{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:2LHIJ7ASYWPL2SAKNWL5RFVK75","short_pith_number":"pith:2LHIJ7AS","schema_version":"1.0","canonical_sha256":"d2ce84fc12c59ebd480a6d97d896aaff4b172a5df81fed783edfa8afefb9939c","source":{"kind":"arxiv","id":"1901.02860","version":3},"attestation_state":"computed","paper":{"title":"Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov, Yiming Yang, Zhilin Yang, Zihang Dai","submitted_at":"2019-01-09T18:28:19Z","abstract_excerpt":"Transformers have a potential of learning longer-term dependency, but are limited by a fixed-length context in the setting of language modeling. We propose a novel neural architecture Transformer-XL that enables learning dependency beyond a fixed length without disrupting temporal coherence. It consists of a segment-level recurrence mechanism and a novel positional encoding scheme. Our method not only enables capturing longer-term dependency, but also resolves the context fragmentation problem. As a result, Transformer-XL learns dependency that is 80% longer than RNNs and 450% longer than vani"},"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":"1901.02860","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2019-01-09T18:28:19Z","cross_cats_sorted":["cs.CL","stat.ML"],"title_canon_sha256":"1bf7f897a9ed43505f9fb7a4c2076daecf84141720adf96ea983278240ab576f","abstract_canon_sha256":"1be40747d33be11293c75e4a19ad3573657e67ecba9d9d047d5af2591d95c2bc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:44:29.634667Z","signature_b64":"BegWF/g761Hqb9zv1IZxHY5s7INjWDNSsAUs+VLyEKi4f6iarnS43sYNZZeR5AA7lYANKWUNl3AbAeEsqhWEDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d2ce84fc12c59ebd480a6d97d896aaff4b172a5df81fed783edfa8afefb9939c","last_reissued_at":"2026-05-17T23:44:29.634168Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:44:29.634168Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov, Yiming Yang, Zhilin Yang, Zihang Dai","submitted_at":"2019-01-09T18:28:19Z","abstract_excerpt":"Transformers have a potential of learning longer-term dependency, but are limited by a fixed-length context in the setting of language modeling. We propose a novel neural architecture Transformer-XL that enables learning dependency beyond a fixed length without disrupting temporal coherence. It consists of a segment-level recurrence mechanism and a novel positional encoding scheme. Our method not only enables capturing longer-term dependency, but also resolves the context fragmentation problem. As a result, Transformer-XL learns dependency that is 80% longer than RNNs and 450% longer than vani"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.02860","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":""},"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":"1901.02860","created_at":"2026-05-17T23:44:29.634241+00:00"},{"alias_kind":"arxiv_version","alias_value":"1901.02860v3","created_at":"2026-05-17T23:44:29.634241+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.02860","created_at":"2026-05-17T23:44:29.634241+00:00"},{"alias_kind":"pith_short_12","alias_value":"2LHIJ7ASYWPL","created_at":"2026-05-18T12:33:07.085635+00:00"},{"alias_kind":"pith_short_16","alias_value":"2LHIJ7ASYWPL2SAK","created_at":"2026-05-18T12:33:07.085635+00:00"},{"alias_kind":"pith_short_8","alias_value":"2LHIJ7AS","created_at":"2026-05-18T12:33:07.085635+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":39,"internal_anchor_count":23,"sample":[{"citing_arxiv_id":"1907.01470","citing_title":"Augmenting Self-attention with Persistent Memory","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2507.06252","citing_title":"False Alarms, Real Damage: Adversarial Attacks Using LLM-based Models on Text-based Cyber Threat Intelligence Systems","ref_index":58,"is_internal_anchor":true},{"citing_arxiv_id":"1907.04868","citing_title":"LakhNES: Improving multi-instrumental music generation with cross-domain pre-training","ref_index":42,"is_internal_anchor":true},{"citing_arxiv_id":"1907.05572","citing_title":"R-Transformer: Recurrent Neural Network Enhanced Transformer","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"1907.06607","citing_title":"Agglomerative Attention","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"1907.09669","citing_title":"EmotionX-HSU: Adopting Pre-trained BERT for Emotion Classification","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2107.06499","citing_title":"Deduplicating Training Data Makes Language Models Better","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2503.18970","citing_title":"Advancing Intelligent Sequence Modeling: Evolution, Trade-offs, and Applications of State- Space Architectures from S4 to Mamba","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2504.14386","citing_title":"LOOPE: Learnable Optimal Patch Order in Positional Embeddings for Vision Transformers","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2404.07143","citing_title":"Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18856","citing_title":"SPHERICAL KV: Angle-Domain Attention and Rate-Distortion Retention for Efficient Long-Context Inference","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2308.08089","citing_title":"DragNUWA: Fine-grained Control in Video Generation by Integrating Text, Image, and Trajectory","ref_index":172,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18530","citing_title":"Continuous Diffusion Scales Competitively with Discrete Diffusion for Language","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2307.06435","citing_title":"A Comprehensive Overview of Large Language Models","ref_index":111,"is_internal_anchor":true},{"citing_arxiv_id":"2507.00432","citing_title":"Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning","ref_index":245,"is_internal_anchor":true},{"citing_arxiv_id":"1911.05507","citing_title":"Compressive Transformers for Long-Range Sequence Modelling","ref_index":111,"is_internal_anchor":true},{"citing_arxiv_id":"1906.08237","citing_title":"XLNet: Generalized Autoregressive Pretraining for Language Understanding","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"1909.05858","citing_title":"CTRL: A Conditional Transformer Language Model for Controllable Generation","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2310.01852","citing_title":"LanguageBind: Extending Video-Language Pretraining to N-modality by Language-based Semantic Alignment","ref_index":70,"is_internal_anchor":true},{"citing_arxiv_id":"2309.16797","citing_title":"Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution","ref_index":85,"is_internal_anchor":true},{"citing_arxiv_id":"2404.18416","citing_title":"Capabilities of Gemini Models in Medicine","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2604.03263","citing_title":"LPC-SM: Local Predictive Coding and Sparse Memory for Long-Context Language Modeling","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2303.16199","citing_title":"LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention","ref_index":103,"is_internal_anchor":true},{"citing_arxiv_id":"1909.11942","citing_title":"ALBERT: A Lite BERT for Self-supervised Learning of Language Representations","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11007","citing_title":"RT-Transformer: The Transformer Block as a Spherical State Estimator","ref_index":148,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2LHIJ7ASYWPL2SAKNWL5RFVK75","json":"https://pith.science/pith/2LHIJ7ASYWPL2SAKNWL5RFVK75.json","graph_json":"https://pith.science/api/pith-number/2LHIJ7ASYWPL2SAKNWL5RFVK75/graph.json","events_json":"https://pith.science/api/pith-number/2LHIJ7ASYWPL2SAKNWL5RFVK75/events.json","paper":"https://pith.science/paper/2LHIJ7AS"},"agent_actions":{"view_html":"https://pith.science/pith/2LHIJ7ASYWPL2SAKNWL5RFVK75","download_json":"https://pith.science/pith/2LHIJ7ASYWPL2SAKNWL5RFVK75.json","view_paper":"https://pith.science/paper/2LHIJ7AS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1901.02860&json=true","fetch_graph":"https://pith.science/api/pith-number/2LHIJ7ASYWPL2SAKNWL5RFVK75/graph.json","fetch_events":"https://pith.science/api/pith-number/2LHIJ7ASYWPL2SAKNWL5RFVK75/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2LHIJ7ASYWPL2SAKNWL5RFVK75/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2LHIJ7ASYWPL2SAKNWL5RFVK75/action/storage_attestation","attest_author":"https://pith.science/pith/2LHIJ7ASYWPL2SAKNWL5RFVK75/action/author_attestation","sign_citation":"https://pith.science/pith/2LHIJ7ASYWPL2SAKNWL5RFVK75/action/citation_signature","submit_replication":"https://pith.science/pith/2LHIJ7ASYWPL2SAKNWL5RFVK75/action/replication_record"}},"created_at":"2026-05-17T23:44:29.634241+00:00","updated_at":"2026-05-17T23:44:29.634241+00:00"}