{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:K5HCITCHLMEVYDVTBFYGESYYGB","short_pith_number":"pith:K5HCITCH","schema_version":"1.0","canonical_sha256":"574e244c475b095c0eb30970624b18304bc63a6c8dace1ef459336b319d3e0c1","source":{"kind":"arxiv","id":"2606.28516","version":1},"attestation_state":"computed","paper":{"title":"CLEAR-MoE: Shared-Basis Expert Extraction from Frozen Vision Transformers via Calibration-Driven Layer Selection","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.CV","authors_text":"Humaira Ayesha, Junaid Ahmed Sifat, Md Irtiza Hossain","submitted_at":"2026-06-26T18:12:03Z","abstract_excerpt":"We present CLEAR-MoE, a four-phase post-training pipeline that converts a frozen pretrained Vision Transformer (ViT) into a sparse Mixture-of-Experts (MoE) model without updating backbone weights. The pipeline (i) scores feed-forward network (FFN) layers by sparsity, clusterability, and output sensitivity; (ii) decomposes selected layers into a shared low-rank SVD basis and per-cluster residual experts using k-means clustering; (iii) trains lightweight routers supervised by cluster labels; and (iv) dispatches tokens through pluggable CUDA backends. On Imagenette with DeiT-Small, CLEAR-MoE reta"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.28516","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-26T18:12:03Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"c3ca9490863e500959c31acc1a5a4bd76dc5f16353cc21707f52e07253ff5e3a","abstract_canon_sha256":"df14a6759e525d6f4e64d65338b19297c483db4cd354a2f47d58f1a00284da79"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T00:15:16.639558Z","signature_b64":"GeyQnXnHLeUgqKcoc22S17YcDF1RRGqULsQQkjkoroV4JqYaprAz2cB5MW9hvAEuGnzGvZPsXtZQ8ElZc4c6BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"574e244c475b095c0eb30970624b18304bc63a6c8dace1ef459336b319d3e0c1","last_reissued_at":"2026-06-30T00:15:16.639070Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T00:15:16.639070Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CLEAR-MoE: Shared-Basis Expert Extraction from Frozen Vision Transformers via Calibration-Driven Layer Selection","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.CV","authors_text":"Humaira Ayesha, Junaid Ahmed Sifat, Md Irtiza Hossain","submitted_at":"2026-06-26T18:12:03Z","abstract_excerpt":"We present CLEAR-MoE, a four-phase post-training pipeline that converts a frozen pretrained Vision Transformer (ViT) into a sparse Mixture-of-Experts (MoE) model without updating backbone weights. The pipeline (i) scores feed-forward network (FFN) layers by sparsity, clusterability, and output sensitivity; (ii) decomposes selected layers into a shared low-rank SVD basis and per-cluster residual experts using k-means clustering; (iii) trains lightweight routers supervised by cluster labels; and (iv) dispatches tokens through pluggable CUDA backends. On Imagenette with DeiT-Small, CLEAR-MoE reta"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.28516","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/2606.28516/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.28516","created_at":"2026-06-30T00:15:16.639122+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.28516v1","created_at":"2026-06-30T00:15:16.639122+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.28516","created_at":"2026-06-30T00:15:16.639122+00:00"},{"alias_kind":"pith_short_12","alias_value":"K5HCITCHLMEV","created_at":"2026-06-30T00:15:16.639122+00:00"},{"alias_kind":"pith_short_16","alias_value":"K5HCITCHLMEVYDVT","created_at":"2026-06-30T00:15:16.639122+00:00"},{"alias_kind":"pith_short_8","alias_value":"K5HCITCH","created_at":"2026-06-30T00:15:16.639122+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/K5HCITCHLMEVYDVTBFYGESYYGB","json":"https://pith.science/pith/K5HCITCHLMEVYDVTBFYGESYYGB.json","graph_json":"https://pith.science/api/pith-number/K5HCITCHLMEVYDVTBFYGESYYGB/graph.json","events_json":"https://pith.science/api/pith-number/K5HCITCHLMEVYDVTBFYGESYYGB/events.json","paper":"https://pith.science/paper/K5HCITCH"},"agent_actions":{"view_html":"https://pith.science/pith/K5HCITCHLMEVYDVTBFYGESYYGB","download_json":"https://pith.science/pith/K5HCITCHLMEVYDVTBFYGESYYGB.json","view_paper":"https://pith.science/paper/K5HCITCH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.28516&json=true","fetch_graph":"https://pith.science/api/pith-number/K5HCITCHLMEVYDVTBFYGESYYGB/graph.json","fetch_events":"https://pith.science/api/pith-number/K5HCITCHLMEVYDVTBFYGESYYGB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/K5HCITCHLMEVYDVTBFYGESYYGB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/K5HCITCHLMEVYDVTBFYGESYYGB/action/storage_attestation","attest_author":"https://pith.science/pith/K5HCITCHLMEVYDVTBFYGESYYGB/action/author_attestation","sign_citation":"https://pith.science/pith/K5HCITCHLMEVYDVTBFYGESYYGB/action/citation_signature","submit_replication":"https://pith.science/pith/K5HCITCHLMEVYDVTBFYGESYYGB/action/replication_record"}},"created_at":"2026-06-30T00:15:16.639122+00:00","updated_at":"2026-06-30T00:15:16.639122+00:00"}