{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VBPHCONBR6XUYGXMQOIWT7R26M","short_pith_number":"pith:VBPHCONB","schema_version":"1.0","canonical_sha256":"a85e7139a18faf4c1aec839169fe3af30abf9c3f05a73a1a4c1564cca20236c9","source":{"kind":"arxiv","id":"2509.02512","version":1},"attestation_state":"computed","paper":{"title":"MoPEQ: Mixture of Mixed Precision Quantized Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jie Ye, Krishna Teja Chitty-Venkata, Murali Emani","submitted_at":"2025-09-02T17:04:59Z","abstract_excerpt":"Large Language and Vision Models using a Mixture-of-Experts (MoE) architecture pose significant challenges for deployment due to their computational and memory demands. Mixed Precision Quantization assigns different precisions to different layers of an LLM/VLM based on layer sensitivity and importance within the model. In this work, we propose a Post Training Quantization algorithm, MoPEQ, that assigns optimal bit width to each expert. Our method balances accuracy and model size by analyzing each expert's sensitivity using Hessian trace approximation instead of relying on the activation freque"},"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":"2509.02512","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-02T17:04:59Z","cross_cats_sorted":[],"title_canon_sha256":"703f591fe75c22fc88704ed9d67c56282db04c95ecc95cf82b8706bb4490c4b7","abstract_canon_sha256":"6585e987056ecc01d78361b5255375ea495f0e8d3c8f0f28bb0006c376445050"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:03:39.888520Z","signature_b64":"HAb1UqlGWAfxBnk8NhBW8abntudmI53+ecy2WqHa9rX2w92f1qRDhzAOdgxDO8X9Gz1p5uTAiws8GTl/ZECJDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a85e7139a18faf4c1aec839169fe3af30abf9c3f05a73a1a4c1564cca20236c9","last_reissued_at":"2026-07-05T12:03:39.888020Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:03:39.888020Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MoPEQ: Mixture of Mixed Precision Quantized Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jie Ye, Krishna Teja Chitty-Venkata, Murali Emani","submitted_at":"2025-09-02T17:04:59Z","abstract_excerpt":"Large Language and Vision Models using a Mixture-of-Experts (MoE) architecture pose significant challenges for deployment due to their computational and memory demands. Mixed Precision Quantization assigns different precisions to different layers of an LLM/VLM based on layer sensitivity and importance within the model. In this work, we propose a Post Training Quantization algorithm, MoPEQ, that assigns optimal bit width to each expert. Our method balances accuracy and model size by analyzing each expert's sensitivity using Hessian trace approximation instead of relying on the activation freque"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.02512","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/2509.02512/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":"2509.02512","created_at":"2026-07-05T12:03:39.888079+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.02512v1","created_at":"2026-07-05T12:03:39.888079+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.02512","created_at":"2026-07-05T12:03:39.888079+00:00"},{"alias_kind":"pith_short_12","alias_value":"VBPHCONBR6XU","created_at":"2026-07-05T12:03:39.888079+00:00"},{"alias_kind":"pith_short_16","alias_value":"VBPHCONBR6XUYGXM","created_at":"2026-07-05T12:03:39.888079+00:00"},{"alias_kind":"pith_short_8","alias_value":"VBPHCONB","created_at":"2026-07-05T12:03:39.888079+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/VBPHCONBR6XUYGXMQOIWT7R26M","json":"https://pith.science/pith/VBPHCONBR6XUYGXMQOIWT7R26M.json","graph_json":"https://pith.science/api/pith-number/VBPHCONBR6XUYGXMQOIWT7R26M/graph.json","events_json":"https://pith.science/api/pith-number/VBPHCONBR6XUYGXMQOIWT7R26M/events.json","paper":"https://pith.science/paper/VBPHCONB"},"agent_actions":{"view_html":"https://pith.science/pith/VBPHCONBR6XUYGXMQOIWT7R26M","download_json":"https://pith.science/pith/VBPHCONBR6XUYGXMQOIWT7R26M.json","view_paper":"https://pith.science/paper/VBPHCONB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.02512&json=true","fetch_graph":"https://pith.science/api/pith-number/VBPHCONBR6XUYGXMQOIWT7R26M/graph.json","fetch_events":"https://pith.science/api/pith-number/VBPHCONBR6XUYGXMQOIWT7R26M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VBPHCONBR6XUYGXMQOIWT7R26M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VBPHCONBR6XUYGXMQOIWT7R26M/action/storage_attestation","attest_author":"https://pith.science/pith/VBPHCONBR6XUYGXMQOIWT7R26M/action/author_attestation","sign_citation":"https://pith.science/pith/VBPHCONBR6XUYGXMQOIWT7R26M/action/citation_signature","submit_replication":"https://pith.science/pith/VBPHCONBR6XUYGXMQOIWT7R26M/action/replication_record"}},"created_at":"2026-07-05T12:03:39.888079+00:00","updated_at":"2026-07-05T12:03:39.888079+00:00"}