{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:QCCNOBRYWTP5W3B3RANMJBJY2B","short_pith_number":"pith:QCCNOBRY","canonical_record":{"source":{"id":"2310.00535","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-01T01:21:35Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"8897a6397e61f48f2dc8451aeb52a90b2312d2b323d847990b414b995939f6fb","abstract_canon_sha256":"81025011339006ecda0e5d8534cf344aa02cb41faecd5d526d964704dea76df6"},"schema_version":"1.0"},"canonical_sha256":"8084d70638b4dfdb6c3b881ac48538d07b2ae5d6f86463aaeb484aecf339ec99","source":{"kind":"arxiv","id":"2310.00535","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.00535","created_at":"2026-07-05T07:56:17Z"},{"alias_kind":"arxiv_version","alias_value":"2310.00535v3","created_at":"2026-07-05T07:56:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.00535","created_at":"2026-07-05T07:56:17Z"},{"alias_kind":"pith_short_12","alias_value":"QCCNOBRYWTP5","created_at":"2026-07-05T07:56:17Z"},{"alias_kind":"pith_short_16","alias_value":"QCCNOBRYWTP5W3B3","created_at":"2026-07-05T07:56:17Z"},{"alias_kind":"pith_short_8","alias_value":"QCCNOBRY","created_at":"2026-07-05T07:56:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:QCCNOBRYWTP5W3B3RANMJBJY2B","target":"record","payload":{"canonical_record":{"source":{"id":"2310.00535","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-01T01:21:35Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"8897a6397e61f48f2dc8451aeb52a90b2312d2b323d847990b414b995939f6fb","abstract_canon_sha256":"81025011339006ecda0e5d8534cf344aa02cb41faecd5d526d964704dea76df6"},"schema_version":"1.0"},"canonical_sha256":"8084d70638b4dfdb6c3b881ac48538d07b2ae5d6f86463aaeb484aecf339ec99","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:56:17.249595Z","signature_b64":"LNDIeXm2mL1c6WNVQq3mWEhbpDc1WCf5BHYP4gQD8xMwP59h0pylRCIE0BRRUxkNFGQb8FXqBl9BXc9xZLfUBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8084d70638b4dfdb6c3b881ac48538d07b2ae5d6f86463aaeb484aecf339ec99","last_reissued_at":"2026-07-05T07:56:17.249141Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:56:17.249141Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2310.00535","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:56:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Uy6qjwvLLna6WGHhIXvzI9BC3+gVYzmzlbGx70gAGd3K+7E6KaIvOLssICYvxdO3QjCqDt5vigjdQaVSoZoeBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T13:13:57.812662Z"},"content_sha256":"ef87212c632e448d842f83bf01c7f98448b523fb8d16ced51640f264060ec625","schema_version":"1.0","event_id":"sha256:ef87212c632e448d842f83bf01c7f98448b523fb8d16ced51640f264060ec625"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:QCCNOBRYWTP5W3B3RANMJBJY2B","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"JoMA: Demystifying Multilayer Transformers via JOint Dynamics of MLP and Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Beidi Chen, Simon Du, Yiping Wang, Yuandong Tian, Zhenyu Zhang","submitted_at":"2023-10-01T01:21:35Z","abstract_excerpt":"We propose Joint MLP/Attention (JoMA) dynamics, a novel mathematical framework to understand the training procedure of multilayer Transformer architectures. This is achieved by integrating out the self-attention layer in Transformers, producing a modified dynamics of MLP layers only. JoMA removes unrealistic assumptions in previous analysis (e.g., lack of residual connection) and predicts that the attention first becomes sparse (to learn salient tokens), then dense (to learn less salient tokens) in the presence of nonlinear activations, while in the linear case, it is consistent with existing "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.00535","kind":"arxiv","version":3},"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/2310.00535/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":"2026-07-05T07:56:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"roTlcTwQIzJMP91gGg+mSqpsMFfXOFpyYy307XRnBost0OpnZYa3j2eC9Ae/j0YmFpBQeVES9glyf9zsBdNvAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T13:13:57.813737Z"},"content_sha256":"f603cb512eb85e25aebfd43a75654da845c3dbde3e0efcb2a30ee59664b115a7","schema_version":"1.0","event_id":"sha256:f603cb512eb85e25aebfd43a75654da845c3dbde3e0efcb2a30ee59664b115a7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QCCNOBRYWTP5W3B3RANMJBJY2B/bundle.json","state_url":"https://pith.science/pith/QCCNOBRYWTP5W3B3RANMJBJY2B/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QCCNOBRYWTP5W3B3RANMJBJY2B/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":"2026-08-08T13:13:57Z","links":{"resolver":"https://pith.science/pith/QCCNOBRYWTP5W3B3RANMJBJY2B","bundle":"https://pith.science/pith/QCCNOBRYWTP5W3B3RANMJBJY2B/bundle.json","state":"https://pith.science/pith/QCCNOBRYWTP5W3B3RANMJBJY2B/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QCCNOBRYWTP5W3B3RANMJBJY2B/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:QCCNOBRYWTP5W3B3RANMJBJY2B","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":"81025011339006ecda0e5d8534cf344aa02cb41faecd5d526d964704dea76df6","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-01T01:21:35Z","title_canon_sha256":"8897a6397e61f48f2dc8451aeb52a90b2312d2b323d847990b414b995939f6fb"},"schema_version":"1.0","source":{"id":"2310.00535","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.00535","created_at":"2026-07-05T07:56:17Z"},{"alias_kind":"arxiv_version","alias_value":"2310.00535v3","created_at":"2026-07-05T07:56:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.00535","created_at":"2026-07-05T07:56:17Z"},{"alias_kind":"pith_short_12","alias_value":"QCCNOBRYWTP5","created_at":"2026-07-05T07:56:17Z"},{"alias_kind":"pith_short_16","alias_value":"QCCNOBRYWTP5W3B3","created_at":"2026-07-05T07:56:17Z"},{"alias_kind":"pith_short_8","alias_value":"QCCNOBRY","created_at":"2026-07-05T07:56:17Z"}],"graph_snapshots":[{"event_id":"sha256:f603cb512eb85e25aebfd43a75654da845c3dbde3e0efcb2a30ee59664b115a7","target":"graph","created_at":"2026-07-05T07:56:17Z","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/2310.00535/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We propose Joint MLP/Attention (JoMA) dynamics, a novel mathematical framework to understand the training procedure of multilayer Transformer architectures. This is achieved by integrating out the self-attention layer in Transformers, producing a modified dynamics of MLP layers only. JoMA removes unrealistic assumptions in previous analysis (e.g., lack of residual connection) and predicts that the attention first becomes sparse (to learn salient tokens), then dense (to learn less salient tokens) in the presence of nonlinear activations, while in the linear case, it is consistent with existing ","authors_text":"Beidi Chen, Simon Du, Yiping Wang, Yuandong Tian, Zhenyu Zhang","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-01T01:21:35Z","title":"JoMA: Demystifying Multilayer Transformers via JOint Dynamics of MLP and Attention"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.00535","kind":"arxiv","version":3},"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:ef87212c632e448d842f83bf01c7f98448b523fb8d16ced51640f264060ec625","target":"record","created_at":"2026-07-05T07:56:17Z","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":"81025011339006ecda0e5d8534cf344aa02cb41faecd5d526d964704dea76df6","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-01T01:21:35Z","title_canon_sha256":"8897a6397e61f48f2dc8451aeb52a90b2312d2b323d847990b414b995939f6fb"},"schema_version":"1.0","source":{"id":"2310.00535","kind":"arxiv","version":3}},"canonical_sha256":"8084d70638b4dfdb6c3b881ac48538d07b2ae5d6f86463aaeb484aecf339ec99","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8084d70638b4dfdb6c3b881ac48538d07b2ae5d6f86463aaeb484aecf339ec99","first_computed_at":"2026-07-05T07:56:17.249141Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:56:17.249141Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LNDIeXm2mL1c6WNVQq3mWEhbpDc1WCf5BHYP4gQD8xMwP59h0pylRCIE0BRRUxkNFGQb8FXqBl9BXc9xZLfUBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:56:17.249595Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.00535","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ef87212c632e448d842f83bf01c7f98448b523fb8d16ced51640f264060ec625","sha256:f603cb512eb85e25aebfd43a75654da845c3dbde3e0efcb2a30ee59664b115a7"],"state_sha256":"1c52892b79b58146faa55ff70227493c0058bb24949f8dfae42d3a7d0d9ec0c4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"c9wdOuILozn30PESr3aoRcuEzNxSqo75sqm5c88oNbpY8wwCJKjX/L6Pmzf1FNYDMi4yMGk3SzzfBLGYp+tvBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T13:13:57.820005Z","bundle_sha256":"dd700bf32a503ccdd9763aa85b5ee1017ea4e643a41b317fe6cf66c87c6ee3d9"}}