{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:JQSXXCPUE367ZCZGZP5HEA2V74","short_pith_number":"pith:JQSXXCPU","canonical_record":{"source":{"id":"2206.00277","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-06-01T07:09:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2c86d4fb47418bd4ad58e8200740d1434d440ee1ca909c4c4904c90ebc9e8781","abstract_canon_sha256":"99933bb70fe5a21b95d381a211623dd19b90302a3ff60e9672b6b8eace243cdf"},"schema_version":"1.0"},"canonical_sha256":"4c257b89f426fdfc8b26cbfa720355ff065fd7189b6adcf96ca19eb2d75cd919","source":{"kind":"arxiv","id":"2206.00277","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2206.00277","created_at":"2026-07-05T04:28:32Z"},{"alias_kind":"arxiv_version","alias_value":"2206.00277v2","created_at":"2026-07-05T04:28:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.00277","created_at":"2026-07-05T04:28:32Z"},{"alias_kind":"pith_short_12","alias_value":"JQSXXCPUE367","created_at":"2026-07-05T04:28:32Z"},{"alias_kind":"pith_short_16","alias_value":"JQSXXCPUE367ZCZG","created_at":"2026-07-05T04:28:32Z"},{"alias_kind":"pith_short_8","alias_value":"JQSXXCPU","created_at":"2026-07-05T04:28:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:JQSXXCPUE367ZCZGZP5HEA2V74","target":"record","payload":{"canonical_record":{"source":{"id":"2206.00277","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-06-01T07:09:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2c86d4fb47418bd4ad58e8200740d1434d440ee1ca909c4c4904c90ebc9e8781","abstract_canon_sha256":"99933bb70fe5a21b95d381a211623dd19b90302a3ff60e9672b6b8eace243cdf"},"schema_version":"1.0"},"canonical_sha256":"4c257b89f426fdfc8b26cbfa720355ff065fd7189b6adcf96ca19eb2d75cd919","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:28:32.485167Z","signature_b64":"BFsTl2HxomIdgcAjLJ05gsEwJihLJlpzUOii/E/ckPT1hv6FNicVQsg5Ebnooau6IBzYX2GrHSctovSkjnY1Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4c257b89f426fdfc8b26cbfa720355ff065fd7189b6adcf96ca19eb2d75cd919","last_reissued_at":"2026-07-05T04:28:32.484684Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:28:32.484684Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2206.00277","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:28:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PduXInqlCerO/1p29WJV6utLHUkdI9iNqka2FyTadOYfgfmaPnoauq0uPtr5BW8dp23ucDpzyyp2oTtlSMQrCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T18:41:51.416961Z"},"content_sha256":"234dda284e91ff4c8c5b010e82ac9db9aa29b54fa1c583cab82ab58d16cb35c2","schema_version":"1.0","event_id":"sha256:234dda284e91ff4c8c5b010e82ac9db9aa29b54fa1c583cab82ab58d16cb35c2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:JQSXXCPUE367ZCZGZP5HEA2V74","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Task-Specific Expert Pruning for Sparse Mixture-of-Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Binxing Jiao, Daxin Jiang, Furu Wei, Haoyi Zhou, Jianxin Li, Shaohan Huang, Tianyu Chen, Yuan Xie","submitted_at":"2022-06-01T07:09:01Z","abstract_excerpt":"The sparse Mixture-of-Experts (MoE) model is powerful for large-scale pre-training and has achieved promising results due to its model capacity. However, with trillions of parameters, MoE is hard to be deployed on cloud or mobile environment. The inference of MoE requires expert parallelism, which is not hardware-friendly and communication expensive. Especially for resource-limited downstream tasks, such sparse structure has to sacrifice a lot of computing efficiency for limited performance gains. In this work, we observe most experts contribute scarcely little to the MoE fine-tuning and infer"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.00277","kind":"arxiv","version":2},"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/2206.00277/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":"2026-07-05T04:28:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dkTyPb2L+/zmrzoxYYUY87JgeV3dkf4IdenkZy9jY1ZQr9UHnlL0TYbxnM9E60kzBuSaTYiGqs5+ymR0fPsjDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T18:41:51.417341Z"},"content_sha256":"de366463462692903aff19242d0af9034f0e00669659f12126f2ab528946e7dd","schema_version":"1.0","event_id":"sha256:de366463462692903aff19242d0af9034f0e00669659f12126f2ab528946e7dd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JQSXXCPUE367ZCZGZP5HEA2V74/bundle.json","state_url":"https://pith.science/pith/JQSXXCPUE367ZCZGZP5HEA2V74/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JQSXXCPUE367ZCZGZP5HEA2V74/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":"2026-07-06T18:41:51Z","links":{"resolver":"https://pith.science/pith/JQSXXCPUE367ZCZGZP5HEA2V74","bundle":"https://pith.science/pith/JQSXXCPUE367ZCZGZP5HEA2V74/bundle.json","state":"https://pith.science/pith/JQSXXCPUE367ZCZGZP5HEA2V74/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JQSXXCPUE367ZCZGZP5HEA2V74/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:JQSXXCPUE367ZCZGZP5HEA2V74","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":"99933bb70fe5a21b95d381a211623dd19b90302a3ff60e9672b6b8eace243cdf","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-06-01T07:09:01Z","title_canon_sha256":"2c86d4fb47418bd4ad58e8200740d1434d440ee1ca909c4c4904c90ebc9e8781"},"schema_version":"1.0","source":{"id":"2206.00277","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2206.00277","created_at":"2026-07-05T04:28:32Z"},{"alias_kind":"arxiv_version","alias_value":"2206.00277v2","created_at":"2026-07-05T04:28:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.00277","created_at":"2026-07-05T04:28:32Z"},{"alias_kind":"pith_short_12","alias_value":"JQSXXCPUE367","created_at":"2026-07-05T04:28:32Z"},{"alias_kind":"pith_short_16","alias_value":"JQSXXCPUE367ZCZG","created_at":"2026-07-05T04:28:32Z"},{"alias_kind":"pith_short_8","alias_value":"JQSXXCPU","created_at":"2026-07-05T04:28:32Z"}],"graph_snapshots":[{"event_id":"sha256:de366463462692903aff19242d0af9034f0e00669659f12126f2ab528946e7dd","target":"graph","created_at":"2026-07-05T04:28:32Z","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/2206.00277/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The sparse Mixture-of-Experts (MoE) model is powerful for large-scale pre-training and has achieved promising results due to its model capacity. However, with trillions of parameters, MoE is hard to be deployed on cloud or mobile environment. The inference of MoE requires expert parallelism, which is not hardware-friendly and communication expensive. Especially for resource-limited downstream tasks, such sparse structure has to sacrifice a lot of computing efficiency for limited performance gains. In this work, we observe most experts contribute scarcely little to the MoE fine-tuning and infer","authors_text":"Binxing Jiao, Daxin Jiang, Furu Wei, Haoyi Zhou, Jianxin Li, Shaohan Huang, Tianyu Chen, Yuan Xie","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-06-01T07:09:01Z","title":"Task-Specific Expert Pruning for Sparse Mixture-of-Experts"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.00277","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":"sha256:234dda284e91ff4c8c5b010e82ac9db9aa29b54fa1c583cab82ab58d16cb35c2","target":"record","created_at":"2026-07-05T04:28:32Z","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":"99933bb70fe5a21b95d381a211623dd19b90302a3ff60e9672b6b8eace243cdf","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-06-01T07:09:01Z","title_canon_sha256":"2c86d4fb47418bd4ad58e8200740d1434d440ee1ca909c4c4904c90ebc9e8781"},"schema_version":"1.0","source":{"id":"2206.00277","kind":"arxiv","version":2}},"canonical_sha256":"4c257b89f426fdfc8b26cbfa720355ff065fd7189b6adcf96ca19eb2d75cd919","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4c257b89f426fdfc8b26cbfa720355ff065fd7189b6adcf96ca19eb2d75cd919","first_computed_at":"2026-07-05T04:28:32.484684Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:28:32.484684Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"BFsTl2HxomIdgcAjLJ05gsEwJihLJlpzUOii/E/ckPT1hv6FNicVQsg5Ebnooau6IBzYX2GrHSctovSkjnY1Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T04:28:32.485167Z","signed_message":"canonical_sha256_bytes"},"source_id":"2206.00277","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:234dda284e91ff4c8c5b010e82ac9db9aa29b54fa1c583cab82ab58d16cb35c2","sha256:de366463462692903aff19242d0af9034f0e00669659f12126f2ab528946e7dd"],"state_sha256":"71d34187145f61717c5286a5af3c2b6460530bd60d179491d8c050d1cc4341a7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Kn3cMa2HobMxiSlx4n8zggTc3lbFIFsD0j1ChS7Rz9MQNSGLBbPio/fg92W4ZFtp51Og7b0s+qzkUhx7DJbfBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T18:41:51.419334Z","bundle_sha256":"51ede89650534bab87810e69418368d3a5d4cd51e051f32d00f923b58cf0793f"}}