{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:L753TDSLC55EF7WJJOJ43AWU4B","short_pith_number":"pith:L753TDSL","canonical_record":{"source":{"id":"2604.21355","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-04-23T07:14:35Z","cross_cats_sorted":[],"title_canon_sha256":"d0d69efc6442705db672094d080d603bc1374205ff97a89ebb0be2653cebcec3","abstract_canon_sha256":"c404544ebbc6a763666dd52328f2077d6b074a176294bc7cfa1e3af8085684f6"},"schema_version":"1.0"},"canonical_sha256":"5ffbb98e4b177a42fec94b93cd82d4e0407dfd0525dbe7ad1450b44ddff8eaee","source":{"kind":"arxiv","id":"2604.21355","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.21355","created_at":"2026-07-01T01:17:51Z"},{"alias_kind":"arxiv_version","alias_value":"2604.21355v2","created_at":"2026-07-01T01:17:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.21355","created_at":"2026-07-01T01:17:51Z"},{"alias_kind":"pith_short_12","alias_value":"L753TDSLC55E","created_at":"2026-07-01T01:17:51Z"},{"alias_kind":"pith_short_16","alias_value":"L753TDSLC55EF7WJ","created_at":"2026-07-01T01:17:51Z"},{"alias_kind":"pith_short_8","alias_value":"L753TDSL","created_at":"2026-07-01T01:17:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:L753TDSLC55EF7WJJOJ43AWU4B","target":"record","payload":{"canonical_record":{"source":{"id":"2604.21355","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-04-23T07:14:35Z","cross_cats_sorted":[],"title_canon_sha256":"d0d69efc6442705db672094d080d603bc1374205ff97a89ebb0be2653cebcec3","abstract_canon_sha256":"c404544ebbc6a763666dd52328f2077d6b074a176294bc7cfa1e3af8085684f6"},"schema_version":"1.0"},"canonical_sha256":"5ffbb98e4b177a42fec94b93cd82d4e0407dfd0525dbe7ad1450b44ddff8eaee","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-01T01:17:51.138878Z","signature_b64":"Zivy7/pb0WTlRyGUbwHir+QgyvyYwx3tuumtmZlcyoN6K8xE3seRNqf/2JNkMnDNbjGFKFzj3hnbbvxXz3ZzAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5ffbb98e4b177a42fec94b93cd82d4e0407dfd0525dbe7ad1450b44ddff8eaee","last_reissued_at":"2026-07-01T01:17:51.138459Z","signature_status":"signed_v1","first_computed_at":"2026-07-01T01:17:51.138459Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.21355","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":"2026-07-01T01:17:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"T8R4UxvwxV5g9Ttn8sJjPHNZzBjOiKujpjdDMpvj8ZvPWwHbXhEYE7Dfap4hbLX888WCFFAmBkjCkozq2VHSCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T09:08:28.592415Z"},"content_sha256":"9e2e29d35648d2cad3fa2ec3206fd645b57716198d29c6f998721bded4ab37df","schema_version":"1.0","event_id":"sha256:9e2e29d35648d2cad3fa2ec3206fd645b57716198d29c6f998721bded4ab37df"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:L753TDSLC55EF7WJJOJ43AWU4B","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RPG: Robust Policy Gating for Smooth Multi-Skill Transitions in Humanoid Fighting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Randomizing skill transitions and timings during training lets one policy produce stable, smooth multi-skill humanoid fighting.","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Bin Zhao, Dong Wang, Jiacheng Bao, Junbo Tan, Xuelong Li, Xueqian Wang, Yubo Dong, Yucheng Xin","submitted_at":"2026-04-23T07:14:35Z","abstract_excerpt":"Humanoid robots have demonstrated impressive motor skills in a wide range of tasks, yet whole-body control for humanlike long-time, dynamic fighting remains particularly challenging due to the stringent requirements on agility and stability. While imitation learning enables robots to execute human-like fighting skills, existing approaches often rely on switching among multiple single-skill policies or employing a general policy to imitate input reference motions. These strategies suffer from instability when transitioning between skills, as the mismatch of initial and terminal states across sk"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our approach incorporates motion transition randomization and temporal randomization to train a unified policy that generates agile fighting actions with stability and smoothness during skill transitions.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That deliberately randomizing skill transitions and timing during training will sufficiently eliminate out-of-domain disturbances caused by state mismatches across skills, both in simulation and on the physical Unitree G1.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"RPG trains a single policy with transition and timing randomization for stable multi-skill fighting on humanoids, integrated with locomotion for arbitrary-duration combat.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Randomizing skill transitions and timings during training lets one policy produce stable, smooth multi-skill humanoid fighting.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"8ce3f298c8dab46e4ad273715181357082d5d7b4b1c200dc47417f3fe9c33440"},"source":{"id":"2604.21355","kind":"arxiv","version":2},"verdict":{"id":"30cd55d5-7b5f-440a-87f4-308aaf628e78","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-09T22:10:52.668627Z","strongest_claim":"Our approach incorporates motion transition randomization and temporal randomization to train a unified policy that generates agile fighting actions with stability and smoothness during skill transitions.","one_line_summary":"RPG trains a single policy with transition and timing randomization for stable multi-skill fighting on humanoids, integrated with locomotion for arbitrary-duration combat.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That deliberately randomizing skill transitions and timing during training will sufficiently eliminate out-of-domain disturbances caused by state mismatches across skills, both in simulation and on the physical Unitree G1.","pith_extraction_headline":"Randomizing skill transitions and timings during training lets one policy produce stable, smooth multi-skill humanoid fighting."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.21355/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-21T12:43:43.829429Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-20T01:06:29.906164Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"17d4b1107c43753cb08b1470bc008c780a40b76ceef5bfb4f94c79f8cdce49cf"},"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":"30cd55d5-7b5f-440a-87f4-308aaf628e78"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-01T01:17:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y1BlchHsLoluP7Ac0TUf/FzDn6xP2x2mAqAyH3DYieva6664pFHDsZKiHQa0axqpfuz7hKcGAt2P0BU+j25bBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T09:08:28.593102Z"},"content_sha256":"7a8c1b0a0147e23bbbdd8448dc82a2f26389f21d3219f37ef02ba2e5537c1f54","schema_version":"1.0","event_id":"sha256:7a8c1b0a0147e23bbbdd8448dc82a2f26389f21d3219f37ef02ba2e5537c1f54"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/L753TDSLC55EF7WJJOJ43AWU4B/bundle.json","state_url":"https://pith.science/pith/L753TDSLC55EF7WJJOJ43AWU4B/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/L753TDSLC55EF7WJJOJ43AWU4B/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":"2026-08-06T09:08:28Z","links":{"resolver":"https://pith.science/pith/L753TDSLC55EF7WJJOJ43AWU4B","bundle":"https://pith.science/pith/L753TDSLC55EF7WJJOJ43AWU4B/bundle.json","state":"https://pith.science/pith/L753TDSLC55EF7WJJOJ43AWU4B/state.json","well_known_bundle":"https://pith.science/.well-known/pith/L753TDSLC55EF7WJJOJ43AWU4B/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:L753TDSLC55EF7WJJOJ43AWU4B","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":"c404544ebbc6a763666dd52328f2077d6b074a176294bc7cfa1e3af8085684f6","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-04-23T07:14:35Z","title_canon_sha256":"d0d69efc6442705db672094d080d603bc1374205ff97a89ebb0be2653cebcec3"},"schema_version":"1.0","source":{"id":"2604.21355","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.21355","created_at":"2026-07-01T01:17:51Z"},{"alias_kind":"arxiv_version","alias_value":"2604.21355v2","created_at":"2026-07-01T01:17:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.21355","created_at":"2026-07-01T01:17:51Z"},{"alias_kind":"pith_short_12","alias_value":"L753TDSLC55E","created_at":"2026-07-01T01:17:51Z"},{"alias_kind":"pith_short_16","alias_value":"L753TDSLC55EF7WJ","created_at":"2026-07-01T01:17:51Z"},{"alias_kind":"pith_short_8","alias_value":"L753TDSL","created_at":"2026-07-01T01:17:51Z"}],"graph_snapshots":[{"event_id":"sha256:7a8c1b0a0147e23bbbdd8448dc82a2f26389f21d3219f37ef02ba2e5537c1f54","target":"graph","created_at":"2026-07-01T01:17:51Z","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":"Our approach incorporates motion transition randomization and temporal randomization to train a unified policy that generates agile fighting actions with stability and smoothness during skill transitions."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That deliberately randomizing skill transitions and timing during training will sufficiently eliminate out-of-domain disturbances caused by state mismatches across skills, both in simulation and on the physical Unitree G1."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"RPG trains a single policy with transition and timing randomization for stable multi-skill fighting on humanoids, integrated with locomotion for arbitrary-duration combat."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Randomizing skill transitions and timings during training lets one policy produce stable, smooth multi-skill humanoid fighting."}],"snapshot_sha256":"8ce3f298c8dab46e4ad273715181357082d5d7b4b1c200dc47417f3fe9c33440"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-21T12:43:43.829429Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-20T01:06:29.906164Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.21355/integrity.json","findings":[],"snapshot_sha256":"17d4b1107c43753cb08b1470bc008c780a40b76ceef5bfb4f94c79f8cdce49cf","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Humanoid robots have demonstrated impressive motor skills in a wide range of tasks, yet whole-body control for humanlike long-time, dynamic fighting remains particularly challenging due to the stringent requirements on agility and stability. While imitation learning enables robots to execute human-like fighting skills, existing approaches often rely on switching among multiple single-skill policies or employing a general policy to imitate input reference motions. These strategies suffer from instability when transitioning between skills, as the mismatch of initial and terminal states across sk","authors_text":"Bin Zhao, Dong Wang, Jiacheng Bao, Junbo Tan, Xuelong Li, Xueqian Wang, Yubo Dong, Yucheng Xin","cross_cats":[],"headline":"Randomizing skill transitions and timings during training lets one policy produce stable, smooth multi-skill humanoid fighting.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-04-23T07:14:35Z","title":"RPG: Robust Policy Gating for Smooth Multi-Skill Transitions in Humanoid Fighting"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.21355","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-09T22:10:52.668627Z","id":"30cd55d5-7b5f-440a-87f4-308aaf628e78","model_set":{"reader":"grok-4.3"},"one_line_summary":"RPG trains a single policy with transition and timing randomization for stable multi-skill fighting on humanoids, integrated with locomotion for arbitrary-duration combat.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Randomizing skill transitions and timings during training lets one policy produce stable, smooth multi-skill humanoid fighting.","strongest_claim":"Our approach incorporates motion transition randomization and temporal randomization to train a unified policy that generates agile fighting actions with stability and smoothness during skill transitions.","weakest_assumption":"That deliberately randomizing skill transitions and timing during training will sufficiently eliminate out-of-domain disturbances caused by state mismatches across skills, both in simulation and on the physical Unitree G1."}},"verdict_id":"30cd55d5-7b5f-440a-87f4-308aaf628e78"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:9e2e29d35648d2cad3fa2ec3206fd645b57716198d29c6f998721bded4ab37df","target":"record","created_at":"2026-07-01T01:17:51Z","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":"c404544ebbc6a763666dd52328f2077d6b074a176294bc7cfa1e3af8085684f6","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-04-23T07:14:35Z","title_canon_sha256":"d0d69efc6442705db672094d080d603bc1374205ff97a89ebb0be2653cebcec3"},"schema_version":"1.0","source":{"id":"2604.21355","kind":"arxiv","version":2}},"canonical_sha256":"5ffbb98e4b177a42fec94b93cd82d4e0407dfd0525dbe7ad1450b44ddff8eaee","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5ffbb98e4b177a42fec94b93cd82d4e0407dfd0525dbe7ad1450b44ddff8eaee","first_computed_at":"2026-07-01T01:17:51.138459Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-01T01:17:51.138459Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Zivy7/pb0WTlRyGUbwHir+QgyvyYwx3tuumtmZlcyoN6K8xE3seRNqf/2JNkMnDNbjGFKFzj3hnbbvxXz3ZzAA==","signature_status":"signed_v1","signed_at":"2026-07-01T01:17:51.138878Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.21355","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9e2e29d35648d2cad3fa2ec3206fd645b57716198d29c6f998721bded4ab37df","sha256:7a8c1b0a0147e23bbbdd8448dc82a2f26389f21d3219f37ef02ba2e5537c1f54"],"state_sha256":"f71dc23a9e409ca60dece9dfce3c009922376ef1ac1c302a8c69205b6090dd7e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BRYqkphhUAFzucCkTfwnal2jzEcv5gmwVake0WzGjg6fVUhfd1j9zaNtfAGdoqdy32wP0x3ZrmwqjTGbhtKKBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T09:08:28.596815Z","bundle_sha256":"636fc98b38d5bced1fbc10a78ac7565a5834488649ecb2c65009141f9a5b5d4a"}}