{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:BTQTFSUNVKZYSDYGULPUCY3LQV","short_pith_number":"pith:BTQTFSUN","canonical_record":{"source":{"id":"2310.16795","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-25T17:24:53Z","cross_cats_sorted":[],"title_canon_sha256":"b9942db8ef82308a96588a88f5c6564f44601fff28f4823c668aa01fb6823f7f","abstract_canon_sha256":"84eba9a38751f4677359c46acf399dc9decb584414af225cacd0d3316506ab92"},"schema_version":"1.0"},"canonical_sha256":"0ce132ca8daab3890f06a2df41636b85789c73bf350f2f7c5fceeb1b494bcedc","source":{"kind":"arxiv","id":"2310.16795","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.16795","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"arxiv_version","alias_value":"2310.16795v1","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.16795","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"pith_short_12","alias_value":"BTQTFSUNVKZY","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"pith_short_16","alias_value":"BTQTFSUNVKZYSDYG","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"pith_short_8","alias_value":"BTQTFSUN","created_at":"2026-07-05T07:05:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:BTQTFSUNVKZYSDYGULPUCY3LQV","target":"record","payload":{"canonical_record":{"source":{"id":"2310.16795","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-25T17:24:53Z","cross_cats_sorted":[],"title_canon_sha256":"b9942db8ef82308a96588a88f5c6564f44601fff28f4823c668aa01fb6823f7f","abstract_canon_sha256":"84eba9a38751f4677359c46acf399dc9decb584414af225cacd0d3316506ab92"},"schema_version":"1.0"},"canonical_sha256":"0ce132ca8daab3890f06a2df41636b85789c73bf350f2f7c5fceeb1b494bcedc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:05:02.993437Z","signature_b64":"rkQ6YE47yYZvyFmMqzNAWhrxqB/gq69FJg7PWSrAsxnVOjLCTOCsJXfBzkxG97GGjXauN+d+9WgQja64Nk1LAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0ce132ca8daab3890f06a2df41636b85789c73bf350f2f7c5fceeb1b494bcedc","last_reissued_at":"2026-07-05T07:05:02.992877Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:05:02.992877Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2310.16795","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":"2026-07-05T07:05:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wa5PeXQgOuOnXSSAIgIj+yb1loyrZEYfKEFLaKqiqKKhbgAk5iWmotw03HclsuhnDsdJudCXlfFdOSQ4arUyAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T14:28:27.274668Z"},"content_sha256":"d69ff2f14776ec7b06fb0f45f2409d5b03a36bb764f5143c75752df015de6efe","schema_version":"1.0","event_id":"sha256:d69ff2f14776ec7b06fb0f45f2409d5b03a36bb764f5143c75752df015de6efe"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:BTQTFSUNVKZYSDYGULPUCY3LQV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dan Alistarh, Elias Frantar","submitted_at":"2023-10-25T17:24:53Z","abstract_excerpt":"Mixture-of-Experts (MoE) architectures offer a general solution to the high inference costs of large language models (LLMs) via sparse routing, bringing faster and more accurate models, at the cost of massive parameter counts. For example, the SwitchTransformer-c2048 model has 1.6 trillion parameters, requiring 3.2TB of accelerator memory to run efficiently, which makes practical deployment challenging and expensive. In this paper, we present a solution to this memory problem, in form of a new compression and execution framework called QMoE. Specifically, QMoE consists of a scalable algorithm "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.16795","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/2310.16795/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":"2026-07-05T07:05:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u8z6idiaLXAibRHdmDfFH0uWjQM4COBaYD8YrT7QSTGnPw2dUnHyVbscaHZHHR9fHPNoLC5NfHSBI1uxHeGpCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T14:28:27.275348Z"},"content_sha256":"ff4cce63251a23493fa11cd4599dc1bc79c12ccf017dfa0bd6c365f218e39cdc","schema_version":"1.0","event_id":"sha256:ff4cce63251a23493fa11cd4599dc1bc79c12ccf017dfa0bd6c365f218e39cdc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BTQTFSUNVKZYSDYGULPUCY3LQV/bundle.json","state_url":"https://pith.science/pith/BTQTFSUNVKZYSDYGULPUCY3LQV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BTQTFSUNVKZYSDYGULPUCY3LQV/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":"2026-08-20T14:28:27Z","links":{"resolver":"https://pith.science/pith/BTQTFSUNVKZYSDYGULPUCY3LQV","bundle":"https://pith.science/pith/BTQTFSUNVKZYSDYGULPUCY3LQV/bundle.json","state":"https://pith.science/pith/BTQTFSUNVKZYSDYGULPUCY3LQV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BTQTFSUNVKZYSDYGULPUCY3LQV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:BTQTFSUNVKZYSDYGULPUCY3LQV","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":"84eba9a38751f4677359c46acf399dc9decb584414af225cacd0d3316506ab92","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-25T17:24:53Z","title_canon_sha256":"b9942db8ef82308a96588a88f5c6564f44601fff28f4823c668aa01fb6823f7f"},"schema_version":"1.0","source":{"id":"2310.16795","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.16795","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"arxiv_version","alias_value":"2310.16795v1","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.16795","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"pith_short_12","alias_value":"BTQTFSUNVKZY","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"pith_short_16","alias_value":"BTQTFSUNVKZYSDYG","created_at":"2026-07-05T07:05:02Z"},{"alias_kind":"pith_short_8","alias_value":"BTQTFSUN","created_at":"2026-07-05T07:05:02Z"}],"graph_snapshots":[{"event_id":"sha256:ff4cce63251a23493fa11cd4599dc1bc79c12ccf017dfa0bd6c365f218e39cdc","target":"graph","created_at":"2026-07-05T07:05:02Z","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/2310.16795/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Mixture-of-Experts (MoE) architectures offer a general solution to the high inference costs of large language models (LLMs) via sparse routing, bringing faster and more accurate models, at the cost of massive parameter counts. For example, the SwitchTransformer-c2048 model has 1.6 trillion parameters, requiring 3.2TB of accelerator memory to run efficiently, which makes practical deployment challenging and expensive. In this paper, we present a solution to this memory problem, in form of a new compression and execution framework called QMoE. Specifically, QMoE consists of a scalable algorithm ","authors_text":"Dan Alistarh, Elias Frantar","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-25T17:24:53Z","title":"QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.16795","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:d69ff2f14776ec7b06fb0f45f2409d5b03a36bb764f5143c75752df015de6efe","target":"record","created_at":"2026-07-05T07:05:02Z","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":"84eba9a38751f4677359c46acf399dc9decb584414af225cacd0d3316506ab92","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-25T17:24:53Z","title_canon_sha256":"b9942db8ef82308a96588a88f5c6564f44601fff28f4823c668aa01fb6823f7f"},"schema_version":"1.0","source":{"id":"2310.16795","kind":"arxiv","version":1}},"canonical_sha256":"0ce132ca8daab3890f06a2df41636b85789c73bf350f2f7c5fceeb1b494bcedc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0ce132ca8daab3890f06a2df41636b85789c73bf350f2f7c5fceeb1b494bcedc","first_computed_at":"2026-07-05T07:05:02.992877Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:05:02.992877Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rkQ6YE47yYZvyFmMqzNAWhrxqB/gq69FJg7PWSrAsxnVOjLCTOCsJXfBzkxG97GGjXauN+d+9WgQja64Nk1LAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:05:02.993437Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.16795","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d69ff2f14776ec7b06fb0f45f2409d5b03a36bb764f5143c75752df015de6efe","sha256:ff4cce63251a23493fa11cd4599dc1bc79c12ccf017dfa0bd6c365f218e39cdc"],"state_sha256":"6f7c1aad26a94038d24275fd96786d298bda69676e24d5c1a7dee1d5c52a7f58"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Tp/r6qs1uIjr9BHT1wd2oGFw3dJU1YV5jd73mbEu5V9Y17kGLWpEw6YvT6uddHwFPO3Ra2ZG5UvJdOHTls6lDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T14:28:27.281504Z","bundle_sha256":"d7d3abd041f0abdf22cbefbabd359b276fb887dda239f3f093474343c757a310"}}