{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:SEFHANCVMB2BHWEVQV3FNP7AV3","short_pith_number":"pith:SEFHANCV","canonical_record":{"source":{"id":"2502.12928","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-18T15:09:58Z","cross_cats_sorted":[],"title_canon_sha256":"120348668b686c6d33681c6c95886389f78f068b817fc7a745ee76e48876620f","abstract_canon_sha256":"b7311791acab1f653309151e94bad39a763b54bf70168e416426cbd02ecb3b27"},"schema_version":"1.0"},"canonical_sha256":"910a703455607413d895857656bfe0aed4bab01a54fa00bdf5c931172e0b350c","source":{"kind":"arxiv","id":"2502.12928","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.12928","created_at":"2026-07-05T10:16:26Z"},{"alias_kind":"arxiv_version","alias_value":"2502.12928v1","created_at":"2026-07-05T10:16:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.12928","created_at":"2026-07-05T10:16:26Z"},{"alias_kind":"pith_short_12","alias_value":"SEFHANCVMB2B","created_at":"2026-07-05T10:16:26Z"},{"alias_kind":"pith_short_16","alias_value":"SEFHANCVMB2BHWEV","created_at":"2026-07-05T10:16:26Z"},{"alias_kind":"pith_short_8","alias_value":"SEFHANCV","created_at":"2026-07-05T10:16:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:SEFHANCVMB2BHWEVQV3FNP7AV3","target":"record","payload":{"canonical_record":{"source":{"id":"2502.12928","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-18T15:09:58Z","cross_cats_sorted":[],"title_canon_sha256":"120348668b686c6d33681c6c95886389f78f068b817fc7a745ee76e48876620f","abstract_canon_sha256":"b7311791acab1f653309151e94bad39a763b54bf70168e416426cbd02ecb3b27"},"schema_version":"1.0"},"canonical_sha256":"910a703455607413d895857656bfe0aed4bab01a54fa00bdf5c931172e0b350c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:16:26.450656Z","signature_b64":"VwQt2XeQB9v8JGQRmgiuywWebU1P2oMhy6NPSU2ba6jsxaVPv8lZyfh/sZx+8QRYM1tM4xWN8fR+CiIK5bN0Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"910a703455607413d895857656bfe0aed4bab01a54fa00bdf5c931172e0b350c","last_reissued_at":"2026-07-05T10:16:26.450226Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:16:26.450226Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.12928","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":"2026-07-05T10:16:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gOMzs9yfcCRi/vOx2EFTChqCr6CqCt82esRmuJZIShIXUB9cXLGBmuAcdg/A6I9Lsy61Wwm+gIoaAv2Pd9aWCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-16T04:36:29.886719Z"},"content_sha256":"84e97b760a8a364070287f694d89414ca801989fa7ceb78cd65753c9ea7e82ac","schema_version":"1.0","event_id":"sha256:84e97b760a8a364070287f694d89414ca801989fa7ceb78cd65753c9ea7e82ac"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:SEFHANCVMB2BHWEVQV3FNP7AV3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Finedeep: Mitigating Sparse Activation in Dense LLMs via Multi-Layer Fine-Grained Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cheng Luo, Deyi Xiong, Di Zhang, Guiguang Ding, Hui Chen, Jungong Han, Kun Gai, Leiyu Pan, Minxuan Lv, Xiangwen Zhang, Yizhe Xiong, Zhenpeng Su, Zijia Lin","submitted_at":"2025-02-18T15:09:58Z","abstract_excerpt":"Large language models have demonstrated exceptional performance across a wide range of tasks. However, dense models usually suffer from sparse activation, where many activation values tend towards zero (i.e., being inactivated). We argue that this could restrict the efficient exploration of model representation space. To mitigate this issue, we propose Finedeep, a deep-layered fine-grained expert architecture for dense models. Our framework partitions the feed-forward neural network layers of traditional dense models into small experts, arranges them across multiple sub-layers. A novel routing"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.12928","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/2502.12928/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":"2026-07-05T10:16:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AvAFlpo1jnni9BOuqCiUMovRo9AeLlld4IFUBZocJrmL0EgLyx3URUMG1RmDVw9end7K1iOijKXUJAoy4QLgCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-16T04:36:29.887094Z"},"content_sha256":"e3af2bbdb773e654af1014628f341b253ee034fa7178725507aefd826872333a","schema_version":"1.0","event_id":"sha256:e3af2bbdb773e654af1014628f341b253ee034fa7178725507aefd826872333a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SEFHANCVMB2BHWEVQV3FNP7AV3/bundle.json","state_url":"https://pith.science/pith/SEFHANCVMB2BHWEVQV3FNP7AV3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SEFHANCVMB2BHWEVQV3FNP7AV3/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":"2026-07-16T04:36:29Z","links":{"resolver":"https://pith.science/pith/SEFHANCVMB2BHWEVQV3FNP7AV3","bundle":"https://pith.science/pith/SEFHANCVMB2BHWEVQV3FNP7AV3/bundle.json","state":"https://pith.science/pith/SEFHANCVMB2BHWEVQV3FNP7AV3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SEFHANCVMB2BHWEVQV3FNP7AV3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SEFHANCVMB2BHWEVQV3FNP7AV3","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":"b7311791acab1f653309151e94bad39a763b54bf70168e416426cbd02ecb3b27","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-18T15:09:58Z","title_canon_sha256":"120348668b686c6d33681c6c95886389f78f068b817fc7a745ee76e48876620f"},"schema_version":"1.0","source":{"id":"2502.12928","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.12928","created_at":"2026-07-05T10:16:26Z"},{"alias_kind":"arxiv_version","alias_value":"2502.12928v1","created_at":"2026-07-05T10:16:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.12928","created_at":"2026-07-05T10:16:26Z"},{"alias_kind":"pith_short_12","alias_value":"SEFHANCVMB2B","created_at":"2026-07-05T10:16:26Z"},{"alias_kind":"pith_short_16","alias_value":"SEFHANCVMB2BHWEV","created_at":"2026-07-05T10:16:26Z"},{"alias_kind":"pith_short_8","alias_value":"SEFHANCV","created_at":"2026-07-05T10:16:26Z"}],"graph_snapshots":[{"event_id":"sha256:e3af2bbdb773e654af1014628f341b253ee034fa7178725507aefd826872333a","target":"graph","created_at":"2026-07-05T10:16:26Z","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/2502.12928/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models have demonstrated exceptional performance across a wide range of tasks. However, dense models usually suffer from sparse activation, where many activation values tend towards zero (i.e., being inactivated). We argue that this could restrict the efficient exploration of model representation space. To mitigate this issue, we propose Finedeep, a deep-layered fine-grained expert architecture for dense models. Our framework partitions the feed-forward neural network layers of traditional dense models into small experts, arranges them across multiple sub-layers. A novel routing","authors_text":"Cheng Luo, Deyi Xiong, Di Zhang, Guiguang Ding, Hui Chen, Jungong Han, Kun Gai, Leiyu Pan, Minxuan Lv, Xiangwen Zhang, Yizhe Xiong, Zhenpeng Su, Zijia Lin","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-18T15:09:58Z","title":"Finedeep: Mitigating Sparse Activation in Dense LLMs via Multi-Layer Fine-Grained Experts"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.12928","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":"sha256:84e97b760a8a364070287f694d89414ca801989fa7ceb78cd65753c9ea7e82ac","target":"record","created_at":"2026-07-05T10:16:26Z","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":"b7311791acab1f653309151e94bad39a763b54bf70168e416426cbd02ecb3b27","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-18T15:09:58Z","title_canon_sha256":"120348668b686c6d33681c6c95886389f78f068b817fc7a745ee76e48876620f"},"schema_version":"1.0","source":{"id":"2502.12928","kind":"arxiv","version":1}},"canonical_sha256":"910a703455607413d895857656bfe0aed4bab01a54fa00bdf5c931172e0b350c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"910a703455607413d895857656bfe0aed4bab01a54fa00bdf5c931172e0b350c","first_computed_at":"2026-07-05T10:16:26.450226Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:16:26.450226Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VwQt2XeQB9v8JGQRmgiuywWebU1P2oMhy6NPSU2ba6jsxaVPv8lZyfh/sZx+8QRYM1tM4xWN8fR+CiIK5bN0Cw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:16:26.450656Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.12928","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:84e97b760a8a364070287f694d89414ca801989fa7ceb78cd65753c9ea7e82ac","sha256:e3af2bbdb773e654af1014628f341b253ee034fa7178725507aefd826872333a"],"state_sha256":"a6bbdceb1a80e617ad34ef59b5ddf4be143162b2590ac61082cc2337ced6cf8e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dr/WdFsrbUFXnhEiADCmnTzDpvJHUAC7J85HXH3F8fpImc5AVahaTjg93rIQ3M4RZjQqbWGM9gT1556FLd53Ag==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-16T04:36:29.889363Z","bundle_sha256":"2a8c820de66eb2e6b2c20649322e1e8a7da65e8a9de86b40eb951fb840e0c75e"}}