{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TW4ZVCRF4XA5DDBG3DXYAZZYA3","short_pith_number":"pith:TW4ZVCRF","schema_version":"1.0","canonical_sha256":"9db99a8a25e5c1d18c26d8ef80673806f11023c6399251bdeeaac18bdc27f7e4","source":{"kind":"arxiv","id":"2605.17598","version":1},"attestation_state":"computed","paper":{"title":"Mixture of Experts for Low-Resource LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dan Revital, Noam Kayzer, Ori Bar Joseph, Sarel Weinberger, Smadar Arvatz","submitted_at":"2026-05-17T18:50:50Z","abstract_excerpt":"Mixture-of-Experts (MoE) architectures enable efficient model scaling, yet expert routing behavior across underrepresented languages remains poorly understood. We analyze routing dynamics in two architecturally distinct MoE models -- a pure Transformer (Qwen3-30B-A3B) and a hybrid Mamba-Transformer (Nemotron-3-Nano-30B-A3B) -- using Hebrew as a morphologically rich, low-resource testbed. Both pre-trained models exhibit \\emph{deep-layer routing collapse}: usage entropy drops sharply in final layers and tokens concentrate on a narrow expert subset, a pattern largely absent for English. Continual"},"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":"2605.17598","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-17T18:50:50Z","cross_cats_sorted":[],"title_canon_sha256":"271664b7b62d470f27b0876a091cf036e999321b9449a8503d88e463d76dedb2","abstract_canon_sha256":"46b6860b96247755a0e4647de2923971851fba29aca9b41f9607c1dea1db7682"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:47.877731Z","signature_b64":"d5lkyzqEv6/nhjH6bLSBxejMdttWfPjIIsrvTmMO5N6W8YrG3Yi7bK8sQu2F94yrJypn/x94s4qyMlyGgtxCBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9db99a8a25e5c1d18c26d8ef80673806f11023c6399251bdeeaac18bdc27f7e4","last_reissued_at":"2026-05-20T00:04:47.876922Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:47.876922Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mixture of Experts for Low-Resource LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dan Revital, Noam Kayzer, Ori Bar Joseph, Sarel Weinberger, Smadar Arvatz","submitted_at":"2026-05-17T18:50:50Z","abstract_excerpt":"Mixture-of-Experts (MoE) architectures enable efficient model scaling, yet expert routing behavior across underrepresented languages remains poorly understood. We analyze routing dynamics in two architecturally distinct MoE models -- a pure Transformer (Qwen3-30B-A3B) and a hybrid Mamba-Transformer (Nemotron-3-Nano-30B-A3B) -- using Hebrew as a morphologically rich, low-resource testbed. Both pre-trained models exhibit \\emph{deep-layer routing collapse}: usage entropy drops sharply in final layers and tokens concentrate on a narrow expert subset, a pattern largely absent for English. Continual"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.17598","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/2605.17598/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T21:33:23.577038Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T21:21:57.506697Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"0617b1901077e47e2e23df02845a6b6aff49283c0dce4899960ea97b3059ad5e"},"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":"2605.17598","created_at":"2026-05-20T00:04:47.877056+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.17598v1","created_at":"2026-05-20T00:04:47.877056+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17598","created_at":"2026-05-20T00:04:47.877056+00:00"},{"alias_kind":"pith_short_12","alias_value":"TW4ZVCRF4XA5","created_at":"2026-05-20T00:04:47.877056+00:00"},{"alias_kind":"pith_short_16","alias_value":"TW4ZVCRF4XA5DDBG","created_at":"2026-05-20T00:04:47.877056+00:00"},{"alias_kind":"pith_short_8","alias_value":"TW4ZVCRF","created_at":"2026-05-20T00:04:47.877056+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/TW4ZVCRF4XA5DDBG3DXYAZZYA3","json":"https://pith.science/pith/TW4ZVCRF4XA5DDBG3DXYAZZYA3.json","graph_json":"https://pith.science/api/pith-number/TW4ZVCRF4XA5DDBG3DXYAZZYA3/graph.json","events_json":"https://pith.science/api/pith-number/TW4ZVCRF4XA5DDBG3DXYAZZYA3/events.json","paper":"https://pith.science/paper/TW4ZVCRF"},"agent_actions":{"view_html":"https://pith.science/pith/TW4ZVCRF4XA5DDBG3DXYAZZYA3","download_json":"https://pith.science/pith/TW4ZVCRF4XA5DDBG3DXYAZZYA3.json","view_paper":"https://pith.science/paper/TW4ZVCRF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.17598&json=true","fetch_graph":"https://pith.science/api/pith-number/TW4ZVCRF4XA5DDBG3DXYAZZYA3/graph.json","fetch_events":"https://pith.science/api/pith-number/TW4ZVCRF4XA5DDBG3DXYAZZYA3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TW4ZVCRF4XA5DDBG3DXYAZZYA3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TW4ZVCRF4XA5DDBG3DXYAZZYA3/action/storage_attestation","attest_author":"https://pith.science/pith/TW4ZVCRF4XA5DDBG3DXYAZZYA3/action/author_attestation","sign_citation":"https://pith.science/pith/TW4ZVCRF4XA5DDBG3DXYAZZYA3/action/citation_signature","submit_replication":"https://pith.science/pith/TW4ZVCRF4XA5DDBG3DXYAZZYA3/action/replication_record"}},"created_at":"2026-05-20T00:04:47.877056+00:00","updated_at":"2026-05-20T00:04:47.877056+00:00"}