper":"https://pith.science/paper/KE3EBX6O"},"agent_actions":{"view_html":"https://pith.science/pith/KE3EBX6O2PC3ROU3C5YS4HXSI2","download_json":"https://pith.science/pith/KE3EBX6O2PC3ROU3C5YS4HXSI2.json","view_paper":"https://pith.science/paper/KE3EBX6O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.26392&json=true","fetch_graph":"https://pith.science/api/pith-number/KE3EBX6O2PC3ROU3C5YS4HXSI2/graph.json","fetch_events":"https://pith.science/api/pith-number/KE3EBX6O2PC3ROU3C5YS4HXSI2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KE3EBX6O2PC3ROU3C5YS4HXSI2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KE3EBX6O2PC3ROU3C5YS4HXSI2/action/storage_attestation","attest_author":"https://pith.science/pith/KE3EBX6O2PC3ROU3C5YS4HXSI2/action/author_attestation","sign_citation":"https://pith.science/pith/KE3EBX6O2PC3ROU3C5YS4HXSI2/action/citation_signature","submit_replication":"https://pith.science/pith/KE3EBX6O2PC3ROU3C5YS4HXSI2/action/replication_record"}},"created_at":"2026-06-26T00:15:39.735702+00:00","updated_at":"2026-06-26T00:15:39.735702+00:00"}