{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:V26QLI5DQF67ZCNMHEEKUMENK3","short_pith_number":"pith:V26QLI5D","canonical_record":{"source":{"id":"2411.04996","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-07T18:59:06Z","cross_cats_sorted":[],"title_canon_sha256":"e079b34f86192301cb847ef4d01e5b1ef00de5dd55cddae41fba26115f8eb32b","abstract_canon_sha256":"63c23ae7820b209e313d61b2ebe13cee78cc1b6fa4f547e41b68ad0f8b5b475d"},"schema_version":"1.0"},"canonical_sha256":"aebd05a3a3817dfc89ac3908aa308d56faf3c6c3fb11da32b7b9c00b60dc10ff","source":{"kind":"arxiv","id":"2411.04996","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.04996","created_at":"2026-05-18T02:43:43Z"},{"alias_kind":"arxiv_version","alias_value":"2411.04996v2","created_at":"2026-05-18T02:43:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.04996","created_at":"2026-05-18T02:43:43Z"},{"alias_kind":"pith_short_12","alias_value":"V26QLI5DQF67","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"V26QLI5DQF67ZCNM","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"V26QLI5D","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:V26QLI5DQF67ZCNMHEEKUMENK3","target":"record","payload":{"canonical_record":{"source":{"id":"2411.04996","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-07T18:59:06Z","cross_cats_sorted":[],"title_canon_sha256":"e079b34f86192301cb847ef4d01e5b1ef00de5dd55cddae41fba26115f8eb32b","abstract_canon_sha256":"63c23ae7820b209e313d61b2ebe13cee78cc1b6fa4f547e41b68ad0f8b5b475d"},"schema_version":"1.0"},"canonical_sha256":"aebd05a3a3817dfc89ac3908aa308d56faf3c6c3fb11da32b7b9c00b60dc10ff","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:43:43.371689Z","signature_b64":"hjGQXdKWLhAAl9necOepof4Ra/4FfiQU02YlgUP9A77XSqPfXU+HmbwjRKVb1gfHLP3UzYmxJEKLaljPg0CICQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aebd05a3a3817dfc89ac3908aa308d56faf3c6c3fb11da32b7b9c00b60dc10ff","last_reissued_at":"2026-05-18T02:43:43.370885Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:43:43.370885Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.04996","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T02:43:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FblNzgakMqqMOf8NJ97ocQ+UtASlsol3KPUCwyeqWUXY7Qk61M/EMVxpbibi2Y9vSqWNh3q/EgCA2mCCpIw3Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-18T19:45:14.270004Z"},"content_sha256":"d6fddabe7e89cd29422f226c481c5f7fb704b1d1ed2e96bc627f062d849066bf","schema_version":"1.0","event_id":"sha256:d6fddabe7e89cd29422f226c481c5f7fb704b1d1ed2e96bc627f062d849066bf"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:V26QLI5DQF67ZCNMHEEKUMENK3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chunting Zhou, Gargi Ghosh, Liang Luo, Lili Yu, Luke Zettlemoyer, Mike Lewis, Ning Dong, Srinivasan Iyer, Weixin Liang, Wen-tau Yih, Xi Victoria Lin","submitted_at":"2024-11-07T18:59:06Z","abstract_excerpt":"The development of large language models (LLMs) has expanded to multi-modal systems capable of processing text, images, and speech within a unified framework. Training these models demands significantly larger datasets and computational resources compared to text-only LLMs. To address the scaling challenges, we introduce Mixture-of-Transformers (MoT), a sparse multi-modal transformer architecture that significantly reduces pretraining computational costs. MoT decouples non-embedding parameters of the model by modality -- including feed-forward networks, attention matrices, and layer normalizat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.04996","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"2026-05-18T02:43:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"v4vzeLDHV0AjQNiXfTIpENi1alPXbkBwtfVGDbY3uTuQkvCAewzB+haPwSItun1S+sOvGfhjqM0NMBnp7CdlDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-18T19:45:14.270424Z"},"content_sha256":"d1d4b8bc1015a550397b8b2a6f5c27abf8a5f74a3d4ccdf255453209aed41318","schema_version":"1.0","event_id":"sha256:d1d4b8bc1015a550397b8b2a6f5c27abf8a5f74a3d4ccdf255453209aed41318"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/V26QLI5DQF67ZCNMHEEKUMENK3/bundle.json","state_url":"https://pith.science/pith/V26QLI5DQF67ZCNMHEEKUMENK3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/V26QLI5DQF67ZCNMHEEKUMENK3/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":"2026-05-18T19:45:14Z","links":{"resolver":"https://pith.science/pith/V26QLI5DQF67ZCNMHEEKUMENK3","bundle":"https://pith.science/pith/V26QLI5DQF67ZCNMHEEKUMENK3/bundle.json","state":"https://pith.science/pith/V26QLI5DQF67ZCNMHEEKUMENK3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/V26QLI5DQF67ZCNMHEEKUMENK3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:V26QLI5DQF67ZCNMHEEKUMENK3","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":"63c23ae7820b209e313d61b2ebe13cee78cc1b6fa4f547e41b68ad0f8b5b475d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-07T18:59:06Z","title_canon_sha256":"e079b34f86192301cb847ef4d01e5b1ef00de5dd55cddae41fba26115f8eb32b"},"schema_version":"1.0","source":{"id":"2411.04996","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.04996","created_at":"2026-05-18T02:43:43Z"},{"alias_kind":"arxiv_version","alias_value":"2411.04996v2","created_at":"2026-05-18T02:43:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.04996","created_at":"2026-05-18T02:43:43Z"},{"alias_kind":"pith_short_12","alias_value":"V26QLI5DQF67","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"V26QLI5DQF67ZCNM","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"V26QLI5D","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:d1d4b8bc1015a550397b8b2a6f5c27abf8a5f74a3d4ccdf255453209aed41318","target":"graph","created_at":"2026-05-18T02:43:43Z","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"},"paper":{"abstract_excerpt":"The development of large language models (LLMs) has expanded to multi-modal systems capable of processing text, images, and speech within a unified framework. Training these models demands significantly larger datasets and computational resources compared to text-only LLMs. To address the scaling challenges, we introduce Mixture-of-Transformers (MoT), a sparse multi-modal transformer architecture that significantly reduces pretraining computational costs. MoT decouples non-embedding parameters of the model by modality -- including feed-forward networks, attention matrices, and layer normalizat","authors_text":"Chunting Zhou, Gargi Ghosh, Liang Luo, Lili Yu, Luke Zettlemoyer, Mike Lewis, Ning Dong, Srinivasan Iyer, Weixin Liang, Wen-tau Yih, Xi Victoria Lin","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-07T18:59:06Z","title":"Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.04996","kind":"arxiv","version":2},"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:d6fddabe7e89cd29422f226c481c5f7fb704b1d1ed2e96bc627f062d849066bf","target":"record","created_at":"2026-05-18T02:43:43Z","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":"63c23ae7820b209e313d61b2ebe13cee78cc1b6fa4f547e41b68ad0f8b5b475d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-07T18:59:06Z","title_canon_sha256":"e079b34f86192301cb847ef4d01e5b1ef00de5dd55cddae41fba26115f8eb32b"},"schema_version":"1.0","source":{"id":"2411.04996","kind":"arxiv","version":2}},"canonical_sha256":"aebd05a3a3817dfc89ac3908aa308d56faf3c6c3fb11da32b7b9c00b60dc10ff","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"aebd05a3a3817dfc89ac3908aa308d56faf3c6c3fb11da32b7b9c00b60dc10ff","first_computed_at":"2026-05-18T02:43:43.370885Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T02:43:43.370885Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hjGQXdKWLhAAl9necOepof4Ra/4FfiQU02YlgUP9A77XSqPfXU+HmbwjRKVb1gfHLP3UzYmxJEKLaljPg0CICQ==","signature_status":"signed_v1","signed_at":"2026-05-18T02:43:43.371689Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.04996","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d6fddabe7e89cd29422f226c481c5f7fb704b1d1ed2e96bc627f062d849066bf","sha256:d1d4b8bc1015a550397b8b2a6f5c27abf8a5f74a3d4ccdf255453209aed41318"],"state_sha256":"188c10d0e52f255bce44a4ef850fb6e4ddc57a7f138af7cce6fd9a2b7389e7a5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HfZRYiCpZbMhOVND/w+/ytqtbtR+oITSrcvfpnsrfnloAGwP5Km7yfbr5ajfpLHB8jg6RwwUMlSkifyiZMWaCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-18T19:45:14.271966Z","bundle_sha256":"42789ad496c27afd4c8485a23c4f346ce2bdb10a22a51a1aff4ccd7ab2728a26"}}