{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:WWZUVLJIXCSBYPGXKDDHRSRWBH","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":"ef659e5dc8b9658f73f50417b28e4657b8469675b5572ab66da2d60ca21e416a","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-04T21:44:09Z","title_canon_sha256":"4384b922d2aff2e4f63175de4baad71d627ba7c49863bdcc0a2b07dd37a81e33"},"schema_version":"1.0","source":{"id":"2402.02622","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.02622","created_at":"2026-07-05T07:59:04Z"},{"alias_kind":"arxiv_version","alias_value":"2402.02622v2","created_at":"2026-07-05T07:59:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.02622","created_at":"2026-07-05T07:59:04Z"},{"alias_kind":"pith_short_12","alias_value":"WWZUVLJIXCSB","created_at":"2026-07-05T07:59:04Z"},{"alias_kind":"pith_short_16","alias_value":"WWZUVLJIXCSBYPGX","created_at":"2026-07-05T07:59:04Z"},{"alias_kind":"pith_short_8","alias_value":"WWZUVLJI","created_at":"2026-07-05T07:59:04Z"}],"graph_snapshots":[{"event_id":"sha256:51ae915416ded1796f1e8a9884bc97b8660d4b9448c880d23b9dea02b236dc90","target":"graph","created_at":"2026-07-05T07:59:04Z","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/2402.02622/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The transformer architecture by Vaswani et al. (2017) is now ubiquitous across application domains, from natural language processing to speech processing and image understanding. We propose DenseFormer, a simple modification to the standard architecture that improves the perplexity of the model without increasing its size -- adding a few thousand parameters for large-scale models in the 100B parameters range. Our approach relies on an additional averaging step after each transformer block, which computes a weighted average of current and past representations -- we refer to this operation as De","authors_text":"Amirkeivan Mohtashami, Francois Fleuret, Martin Jaggi, Matteo Pagliardini","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-04T21:44:09Z","title":"DenseFormer: Enhancing Information Flow in Transformers via Depth Weighted Averaging"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.02622","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":"sha256:62becc44ba01ec057784309d61bfd8e00fd780f6dd4841ada570d7cd286968b2","target":"record","created_at":"2026-07-05T07:59:04Z","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":"ef659e5dc8b9658f73f50417b28e4657b8469675b5572ab66da2d60ca21e416a","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-04T21:44:09Z","title_canon_sha256":"4384b922d2aff2e4f63175de4baad71d627ba7c49863bdcc0a2b07dd37a81e33"},"schema_version":"1.0","source":{"id":"2402.02622","kind":"arxiv","version":2}},"canonical_sha256":"b5b34aad28b8a41c3cd750c678ca3609f3c79d77c3cbe6794e8a6967d3b4c11f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b5b34aad28b8a41c3cd750c678ca3609f3c79d77c3cbe6794e8a6967d3b4c11f","first_computed_at":"2026-07-05T07:59:04.904258Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:59:04.904258Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RXz65sjWuwrx+n3u2MSXDc61bGAjIWJCVthw+0pSRlUFATLlYyRgw7f2pDnBx5rvrf4W9exHXGrOpWT9GhnZDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:59:04.904723Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.02622","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:62becc44ba01ec057784309d61bfd8e00fd780f6dd4841ada570d7cd286968b2","sha256:51ae915416ded1796f1e8a9884bc97b8660d4b9448c880d23b9dea02b236dc90"],"state_sha256":"20d61f5b6902de5a29ea0b713fcd219ca73f365e45ccc5978989868dbe0ba85f"}