{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:XDBBMGNOVAI4MP4E6SDTD4MPIB","short_pith_number":"pith:XDBBMGNO","schema_version":"1.0","canonical_sha256":"b8c21619aea811c63f84f48731f18f406b46b947ec027c18fbdd28fb6dbaf274","source":{"kind":"arxiv","id":"2405.04517","version":2},"attestation_state":"computed","paper":{"title":"xLSTM: Extended Long Short-Term Memory","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andreas Auer, G\\\"unter Klambauer, Johannes Brandstetter, Korbinian P\\\"oppel, Markus Spanring, Maximilian Beck, Michael Kopp, Oleksandra Prudnikova, Sepp Hochreiter","submitted_at":"2024-05-07T17:50:21Z","abstract_excerpt":"In the 1990s, the constant error carousel and gating were introduced as the central ideas of the Long Short-Term Memory (LSTM). Since then, LSTMs have stood the test of time and contributed to numerous deep learning success stories, in particular they constituted the first Large Language Models (LLMs). However, the advent of the Transformer technology with parallelizable self-attention at its core marked the dawn of a new era, outpacing LSTMs at scale. We now raise a simple question: How far do we get in language modeling when scaling LSTMs to billions of parameters, leveraging the latest tech"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2405.04517","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-05-07T17:50:21Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"9a8ca71d27373ba6e98cdc55683ac39a1cee2ad7b0be3e48fc8b4f60a2004cfb","abstract_canon_sha256":"0c4618aabd254d6c737a9374841c6e218462eb789d528dd26ed2923a0f3b0caf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:45:09.928493Z","signature_b64":"sqqI41UyRyNXBxPf5fQJDjyyIwuvMd+VyI7GD+wxk+kuRALFVdhHFqRM3pDML8mDe1u9f2mOrmQ8+DJdZhqbCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b8c21619aea811c63f84f48731f18f406b46b947ec027c18fbdd28fb6dbaf274","last_reissued_at":"2026-07-05T09:45:09.927981Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:45:09.927981Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"xLSTM: Extended Long Short-Term Memory","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andreas Auer, G\\\"unter Klambauer, Johannes Brandstetter, Korbinian P\\\"oppel, Markus Spanring, Maximilian Beck, Michael Kopp, Oleksandra Prudnikova, Sepp Hochreiter","submitted_at":"2024-05-07T17:50:21Z","abstract_excerpt":"In the 1990s, the constant error carousel and gating were introduced as the central ideas of the Long Short-Term Memory (LSTM). Since then, LSTMs have stood the test of time and contributed to numerous deep learning success stories, in particular they constituted the first Large Language Models (LLMs). However, the advent of the Transformer technology with parallelizable self-attention at its core marked the dawn of a new era, outpacing LSTMs at scale. We now raise a simple question: How far do we get in language modeling when scaling LSTMs to billions of parameters, leveraging the latest tech"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.04517","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2405.04517/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2405.04517","created_at":"2026-07-05T09:45:09.928046+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.04517v2","created_at":"2026-07-05T09:45:09.928046+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.04517","created_at":"2026-07-05T09:45:09.928046+00:00"},{"alias_kind":"pith_short_12","alias_value":"XDBBMGNOVAI4","created_at":"2026-07-05T09:45:09.928046+00:00"},{"alias_kind":"pith_short_16","alias_value":"XDBBMGNOVAI4MP4E","created_at":"2026-07-05T09:45:09.928046+00:00"},{"alias_kind":"pith_short_8","alias_value":"XDBBMGNO","created_at":"2026-07-05T09:45:09.928046+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":26,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07386","citing_title":"Sparse Delta Memory: Scaling the State of Linear RNNs through Sparsity","ref_index":81,"is_internal_anchor":true},{"citing_arxiv_id":"2606.25010","citing_title":"Emergent Capabilities Arise Randomly from Learning Sparse Attention Patterns","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.26749","citing_title":"Structure Before Collapse: Transient semantic geometry in next-token prediction","ref_index":66,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03979","citing_title":"Language Models Need Sleep: Learning to Self-Modify and Consolidate Memories","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2605.31163","citing_title":"Memory by Design: Probabilistic Sequence Layers","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27538","citing_title":"The Context-Ready Transformer","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2503.08223","citing_title":"Will LLMs Scaling Hit the Wall? Breaking Barriers via Distributed Resources on Massive Edge Devices","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2504.10013","citing_title":"Training LLMs on HPC Systems: Best Practices from the OpenGPT-X Project","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2407.08608","citing_title":"FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17108","citing_title":"Parallel Recursive LSTM","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2509.24552","citing_title":"Short window attention enables long-term memorization","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2510.17018","citing_title":"CoGate-LSTM: Prototype-Guided Feature-Space Gating for Mitigating Gradient Dilution in Imbalanced Toxic Comment Classification","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2512.21370","citing_title":"Detection of Lensed Gravitational Waves in the Millihertz Band Using Frequency-Domain Lensing Feature Extraction Network","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2407.04620","citing_title":"Learning to (Learn at Test Time): RNNs with Expressive Hidden States","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2312.06635","citing_title":"Gated Linear Attention Transformers with Hardware-Efficient Training","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2501.00663","citing_title":"Titans: Learning to Memorize at Test Time","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13807","citing_title":"Parallel Scan Recurrent Neural Quantum States for Scalable Variational Monte Carlo","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02771","citing_title":"ContractShield: Bridging Semantic-Structural Gaps via Hierarchical Cross-Modal Fusion for Multi-Label Vulnerability Detection in Obfuscated Smart Contracts","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2412.06464","citing_title":"Gated Delta Networks: Improving Mamba2 with Delta Rule","ref_index":299,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08587","citing_title":"Kaczmarz Linear Attention","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10643","citing_title":"A Single-Layer Model Can Do Language Modeling","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19343","citing_title":"Scalable Memristive-Friendly Reservoir Computing for Time Series Classification","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2405.21060","citing_title":"Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2404.06654","citing_title":"RULER: What's the Real Context Size of Your Long-Context Language Models?","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07142","citing_title":"AGA3DNet: Anatomy-Guided Gaussian Priors with Multi-view xLSTM for 3D Brain MRI Subtype Classification","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XDBBMGNOVAI4MP4E6SDTD4MPIB","json":"https://pith.science/pith/XDBBMGNOVAI4MP4E6SDTD4MPIB.json","graph_json":"https://pith.science/api/pith-number/XDBBMGNOVAI4MP4E6SDTD4MPIB/graph.json","events_json":"https://pith.science/api/pith-number/XDBBMGNOVAI4MP4E6SDTD4MPIB/events.json","paper":"https://pith.science/paper/XDBBMGNO"},"agent_actions":{"view_html":"https://pith.science/pith/XDBBMGNOVAI4MP4E6SDTD4MPIB","download_json":"https://pith.science/pith/XDBBMGNOVAI4MP4E6SDTD4MPIB.json","view_paper":"https://pith.science/paper/XDBBMGNO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.04517&json=true","fetch_graph":"https://pith.science/api/pith-number/XDBBMGNOVAI4MP4E6SDTD4MPIB/graph.json","fetch_events":"https://pith.science/api/pith-number/XDBBMGNOVAI4MP4E6SDTD4MPIB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XDBBMGNOVAI4MP4E6SDTD4MPIB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XDBBMGNOVAI4MP4E6SDTD4MPIB/action/storage_attestation","attest_author":"https://pith.science/pith/XDBBMGNOVAI4MP4E6SDTD4MPIB/action/author_attestation","sign_citation":"https://pith.science/pith/XDBBMGNOVAI4MP4E6SDTD4MPIB/action/citation_signature","submit_replication":"https://pith.science/pith/XDBBMGNOVAI4MP4E6SDTD4MPIB/action/replication_record"}},"created_at":"2026-07-05T09:45:09.928046+00:00","updated_at":"2026-07-05T09:45:09.928046+00:00"}