{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JTECTJH26NZ4MEGOEYJANKK2B7","short_pith_number":"pith:JTECTJH2","schema_version":"1.0","canonical_sha256":"4cc829a4faf373c610ce261206a95a0fe84a62ab9983830f154dc4b13c276344","source":{"kind":"arxiv","id":"2505.18451","version":1},"attestation_state":"computed","paper":{"title":"$\\mu$-MoE: Test-Time Pruning as Micro-Grained Mixture-of-Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jing Liu, Toshiaki Koike-Akino, Ye Wang","submitted_at":"2025-05-24T01:23:02Z","abstract_excerpt":"To tackle the huge computational demand of large foundation models, activation-aware compression techniques without retraining have been introduced. However, since these rely on calibration data, domain shift may arise for unknown downstream tasks. With a computationally efficient calibration, activation-aware pruning can be executed for every prompt adaptively, yet achieving reduced complexity at inference. We formulate it as a mixture of micro-experts, called $\\mu$-MoE. Several experiments demonstrate that $\\mu$-MoE can dynamically adapt to task/prompt-dependent structured sparsity on the fl"},"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":"2505.18451","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-24T01:23:02Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"a36b913c1f8b6a64650d54e4e6763bd6dfc32588c1423efbf0cc2bb5b4558248","abstract_canon_sha256":"30d7c06c76d9b83028e329b770d7a5fe070385950cd40bd3140a0059ea3b4304"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:46.652498Z","signature_b64":"9ldKxSUSHHgzyeR5JL0r9hEyqF9aj5wrXLs29mtGc3j3tTuMofhZNuc4/zo558DwcBMgpvf5RcGN/2OAsYPqCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4cc829a4faf373c610ce261206a95a0fe84a62ab9983830f154dc4b13c276344","last_reissued_at":"2026-07-05T11:08:46.652092Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:46.652092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"$\\mu$-MoE: Test-Time Pruning as Micro-Grained Mixture-of-Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jing Liu, Toshiaki Koike-Akino, Ye Wang","submitted_at":"2025-05-24T01:23:02Z","abstract_excerpt":"To tackle the huge computational demand of large foundation models, activation-aware compression techniques without retraining have been introduced. However, since these rely on calibration data, domain shift may arise for unknown downstream tasks. With a computationally efficient calibration, activation-aware pruning can be executed for every prompt adaptively, yet achieving reduced complexity at inference. We formulate it as a mixture of micro-experts, called $\\mu$-MoE. Several experiments demonstrate that $\\mu$-MoE can dynamically adapt to task/prompt-dependent structured sparsity on the fl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18451","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/2505.18451/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":"2505.18451","created_at":"2026-07-05T11:08:46.652145+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.18451v1","created_at":"2026-07-05T11:08:46.652145+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18451","created_at":"2026-07-05T11:08:46.652145+00:00"},{"alias_kind":"pith_short_12","alias_value":"JTECTJH26NZ4","created_at":"2026-07-05T11:08:46.652145+00:00"},{"alias_kind":"pith_short_16","alias_value":"JTECTJH26NZ4MEGO","created_at":"2026-07-05T11:08:46.652145+00:00"},{"alias_kind":"pith_short_8","alias_value":"JTECTJH2","created_at":"2026-07-05T11:08:46.652145+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08565","citing_title":"EinSort: Sorting is All We Need for Tensorizing LLM","ref_index":42,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JTECTJH26NZ4MEGOEYJANKK2B7","json":"https://pith.science/pith/JTECTJH26NZ4MEGOEYJANKK2B7.json","graph_json":"https://pith.science/api/pith-number/JTECTJH26NZ4MEGOEYJANKK2B7/graph.json","events_json":"https://pith.science/api/pith-number/JTECTJH26NZ4MEGOEYJANKK2B7/events.json","paper":"https://pith.science/paper/JTECTJH2"},"agent_actions":{"view_html":"https://pith.science/pith/JTECTJH26NZ4MEGOEYJANKK2B7","download_json":"https://pith.science/pith/JTECTJH26NZ4MEGOEYJANKK2B7.json","view_paper":"https://pith.science/paper/JTECTJH2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.18451&json=true","fetch_graph":"https://pith.science/api/pith-number/JTECTJH26NZ4MEGOEYJANKK2B7/graph.json","fetch_events":"https://pith.science/api/pith-number/JTECTJH26NZ4MEGOEYJANKK2B7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JTECTJH26NZ4MEGOEYJANKK2B7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JTECTJH26NZ4MEGOEYJANKK2B7/action/storage_attestation","attest_author":"https://pith.science/pith/JTECTJH26NZ4MEGOEYJANKK2B7/action/author_attestation","sign_citation":"https://pith.science/pith/JTECTJH26NZ4MEGOEYJANKK2B7/action/citation_signature","submit_replication":"https://pith.science/pith/JTECTJH26NZ4MEGOEYJANKK2B7/action/replication_record"}},"created_at":"2026-07-05T11:08:46.652145+00:00","updated_at":"2026-07-05T11:08:46.652145+00:00"}