{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:P3BZ54GPSKAE6SHSXHUAAMB6WD","short_pith_number":"pith:P3BZ54GP","canonical_record":{"source":{"id":"2608.03872","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-08-04T16:12:05Z","cross_cats_sorted":[],"title_canon_sha256":"3ec964744fb72b90d22e2a6d4a7ed18ea9e8feff357fd073b2ba572086494a08","abstract_canon_sha256":"37546a0448cab3f77265b62115c1eed856c1403d224b7770737486ca3b7c89d6"},"schema_version":"1.0"},"canonical_sha256":"7ec39ef0cf92804f48f2b9e800303eb0c91796e15df011fdd6f4d0981acd9a5c","source":{"kind":"arxiv","id":"2608.03872","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.03872","created_at":"2026-08-05T01:37:17Z"},{"alias_kind":"arxiv_version","alias_value":"2608.03872v1","created_at":"2026-08-05T01:37:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.03872","created_at":"2026-08-05T01:37:17Z"},{"alias_kind":"pith_short_12","alias_value":"P3BZ54GPSKAE","created_at":"2026-08-05T01:37:17Z"},{"alias_kind":"pith_short_16","alias_value":"P3BZ54GPSKAE6SHS","created_at":"2026-08-05T01:37:17Z"},{"alias_kind":"pith_short_8","alias_value":"P3BZ54GP","created_at":"2026-08-05T01:37:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:P3BZ54GPSKAE6SHSXHUAAMB6WD","target":"record","payload":{"canonical_record":{"source":{"id":"2608.03872","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-08-04T16:12:05Z","cross_cats_sorted":[],"title_canon_sha256":"3ec964744fb72b90d22e2a6d4a7ed18ea9e8feff357fd073b2ba572086494a08","abstract_canon_sha256":"37546a0448cab3f77265b62115c1eed856c1403d224b7770737486ca3b7c89d6"},"schema_version":"1.0"},"canonical_sha256":"7ec39ef0cf92804f48f2b9e800303eb0c91796e15df011fdd6f4d0981acd9a5c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-05T01:37:17.735735Z","signature_b64":"PGp6IIlzP1SGQAEsV5FFO7Yef2huBgnjgLQnNPwWLE/NI7I2u/xekyuzAtu/BNTTFV8B/z1knrEipzu+1gVFCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7ec39ef0cf92804f48f2b9e800303eb0c91796e15df011fdd6f4d0981acd9a5c","last_reissued_at":"2026-08-05T01:37:17.734107Z","signature_status":"signed_v1","first_computed_at":"2026-08-05T01:37:17.734107Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.03872","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-05T01:37:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jA1avMYA6dY7BOPDpXT9HsdWFwlkOVw8owebCrWegsMOPPZDzlXdNJtqp0MOFU9cuMttNrKiM7PONLPbROH4BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T23:53:45.277518Z"},"content_sha256":"6c09258569839fb97dd27e17af6875202296de0390573d9e3f9c013cd4c551de","schema_version":"1.0","event_id":"sha256:6c09258569839fb97dd27e17af6875202296de0390573d9e3f9c013cd4c551de"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:P3BZ54GPSKAE6SHSXHUAAMB6WD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"EvoHIL: Self-Evolving Reward and Flow-Matched Policy Optimization for Robust Human-in-the-Loop Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Bin Zheng, Jiahao Tu, Jia Pan, Jinzhuo Peng, Kai Liu, Ke Wang, Shuoqin Zhang, Tongtong Cheng, Xiru Gao, Zhe Hu","submitted_at":"2026-08-04T16:12:05Z","abstract_excerpt":"Human-in-the-loop reinforcement learning (HIL-RL) enables robots to learn contact-rich manipulation from limited real-world interaction, but deployment exposes three coupled limitations: static visual reward models fail under scene changes; independently sampled actions cause temporally inconsistent motion; and vision-based policies remain sensitive to appearance shifts. We present EvoHIL, a unified framework that adapts the reward model, action generator, and visual do main within a staged human-in-the-loop learning process. First, self-evolving reward (SER) adapts the success classifier from"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.03872","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/2608.03872/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-05T01:37:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VQLNy2AgbHhCvVMUY9IYT49ybRt/oUrN7RHeIL9BOLeDF1Iielwaoe+IVJQuIZoN5Xm0r8XbpGfkECRo/hs8Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T23:53:45.278531Z"},"content_sha256":"ea055717892c85c8a38db09e16c4e1b3b253fcb71a872dbdc4b7c6b22ee9f07a","schema_version":"1.0","event_id":"sha256:ea055717892c85c8a38db09e16c4e1b3b253fcb71a872dbdc4b7c6b22ee9f07a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/P3BZ54GPSKAE6SHSXHUAAMB6WD/bundle.json","state_url":"https://pith.science/pith/P3BZ54GPSKAE6SHSXHUAAMB6WD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/P3BZ54GPSKAE6SHSXHUAAMB6WD/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":"2026-08-08T23:53:45Z","links":{"resolver":"https://pith.science/pith/P3BZ54GPSKAE6SHSXHUAAMB6WD","bundle":"https://pith.science/pith/P3BZ54GPSKAE6SHSXHUAAMB6WD/bundle.json","state":"https://pith.science/pith/P3BZ54GPSKAE6SHSXHUAAMB6WD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/P3BZ54GPSKAE6SHSXHUAAMB6WD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:P3BZ54GPSKAE6SHSXHUAAMB6WD","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":"37546a0448cab3f77265b62115c1eed856c1403d224b7770737486ca3b7c89d6","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-08-04T16:12:05Z","title_canon_sha256":"3ec964744fb72b90d22e2a6d4a7ed18ea9e8feff357fd073b2ba572086494a08"},"schema_version":"1.0","source":{"id":"2608.03872","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.03872","created_at":"2026-08-05T01:37:17Z"},{"alias_kind":"arxiv_version","alias_value":"2608.03872v1","created_at":"2026-08-05T01:37:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.03872","created_at":"2026-08-05T01:37:17Z"},{"alias_kind":"pith_short_12","alias_value":"P3BZ54GPSKAE","created_at":"2026-08-05T01:37:17Z"},{"alias_kind":"pith_short_16","alias_value":"P3BZ54GPSKAE6SHS","created_at":"2026-08-05T01:37:17Z"},{"alias_kind":"pith_short_8","alias_value":"P3BZ54GP","created_at":"2026-08-05T01:37:17Z"}],"graph_snapshots":[{"event_id":"sha256:ea055717892c85c8a38db09e16c4e1b3b253fcb71a872dbdc4b7c6b22ee9f07a","target":"graph","created_at":"2026-08-05T01:37:17Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2608.03872/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Human-in-the-loop reinforcement learning (HIL-RL) enables robots to learn contact-rich manipulation from limited real-world interaction, but deployment exposes three coupled limitations: static visual reward models fail under scene changes; independently sampled actions cause temporally inconsistent motion; and vision-based policies remain sensitive to appearance shifts. We present EvoHIL, a unified framework that adapts the reward model, action generator, and visual do main within a staged human-in-the-loop learning process. First, self-evolving reward (SER) adapts the success classifier from","authors_text":"Bin Zheng, Jiahao Tu, Jia Pan, Jinzhuo Peng, Kai Liu, Ke Wang, Shuoqin Zhang, Tongtong Cheng, Xiru Gao, Zhe Hu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-08-04T16:12:05Z","title":"EvoHIL: Self-Evolving Reward and Flow-Matched Policy Optimization for Robust Human-in-the-Loop Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.03872","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6c09258569839fb97dd27e17af6875202296de0390573d9e3f9c013cd4c551de","target":"record","created_at":"2026-08-05T01:37:17Z","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":"37546a0448cab3f77265b62115c1eed856c1403d224b7770737486ca3b7c89d6","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-08-04T16:12:05Z","title_canon_sha256":"3ec964744fb72b90d22e2a6d4a7ed18ea9e8feff357fd073b2ba572086494a08"},"schema_version":"1.0","source":{"id":"2608.03872","kind":"arxiv","version":1}},"canonical_sha256":"7ec39ef0cf92804f48f2b9e800303eb0c91796e15df011fdd6f4d0981acd9a5c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7ec39ef0cf92804f48f2b9e800303eb0c91796e15df011fdd6f4d0981acd9a5c","first_computed_at":"2026-08-05T01:37:17.734107Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-05T01:37:17.734107Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PGp6IIlzP1SGQAEsV5FFO7Yef2huBgnjgLQnNPwWLE/NI7I2u/xekyuzAtu/BNTTFV8B/z1knrEipzu+1gVFCA==","signature_status":"signed_v1","signed_at":"2026-08-05T01:37:17.735735Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.03872","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6c09258569839fb97dd27e17af6875202296de0390573d9e3f9c013cd4c551de","sha256:ea055717892c85c8a38db09e16c4e1b3b253fcb71a872dbdc4b7c6b22ee9f07a"],"state_sha256":"194e0ae97484870d323f708699f4a00dd8922520f1a4f8a5abd4dc3e960dc5fc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qlRAr28UNbwoHexb4TQXruEF1J1AsEmSx5HithHbMkr4HrZB878VYEw3cShG6KGsAwrO5jskZYPC51UvKQkxCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T23:53:45.284998Z","bundle_sha256":"26af18be725d822d950b45352c3ea0121a27058928b1ef31268853caa497c8a4"}}