{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:3APZTNOU6JEGQV76FR45TZR64O","short_pith_number":"pith:3APZTNOU","canonical_record":{"source":{"id":"2505.17852","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T13:04:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c9a2fdaba493dcce4b87ee7c840eb1a7e1000a1ce609310782a29ba4f085ea1d","abstract_canon_sha256":"71a0aee2709c4c5410138886b173b867e187228125e15f03353e29dd20ec1a4f"},"schema_version":"1.0"},"canonical_sha256":"d81f99b5d4f2486857fe2c79d9e63ee39657e61b85661318dd2149272d2c1fca","source":{"kind":"arxiv","id":"2505.17852","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.17852","created_at":"2026-07-05T11:08:32Z"},{"alias_kind":"arxiv_version","alias_value":"2505.17852v1","created_at":"2026-07-05T11:08:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.17852","created_at":"2026-07-05T11:08:32Z"},{"alias_kind":"pith_short_12","alias_value":"3APZTNOU6JEG","created_at":"2026-07-05T11:08:32Z"},{"alias_kind":"pith_short_16","alias_value":"3APZTNOU6JEGQV76","created_at":"2026-07-05T11:08:32Z"},{"alias_kind":"pith_short_8","alias_value":"3APZTNOU","created_at":"2026-07-05T11:08:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:3APZTNOU6JEGQV76FR45TZR64O","target":"record","payload":{"canonical_record":{"source":{"id":"2505.17852","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T13:04:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c9a2fdaba493dcce4b87ee7c840eb1a7e1000a1ce609310782a29ba4f085ea1d","abstract_canon_sha256":"71a0aee2709c4c5410138886b173b867e187228125e15f03353e29dd20ec1a4f"},"schema_version":"1.0"},"canonical_sha256":"d81f99b5d4f2486857fe2c79d9e63ee39657e61b85661318dd2149272d2c1fca","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:32.887283Z","signature_b64":"N1PP7rMtaFV4EqEOHDMbSM4MYxUWCw9I859i+0L6fBvk91KOeStjn7+Nj+eEfH0CeEesVSeN6phULGLyqcC6Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d81f99b5d4f2486857fe2c79d9e63ee39657e61b85661318dd2149272d2c1fca","last_reissued_at":"2026-07-05T11:08:32.886808Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:32.886808Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.17852","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":"2026-07-05T11:08:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"P5UX6w72GXTTNrC3XLE3VxQIuqY1AsTpRUvjGqsD7eEAmmm4AwoYhx14sA0yPLKcCy4dCpletbiNOO4s7c/rDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T19:18:01.004004Z"},"content_sha256":"c34347b46a95cc66db268bf17c315b72b9705d94f8568a60637ba6f1d937e234","schema_version":"1.0","event_id":"sha256:c34347b46a95cc66db268bf17c315b72b9705d94f8568a60637ba6f1d937e234"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:3APZTNOU6JEGQV76FR45TZR64O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Scaling Recurrent Neural Networks to a Billion Parameters with Zero-Order Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Francois Chaubard, Mykel Kochenderfer","submitted_at":"2025-05-23T13:04:06Z","abstract_excerpt":"During inference, Recurrent Neural Networks (RNNs) scale constant in both FLOPs and GPU memory with increasing context length, as they compress all prior tokens into a fixed-size memory. In contrast, transformers scale linearly in FLOPs and, at best, linearly in memory during generation, since they must attend to all previous tokens explicitly. Despite this inference-time advantage, training large RNNs on long contexts remains impractical because standard optimization methods depend on Backpropagation Through Time (BPTT). BPTT requires retention of all intermediate activations during the forwa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.17852","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/2505.17852/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":"2026-07-05T11:08:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6IeamclCw13R6cB7vimbdP/5DodG+bNlBlChDloPCLnPZZeAenXe4pqR+l9fBatrOk62uTVsIFmobP4Qn+j8DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T19:18:01.004532Z"},"content_sha256":"42db128e9995b6e8cbac0bf12a2c9f7b4aab745b6baaf06efdbb01fcf64d2cc7","schema_version":"1.0","event_id":"sha256:42db128e9995b6e8cbac0bf12a2c9f7b4aab745b6baaf06efdbb01fcf64d2cc7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3APZTNOU6JEGQV76FR45TZR64O/bundle.json","state_url":"https://pith.science/pith/3APZTNOU6JEGQV76FR45TZR64O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3APZTNOU6JEGQV76FR45TZR64O/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":"2026-08-07T19:18:01Z","links":{"resolver":"https://pith.science/pith/3APZTNOU6JEGQV76FR45TZR64O","bundle":"https://pith.science/pith/3APZTNOU6JEGQV76FR45TZR64O/bundle.json","state":"https://pith.science/pith/3APZTNOU6JEGQV76FR45TZR64O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3APZTNOU6JEGQV76FR45TZR64O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:3APZTNOU6JEGQV76FR45TZR64O","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":"71a0aee2709c4c5410138886b173b867e187228125e15f03353e29dd20ec1a4f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T13:04:06Z","title_canon_sha256":"c9a2fdaba493dcce4b87ee7c840eb1a7e1000a1ce609310782a29ba4f085ea1d"},"schema_version":"1.0","source":{"id":"2505.17852","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.17852","created_at":"2026-07-05T11:08:32Z"},{"alias_kind":"arxiv_version","alias_value":"2505.17852v1","created_at":"2026-07-05T11:08:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.17852","created_at":"2026-07-05T11:08:32Z"},{"alias_kind":"pith_short_12","alias_value":"3APZTNOU6JEG","created_at":"2026-07-05T11:08:32Z"},{"alias_kind":"pith_short_16","alias_value":"3APZTNOU6JEGQV76","created_at":"2026-07-05T11:08:32Z"},{"alias_kind":"pith_short_8","alias_value":"3APZTNOU","created_at":"2026-07-05T11:08:32Z"}],"graph_snapshots":[{"event_id":"sha256:42db128e9995b6e8cbac0bf12a2c9f7b4aab745b6baaf06efdbb01fcf64d2cc7","target":"graph","created_at":"2026-07-05T11:08:32Z","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/2505.17852/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"During inference, Recurrent Neural Networks (RNNs) scale constant in both FLOPs and GPU memory with increasing context length, as they compress all prior tokens into a fixed-size memory. In contrast, transformers scale linearly in FLOPs and, at best, linearly in memory during generation, since they must attend to all previous tokens explicitly. Despite this inference-time advantage, training large RNNs on long contexts remains impractical because standard optimization methods depend on Backpropagation Through Time (BPTT). BPTT requires retention of all intermediate activations during the forwa","authors_text":"Francois Chaubard, Mykel Kochenderfer","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T13:04:06Z","title":"Scaling Recurrent Neural Networks to a Billion Parameters with Zero-Order Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.17852","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:c34347b46a95cc66db268bf17c315b72b9705d94f8568a60637ba6f1d937e234","target":"record","created_at":"2026-07-05T11:08:32Z","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":"71a0aee2709c4c5410138886b173b867e187228125e15f03353e29dd20ec1a4f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T13:04:06Z","title_canon_sha256":"c9a2fdaba493dcce4b87ee7c840eb1a7e1000a1ce609310782a29ba4f085ea1d"},"schema_version":"1.0","source":{"id":"2505.17852","kind":"arxiv","version":1}},"canonical_sha256":"d81f99b5d4f2486857fe2c79d9e63ee39657e61b85661318dd2149272d2c1fca","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d81f99b5d4f2486857fe2c79d9e63ee39657e61b85661318dd2149272d2c1fca","first_computed_at":"2026-07-05T11:08:32.886808Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:08:32.886808Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"N1PP7rMtaFV4EqEOHDMbSM4MYxUWCw9I859i+0L6fBvk91KOeStjn7+Nj+eEfH0CeEesVSeN6phULGLyqcC6Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:08:32.887283Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.17852","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c34347b46a95cc66db268bf17c315b72b9705d94f8568a60637ba6f1d937e234","sha256:42db128e9995b6e8cbac0bf12a2c9f7b4aab745b6baaf06efdbb01fcf64d2cc7"],"state_sha256":"4e56355519b12308923c8f7809c0346be224ff992b663928e86ca7ae2c7d1a93"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lsTqdJeW/Y2V4jb0YuL/fWFQIuiI/N46qcMpcxuCttEf7RfGVMEi+Tz6H3fQ6n1ikQOWlVohgP+CH2INZA7yAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T19:18:01.011609Z","bundle_sha256":"150e82d3b7d2168cabae6abd4d697ffa86dec455e6b6a0761b465257a81c6aa1"}}