{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:6OAZJRETLC6UF4K7R7RRKRNQPW","short_pith_number":"pith:6OAZJRET","canonical_record":{"source":{"id":"2606.10305","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-06-09T01:46:23Z","cross_cats_sorted":[],"title_canon_sha256":"7006ec12c8b7e79be9e34c97e85ada602c92645d0914deeaac460b5d14bedaa1","abstract_canon_sha256":"95236a3280522cde2aca4f80c46f032615a6737605ae1460e9b267f42eae72c6"},"schema_version":"1.0"},"canonical_sha256":"f38194c49358bd42f15f8fe31545b07d848893f416046fc463029a7679cc3dc4","source":{"kind":"arxiv","id":"2606.10305","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.10305","created_at":"2026-06-10T01:10:10Z"},{"alias_kind":"arxiv_version","alias_value":"2606.10305v1","created_at":"2026-06-10T01:10:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.10305","created_at":"2026-06-10T01:10:10Z"},{"alias_kind":"pith_short_12","alias_value":"6OAZJRETLC6U","created_at":"2026-06-10T01:10:10Z"},{"alias_kind":"pith_short_16","alias_value":"6OAZJRETLC6UF4K7","created_at":"2026-06-10T01:10:10Z"},{"alias_kind":"pith_short_8","alias_value":"6OAZJRET","created_at":"2026-06-10T01:10:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:6OAZJRETLC6UF4K7R7RRKRNQPW","target":"record","payload":{"canonical_record":{"source":{"id":"2606.10305","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-06-09T01:46:23Z","cross_cats_sorted":[],"title_canon_sha256":"7006ec12c8b7e79be9e34c97e85ada602c92645d0914deeaac460b5d14bedaa1","abstract_canon_sha256":"95236a3280522cde2aca4f80c46f032615a6737605ae1460e9b267f42eae72c6"},"schema_version":"1.0"},"canonical_sha256":"f38194c49358bd42f15f8fe31545b07d848893f416046fc463029a7679cc3dc4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T01:10:10.859567Z","signature_b64":"49jXmYF65oSTLZeblRXjfBGPCxzvsM/qRrfhO4cVWcZjhGJAH+ofzWT7Kmsxlu/GtE9hTG+lLWjps/3cmIJdBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f38194c49358bd42f15f8fe31545b07d848893f416046fc463029a7679cc3dc4","last_reissued_at":"2026-06-10T01:10:10.858573Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T01:10:10.858573Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.10305","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":"2026-06-10T01:10:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"etvRK8+uX/3X6imYyprV8WYxGdF74IFBZXRL/K89L0JVU6cupAadxoA7bSqnHc1HmZ4TXBA2qEyYl+zwn+AGBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T18:24:54.709685Z"},"content_sha256":"f22a75a782a6ffe019633ac06cfe210f6804121dfc78de4ec879d10c2927ac34","schema_version":"1.0","event_id":"sha256:f22a75a782a6ffe019633ac06cfe210f6804121dfc78de4ec879d10c2927ac34"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:6OAZJRETLC6UF4K7R7RRKRNQPW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SARM2: Multi-Task Stage Aware Reward Modeling for Self Improving Robotic Manipulation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Chuan Wen, Hau Zheng, Jiankai Sun, Justin Yu, Ken Goldberg, Mac Schwager, Philipp Wu, Pieter Abbeel, Qianzhong Chen, Suning Huang, Yide Shentu","submitted_at":"2026-06-09T01:46:23Z","abstract_excerpt":"Fine-tuning vision-language-action (VLA) policies for long-horizon manipulation still relies heavily on behavior cloning, which requires costly high-quality demonstrations and keeps policies near the demonstration distribution. Reward models can reduce this dependence by reweighting demonstrations and providing dense supervision for on-robot reinforcement learning (RL), but they must be dense, accurate, and general. Existing methods fall short: task-specific stage-aware models are accurate but require per-task annotations, while general vision-language-model (VLM) reward models are broadly app"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.10305","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/2606.10305/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":"2026-06-10T01:10:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"slxQbMVNH3Q7fzNNn/t6F8Hu6bTAK8OIGmsvuA9jyvZzO1n6B1QANsWWrzCVUtfrqM66npRAlgmacvOMmCiiBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T18:24:54.710351Z"},"content_sha256":"0fc5ecf2c994f1431cd2b287507a84af0c7a415687f60e3c7db30147c386a69a","schema_version":"1.0","event_id":"sha256:0fc5ecf2c994f1431cd2b287507a84af0c7a415687f60e3c7db30147c386a69a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6OAZJRETLC6UF4K7R7RRKRNQPW/bundle.json","state_url":"https://pith.science/pith/6OAZJRETLC6UF4K7R7RRKRNQPW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6OAZJRETLC6UF4K7R7RRKRNQPW/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":"2026-06-11T18:24:54Z","links":{"resolver":"https://pith.science/pith/6OAZJRETLC6UF4K7R7RRKRNQPW","bundle":"https://pith.science/pith/6OAZJRETLC6UF4K7R7RRKRNQPW/bundle.json","state":"https://pith.science/pith/6OAZJRETLC6UF4K7R7RRKRNQPW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6OAZJRETLC6UF4K7R7RRKRNQPW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:6OAZJRETLC6UF4K7R7RRKRNQPW","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":"95236a3280522cde2aca4f80c46f032615a6737605ae1460e9b267f42eae72c6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-06-09T01:46:23Z","title_canon_sha256":"7006ec12c8b7e79be9e34c97e85ada602c92645d0914deeaac460b5d14bedaa1"},"schema_version":"1.0","source":{"id":"2606.10305","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.10305","created_at":"2026-06-10T01:10:10Z"},{"alias_kind":"arxiv_version","alias_value":"2606.10305v1","created_at":"2026-06-10T01:10:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.10305","created_at":"2026-06-10T01:10:10Z"},{"alias_kind":"pith_short_12","alias_value":"6OAZJRETLC6U","created_at":"2026-06-10T01:10:10Z"},{"alias_kind":"pith_short_16","alias_value":"6OAZJRETLC6UF4K7","created_at":"2026-06-10T01:10:10Z"},{"alias_kind":"pith_short_8","alias_value":"6OAZJRET","created_at":"2026-06-10T01:10:10Z"}],"graph_snapshots":[{"event_id":"sha256:0fc5ecf2c994f1431cd2b287507a84af0c7a415687f60e3c7db30147c386a69a","target":"graph","created_at":"2026-06-10T01:10:10Z","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/2606.10305/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Fine-tuning vision-language-action (VLA) policies for long-horizon manipulation still relies heavily on behavior cloning, which requires costly high-quality demonstrations and keeps policies near the demonstration distribution. Reward models can reduce this dependence by reweighting demonstrations and providing dense supervision for on-robot reinforcement learning (RL), but they must be dense, accurate, and general. Existing methods fall short: task-specific stage-aware models are accurate but require per-task annotations, while general vision-language-model (VLM) reward models are broadly app","authors_text":"Chuan Wen, Hau Zheng, Jiankai Sun, Justin Yu, Ken Goldberg, Mac Schwager, Philipp Wu, Pieter Abbeel, Qianzhong Chen, Suning Huang, Yide Shentu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-06-09T01:46:23Z","title":"SARM2: Multi-Task Stage Aware Reward Modeling for Self Improving Robotic Manipulation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.10305","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:f22a75a782a6ffe019633ac06cfe210f6804121dfc78de4ec879d10c2927ac34","target":"record","created_at":"2026-06-10T01:10:10Z","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":"95236a3280522cde2aca4f80c46f032615a6737605ae1460e9b267f42eae72c6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-06-09T01:46:23Z","title_canon_sha256":"7006ec12c8b7e79be9e34c97e85ada602c92645d0914deeaac460b5d14bedaa1"},"schema_version":"1.0","source":{"id":"2606.10305","kind":"arxiv","version":1}},"canonical_sha256":"f38194c49358bd42f15f8fe31545b07d848893f416046fc463029a7679cc3dc4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f38194c49358bd42f15f8fe31545b07d848893f416046fc463029a7679cc3dc4","first_computed_at":"2026-06-10T01:10:10.858573Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-10T01:10:10.858573Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"49jXmYF65oSTLZeblRXjfBGPCxzvsM/qRrfhO4cVWcZjhGJAH+ofzWT7Kmsxlu/GtE9hTG+lLWjps/3cmIJdBg==","signature_status":"signed_v1","signed_at":"2026-06-10T01:10:10.859567Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.10305","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f22a75a782a6ffe019633ac06cfe210f6804121dfc78de4ec879d10c2927ac34","sha256:0fc5ecf2c994f1431cd2b287507a84af0c7a415687f60e3c7db30147c386a69a"],"state_sha256":"5e1a8bfba930868b510b6e53bf253a2db05fcda6be6b6d6f67bd92e1b017aeca"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ln7AfYHhmk7TTZOm+n5LONeML19LGjyrgwsGj/m+rWk1UQGtx/9GIQ4bPDHmNZ4i8XHUhyrgQ5zOjm7JfYgoCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-11T18:24:54.714592Z","bundle_sha256":"e34d6dfbb75e2d75a847370ff1492bec15052e74b9ac1f0b34eeab43a94f4785"}}