{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ZHE5LMQIXJRVCGUN44IBNH7JYS","short_pith_number":"pith:ZHE5LMQI","canonical_record":{"source":{"id":"2507.03526","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-04T12:23:45Z","cross_cats_sorted":[],"title_canon_sha256":"7dc1113d1d7d9597abebce7c80c490290dffb51e8213ae08c2cefacc8face931","abstract_canon_sha256":"a46894127806282325edbf1e79d6540e8d737459528ccc3177635a50e0bf86eb"},"schema_version":"1.0"},"canonical_sha256":"c9c9d5b208ba63511a8de710169fe9c4a4f5c8e43069d52c0dbac029028625bc","source":{"kind":"arxiv","id":"2507.03526","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.03526","created_at":"2026-07-05T11:32:01Z"},{"alias_kind":"arxiv_version","alias_value":"2507.03526v1","created_at":"2026-07-05T11:32:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.03526","created_at":"2026-07-05T11:32:01Z"},{"alias_kind":"pith_short_12","alias_value":"ZHE5LMQIXJRV","created_at":"2026-07-05T11:32:01Z"},{"alias_kind":"pith_short_16","alias_value":"ZHE5LMQIXJRVCGUN","created_at":"2026-07-05T11:32:01Z"},{"alias_kind":"pith_short_8","alias_value":"ZHE5LMQI","created_at":"2026-07-05T11:32:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ZHE5LMQIXJRVCGUN44IBNH7JYS","target":"record","payload":{"canonical_record":{"source":{"id":"2507.03526","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-04T12:23:45Z","cross_cats_sorted":[],"title_canon_sha256":"7dc1113d1d7d9597abebce7c80c490290dffb51e8213ae08c2cefacc8face931","abstract_canon_sha256":"a46894127806282325edbf1e79d6540e8d737459528ccc3177635a50e0bf86eb"},"schema_version":"1.0"},"canonical_sha256":"c9c9d5b208ba63511a8de710169fe9c4a4f5c8e43069d52c0dbac029028625bc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:32:01.001834Z","signature_b64":"O1EqSwC59+WOh3vg1U6u1Hswd4J2j1MnnmTgnSBj5rYC+DGYUUbOtmZ/q23IxhOaTZm9aH2WzJFFTP0lck65Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c9c9d5b208ba63511a8de710169fe9c4a4f5c8e43069d52c0dbac029028625bc","last_reissued_at":"2026-07-05T11:32:01.001335Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:32:01.001335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.03526","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":"2026-07-05T11:32:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vZHByHyvyn9RDcZmTPWgqfZ9MNnEM7QWF4uKgJPO/HUsfolDi1W/gllUIzpOALeFxN+SxvwZ0QiMHa2M60+ZCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T15:00:14.314589Z"},"content_sha256":"5132398853182119302cc3cb8db3177e328f71108ab461e644b73fb925f63905","schema_version":"1.0","event_id":"sha256:5132398853182119302cc3cb8db3177e328f71108ab461e644b73fb925f63905"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ZHE5LMQIXJRVCGUN44IBNH7JYS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Decoupled Relative Learning Rate Schedules","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jakub Krajewski, Jan Ludziejewski, Jan Ma{\\l}a\\'snicki, Kamil Adamczewski, Kamil Ciebiera, Maciej Pi\\'oro, Maciej Stefaniak, Marek Cygan, Micha{\\l} Krutul, Piotr Sankowski, Sebastian Jaszczur","submitted_at":"2025-07-04T12:23:45Z","abstract_excerpt":"In this work, we introduce a novel approach for optimizing LLM training by adjusting learning rates across weights of different components in Transformer models. Traditional methods often apply a uniform learning rate across all network layers, potentially overlooking the unique dynamics of each part. Remarkably, our introduced relative learning rates, RLRS, method accelerates the training process by up to $23\\%$, particularly in complex models such as Mixture of Experts (MoE). Hyperparameters of RLRS can be efficiently tuned on smaller models and then effectively reused on models up to $27\\ti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.03526","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/2507.03526/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":"2026-07-05T11:32:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cpOUbpaKbs4GEG0QGoUMW1gkyXw586/1gZtHHT43I4CDgALDp6f/+OhJR73bXFSrLHJsALryIIzdjUkuhC0IAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T15:00:14.314961Z"},"content_sha256":"c09d0dcb34bd77f6409acabbd8f487a6149b8b8993a70af720a53acc89ff7762","schema_version":"1.0","event_id":"sha256:c09d0dcb34bd77f6409acabbd8f487a6149b8b8993a70af720a53acc89ff7762"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZHE5LMQIXJRVCGUN44IBNH7JYS/bundle.json","state_url":"https://pith.science/pith/ZHE5LMQIXJRVCGUN44IBNH7JYS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZHE5LMQIXJRVCGUN44IBNH7JYS/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":"2026-07-06T15:00:14Z","links":{"resolver":"https://pith.science/pith/ZHE5LMQIXJRVCGUN44IBNH7JYS","bundle":"https://pith.science/pith/ZHE5LMQIXJRVCGUN44IBNH7JYS/bundle.json","state":"https://pith.science/pith/ZHE5LMQIXJRVCGUN44IBNH7JYS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZHE5LMQIXJRVCGUN44IBNH7JYS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZHE5LMQIXJRVCGUN44IBNH7JYS","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":"a46894127806282325edbf1e79d6540e8d737459528ccc3177635a50e0bf86eb","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-04T12:23:45Z","title_canon_sha256":"7dc1113d1d7d9597abebce7c80c490290dffb51e8213ae08c2cefacc8face931"},"schema_version":"1.0","source":{"id":"2507.03526","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.03526","created_at":"2026-07-05T11:32:01Z"},{"alias_kind":"arxiv_version","alias_value":"2507.03526v1","created_at":"2026-07-05T11:32:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.03526","created_at":"2026-07-05T11:32:01Z"},{"alias_kind":"pith_short_12","alias_value":"ZHE5LMQIXJRV","created_at":"2026-07-05T11:32:01Z"},{"alias_kind":"pith_short_16","alias_value":"ZHE5LMQIXJRVCGUN","created_at":"2026-07-05T11:32:01Z"},{"alias_kind":"pith_short_8","alias_value":"ZHE5LMQI","created_at":"2026-07-05T11:32:01Z"}],"graph_snapshots":[{"event_id":"sha256:c09d0dcb34bd77f6409acabbd8f487a6149b8b8993a70af720a53acc89ff7762","target":"graph","created_at":"2026-07-05T11:32:01Z","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/2507.03526/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In this work, we introduce a novel approach for optimizing LLM training by adjusting learning rates across weights of different components in Transformer models. Traditional methods often apply a uniform learning rate across all network layers, potentially overlooking the unique dynamics of each part. Remarkably, our introduced relative learning rates, RLRS, method accelerates the training process by up to $23\\%$, particularly in complex models such as Mixture of Experts (MoE). Hyperparameters of RLRS can be efficiently tuned on smaller models and then effectively reused on models up to $27\\ti","authors_text":"Jakub Krajewski, Jan Ludziejewski, Jan Ma{\\l}a\\'snicki, Kamil Adamczewski, Kamil Ciebiera, Maciej Pi\\'oro, Maciej Stefaniak, Marek Cygan, Micha{\\l} Krutul, Piotr Sankowski, Sebastian Jaszczur","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-04T12:23:45Z","title":"Decoupled Relative Learning Rate Schedules"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.03526","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":"sha256:5132398853182119302cc3cb8db3177e328f71108ab461e644b73fb925f63905","target":"record","created_at":"2026-07-05T11:32:01Z","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":"a46894127806282325edbf1e79d6540e8d737459528ccc3177635a50e0bf86eb","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-04T12:23:45Z","title_canon_sha256":"7dc1113d1d7d9597abebce7c80c490290dffb51e8213ae08c2cefacc8face931"},"schema_version":"1.0","source":{"id":"2507.03526","kind":"arxiv","version":1}},"canonical_sha256":"c9c9d5b208ba63511a8de710169fe9c4a4f5c8e43069d52c0dbac029028625bc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c9c9d5b208ba63511a8de710169fe9c4a4f5c8e43069d52c0dbac029028625bc","first_computed_at":"2026-07-05T11:32:01.001335Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:32:01.001335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"O1EqSwC59+WOh3vg1U6u1Hswd4J2j1MnnmTgnSBj5rYC+DGYUUbOtmZ/q23IxhOaTZm9aH2WzJFFTP0lck65Dw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:32:01.001834Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.03526","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5132398853182119302cc3cb8db3177e328f71108ab461e644b73fb925f63905","sha256:c09d0dcb34bd77f6409acabbd8f487a6149b8b8993a70af720a53acc89ff7762"],"state_sha256":"7ded6db5891783bb4f26515dff150d6ff2e02c1dc625fc22fa6b9bef8400ae54"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/745BDaSmHk4tCluLGb9VC0RJGDRmI/CzP77V21NILZATAsxvjS3Z78JWP9ujBXNbE9SjHiFKcInUT2pO7w3Dg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T15:00:14.316941Z","bundle_sha256":"1ffbe1a1c747f711331a9e7560471a564f8350de4d80ca8a714d3e57b4c1c028"}}