{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PIAYRC3NQE44GBMMV5O7JNSWDN","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":"8929134f57cccb7ef57721eb90591d3ff5b5535afbe2d57c1f88adad48491b4a","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-26T16:06:36Z","title_canon_sha256":"ab8508c702ce42d944b7c90e06c3dc28843e02c944c16c717da09ba6acf75c1f"},"schema_version":"1.0","source":{"id":"2502.19261","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.19261","created_at":"2026-07-05T10:31:54Z"},{"alias_kind":"arxiv_version","alias_value":"2502.19261v2","created_at":"2026-07-05T10:31:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.19261","created_at":"2026-07-05T10:31:54Z"},{"alias_kind":"pith_short_12","alias_value":"PIAYRC3NQE44","created_at":"2026-07-05T10:31:54Z"},{"alias_kind":"pith_short_16","alias_value":"PIAYRC3NQE44GBMM","created_at":"2026-07-05T10:31:54Z"},{"alias_kind":"pith_short_8","alias_value":"PIAYRC3N","created_at":"2026-07-05T10:31:54Z"}],"graph_snapshots":[{"event_id":"sha256:a2035a70b6d5f08c96af2adab2479f09a4c18bb774dd8adec8ea685d15b76f73","target":"graph","created_at":"2026-07-05T10:31:54Z","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/2502.19261/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The Mixture of Experts (MoE) architecture reduces the training and inference cost significantly compared to a dense model of equivalent capacity. Upcycling is an approach that initializes and trains an MoE model using a pre-trained dense model. While upcycling leads to initial performance gains, the training progresses slower than when trained from scratch, leading to suboptimal performance in the long term. We propose Drop-Upcycling - a method that effectively addresses this problem. Drop-Upcycling combines two seemingly contradictory approaches: utilizing the knowledge of pre-trained dense m","authors_text":"Jun Suzuki, Kazuki Fujii, Rio Yokota, Taishi Nakamura, Takuya Akiba, Yusuke Oda","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-26T16:06:36Z","title":"Drop-Upcycling: Training Sparse Mixture of Experts with Partial Re-initialization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.19261","kind":"arxiv","version":2},"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":"sha256:01129c85b11c54d9890738ce53ebc761e657b15f67688569e3b22f2f6c6b9198","target":"record","created_at":"2026-07-05T10:31:54Z","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":"8929134f57cccb7ef57721eb90591d3ff5b5535afbe2d57c1f88adad48491b4a","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-26T16:06:36Z","title_canon_sha256":"ab8508c702ce42d944b7c90e06c3dc28843e02c944c16c717da09ba6acf75c1f"},"schema_version":"1.0","source":{"id":"2502.19261","kind":"arxiv","version":2}},"canonical_sha256":"7a01888b6d8139c3058caf5df4b6561b77fff0ed111aa3525a8b779c7599dcaf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7a01888b6d8139c3058caf5df4b6561b77fff0ed111aa3525a8b779c7599dcaf","first_computed_at":"2026-07-05T10:31:54.858107Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:31:54.858107Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3tcj0I/pBE6xfln8LJ77rAqwNZyf4cxZIQem223TiaoODBTJiCN1rEx2P20zMqeMXFoiovUobnqtXT/1YrZ4DQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:31:54.859038Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.19261","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:01129c85b11c54d9890738ce53ebc761e657b15f67688569e3b22f2f6c6b9198","sha256:a2035a70b6d5f08c96af2adab2479f09a4c18bb774dd8adec8ea685d15b76f73"],"state_sha256":"dec3975236f4b6282a18e0d3c7740001b24a7fbfefd197dfd649c0395491b2f1"}