{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:QJIYYNIFEKWA6I3K44K3HQ5Y6I","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":"816b8676e99c6171051faa9d8a2c4ca086b032f465023efb195fa8a8778c371b","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-04T18:14:07Z","title_canon_sha256":"9c7a59da80081c8e0a5000a8defd2263dbd9dfeb2237b81ee7207df3d4d5b1d6"},"schema_version":"1.0","source":{"id":"2411.02344","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.02344","created_at":"2026-07-05T10:36:15Z"},{"alias_kind":"arxiv_version","alias_value":"2411.02344v2","created_at":"2026-07-05T10:36:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.02344","created_at":"2026-07-05T10:36:15Z"},{"alias_kind":"pith_short_12","alias_value":"QJIYYNIFEKWA","created_at":"2026-07-05T10:36:15Z"},{"alias_kind":"pith_short_16","alias_value":"QJIYYNIFEKWA6I3K","created_at":"2026-07-05T10:36:15Z"},{"alias_kind":"pith_short_8","alias_value":"QJIYYNIF","created_at":"2026-07-05T10:36:15Z"}],"graph_snapshots":[{"event_id":"sha256:7b4a9b8ca5ecfbf4ab74f7f8d009b1145199a12f009e2bbcd473a25a2194e8a5","target":"graph","created_at":"2026-07-05T10:36:15Z","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/2411.02344/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Decoder-only Transformers often struggle with complex reasoning tasks, particularly arithmetic reasoning requiring multiple sequential operations. In this work, we identify representation collapse in the model's intermediate layers as a key factor limiting their reasoning capabilities. To address this, we propose Sequential Variance-Covariance Regularization (Seq-VCR), which enhances the entropy of intermediate representations and prevents collapse. Combined with dummy pause tokens as substitutes for chain-of-thought (CoT) tokens, our method significantly improves performance in arithmetic rea","authors_text":"Christopher Pal, Gopeshh Subbaraj, Irina Rish, Md Rifat Arefin, Nicolas Gontier, Ravid Shwartz-Ziv, Yann LeCun","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-04T18:14:07Z","title":"Seq-VCR: Preventing Collapse in Intermediate Transformer Representations for Enhanced Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.02344","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":"sha256:6288e285495718e93ed265ac1240414745452fbfaac4a7f5016990551e8eec7d","target":"record","created_at":"2026-07-05T10:36:15Z","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":"816b8676e99c6171051faa9d8a2c4ca086b032f465023efb195fa8a8778c371b","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-04T18:14:07Z","title_canon_sha256":"9c7a59da80081c8e0a5000a8defd2263dbd9dfeb2237b81ee7207df3d4d5b1d6"},"schema_version":"1.0","source":{"id":"2411.02344","kind":"arxiv","version":2}},"canonical_sha256":"82518c350522ac0f236ae715b3c3b8f234283d0d8dd1dcdbe338a31afc264e7a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"82518c350522ac0f236ae715b3c3b8f234283d0d8dd1dcdbe338a31afc264e7a","first_computed_at":"2026-07-05T10:36:15.266562Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:36:15.266562Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"EpRGBLX5r4cSIvvNJYsT4RbdmImErrgxpWsYU4ugWQ5iPLQi0M3Hyzrgv0Ab2+gfu7TGk3U0wuQ6lejEBRi8AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:36:15.267219Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.02344","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6288e285495718e93ed265ac1240414745452fbfaac4a7f5016990551e8eec7d","sha256:7b4a9b8ca5ecfbf4ab74f7f8d009b1145199a12f009e2bbcd473a25a2194e8a5"],"state_sha256":"fcc2e03c72dcc0cb60e40ddd5bdcb146bd20a38d113ac68279a0c8240290baae"}