{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7GUERYJDQ53UW25SBOYWNBG3O4","short_pith_number":"pith:7GUERYJD","schema_version":"1.0","canonical_sha256":"f9a848e12387774b6bb20bb16684db770ed03d36d296c6da3b743395a277a4fc","source":{"kind":"arxiv","id":"2405.08944","version":1},"attestation_state":"computed","paper":{"title":"Challenges in Deploying Long-Context Transformers: A Theoretical Peak Performance Analysis","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.DC"],"primary_cat":"cs.LG","authors_text":"Yao Fu","submitted_at":"2024-05-14T20:17:22Z","abstract_excerpt":"Transformer-based long context generative models power emerging AI applications like hour-long video understanding and project-level coding agent. Deploying long context transformers (e.g., 100K to 10M tokens) is prohibitively expensive compared to short context (e.g., 4K tokens) model variants. Reducing the cost of long-context transformers is becoming a pressing research and engineering challenge starting from the year of 2024. This work describes a concurrent programming framework for quantitatively analyzing the efficiency challenges in serving multiple long-context requests under limited "},"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":"2405.08944","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-05-14T20:17:22Z","cross_cats_sorted":["cs.AI","cs.CL","cs.DC"],"title_canon_sha256":"18cd756c0d1c0e76ddfb807d164b38c3f0f06a9a2345fefd2aa02e78478ebbf3","abstract_canon_sha256":"67325f3aed05e0d98e81100584a61832eef06fe98dc38c98060e8cfb8a848da7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:19:11.130636Z","signature_b64":"00y0gIpzpJXSjwJ7WzkEtsGE85or1kiyF6oZpE26bou0CqvZWxQUKUsMv9XEuR9uimVKZw0yGmJGRaxP6dtkCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f9a848e12387774b6bb20bb16684db770ed03d36d296c6da3b743395a277a4fc","last_reissued_at":"2026-07-05T08:19:11.130158Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:19:11.130158Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Challenges in Deploying Long-Context Transformers: A Theoretical Peak Performance Analysis","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.DC"],"primary_cat":"cs.LG","authors_text":"Yao Fu","submitted_at":"2024-05-14T20:17:22Z","abstract_excerpt":"Transformer-based long context generative models power emerging AI applications like hour-long video understanding and project-level coding agent. Deploying long context transformers (e.g., 100K to 10M tokens) is prohibitively expensive compared to short context (e.g., 4K tokens) model variants. Reducing the cost of long-context transformers is becoming a pressing research and engineering challenge starting from the year of 2024. This work describes a concurrent programming framework for quantitatively analyzing the efficiency challenges in serving multiple long-context requests under limited "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.08944","kind":"arxiv","version":1},"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/2405.08944/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":"2405.08944","created_at":"2026-07-05T08:19:11.130213+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.08944v1","created_at":"2026-07-05T08:19:11.130213+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.08944","created_at":"2026-07-05T08:19:11.130213+00:00"},{"alias_kind":"pith_short_12","alias_value":"7GUERYJDQ53U","created_at":"2026-07-05T08:19:11.130213+00:00"},{"alias_kind":"pith_short_16","alias_value":"7GUERYJDQ53UW25S","created_at":"2026-07-05T08:19:11.130213+00:00"},{"alias_kind":"pith_short_8","alias_value":"7GUERYJD","created_at":"2026-07-05T08:19:11.130213+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05868","citing_title":"YouZhi: Towards High-Concurrency Financial LLMs via Adaptive GQA-to-MLA Transition","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24820","citing_title":"Salca: A Sparsity-Aware Hardware Accelerator for Efficient Long-Context Attention Decoding","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05219","citing_title":"Sparse Prefix Caching for Hybrid and Recurrent LLM Serving","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16883","citing_title":"SinkRouter: Sink-Aware Routing for Efficient Long-Context Decoding in Large Language and Multimodal Models","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7GUERYJDQ53UW25SBOYWNBG3O4","json":"https://pith.science/pith/7GUERYJDQ53UW25SBOYWNBG3O4.json","graph_json":"https://pith.science/api/pith-number/7GUERYJDQ53UW25SBOYWNBG3O4/graph.json","events_json":"https://pith.science/api/pith-number/7GUERYJDQ53UW25SBOYWNBG3O4/events.json","paper":"https://pith.science/paper/7GUERYJD"},"agent_actions":{"view_html":"https://pith.science/pith/7GUERYJDQ53UW25SBOYWNBG3O4","download_json":"https://pith.science/pith/7GUERYJDQ53UW25SBOYWNBG3O4.json","view_paper":"https://pith.science/paper/7GUERYJD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.08944&json=true","fetch_graph":"https://pith.science/api/pith-number/7GUERYJDQ53UW25SBOYWNBG3O4/graph.json","fetch_events":"https://pith.science/api/pith-number/7GUERYJDQ53UW25SBOYWNBG3O4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7GUERYJDQ53UW25SBOYWNBG3O4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7GUERYJDQ53UW25SBOYWNBG3O4/action/storage_attestation","attest_author":"https://pith.science/pith/7GUERYJDQ53UW25SBOYWNBG3O4/action/author_attestation","sign_citation":"https://pith.science/pith/7GUERYJDQ53UW25SBOYWNBG3O4/action/citation_signature","submit_replication":"https://pith.science/pith/7GUERYJDQ53UW25SBOYWNBG3O4/action/replication_record"}},"created_at":"2026-07-05T08:19:11.130213+00:00","updated_at":"2026-07-05T08:19:11.130213+00:00"}