{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VT6CMHEOWGHKIVKRUZL4J73HC2","short_pith_number":"pith:VT6CMHEO","schema_version":"1.0","canonical_sha256":"acfc261c8eb18ea45551a657c4ff6716ba0366b7560544c44e538ae9889ea187","source":{"kind":"arxiv","id":"2605.15236","version":1},"attestation_state":"computed","paper":{"title":"Learning Selective Merge Policies for Deadline-Constrained Coded Caching via Deep Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A graph-attention policy network trained by reinforcement learning learns selective merge decisions that cut packet expiration rates by 40.9 percent in deadline-constrained coded caching.","cross_cats":["cs.AI","cs.NI","math.IT"],"primary_cat":"cs.IT","authors_text":"Amirhossein Yousefiramandi","submitted_at":"2026-05-13T22:18:30Z","abstract_excerpt":"With the coded caching, the server can use the information the users have cached to serve multiple users at a time by sending a single coded multi-casting message, i.e., the merged message, thereby relieving the peak network loads. However, for the delay-sensitive applications of the users, like the video streaming services, it becomes essential to choose which messages to merge online, considering the strict deadlines for each request. The problem, however, is that while the merge is helpful for the formation of the current coded multi-casting message, it can be harmful for the subsequent one"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.15236","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IT","submitted_at":"2026-05-13T22:18:30Z","cross_cats_sorted":["cs.AI","cs.NI","math.IT"],"title_canon_sha256":"4e96b49917f6fe8d1ed25ea62b6a3b1d687df8424e9129b766b4a8e27b8a6faa","abstract_canon_sha256":"15f71ba89defafc43243ae46b193f3a8d6d00c854c2c570e1e26bc8784848cf7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:47.757160Z","signature_b64":"oVE4CHLTPk934S2FCUx77XvWJm7l30+k0Z9weUEFHLEWQulZG8gUEDafPLkrdeAw547J+4gbHmFGqZjJr4mBDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"acfc261c8eb18ea45551a657c4ff6716ba0366b7560544c44e538ae9889ea187","last_reissued_at":"2026-05-20T00:00:47.756420Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:47.756420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Selective Merge Policies for Deadline-Constrained Coded Caching via Deep Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A graph-attention policy network trained by reinforcement learning learns selective merge decisions that cut packet expiration rates by 40.9 percent in deadline-constrained coded caching.","cross_cats":["cs.AI","cs.NI","math.IT"],"primary_cat":"cs.IT","authors_text":"Amirhossein Yousefiramandi","submitted_at":"2026-05-13T22:18:30Z","abstract_excerpt":"With the coded caching, the server can use the information the users have cached to serve multiple users at a time by sending a single coded multi-casting message, i.e., the merged message, thereby relieving the peak network loads. However, for the delay-sensitive applications of the users, like the video streaming services, it becomes essential to choose which messages to merge online, considering the strict deadlines for each request. The problem, however, is that while the merge is helpful for the formation of the current coded multi-casting message, it can be harmful for the subsequent one"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The policy network reduces the broadcast-packet expiration ratio ρ by 40.9% (0.208 vs. 0.352) with respect to the best coded multi-casting baseline (SACM++) on the uniform-demand benchmark, while also attaining the best broadcast-efficiency score σ across the Track A battery among the coded multi-casting methods.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The simulator used for training and evaluation accurately captures the real-world trade-offs between current multicast opportunities and future deadline violations in coded caching systems (abstract states the formulation as a masked discrete-action queue-state control problem without providing validation against live networks).","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A DRL policy with graph attention learns selective merging for deadline-constrained coded caching, cutting packet expiration ratio by 40.9% versus SACM++ while merging only about 32% of the time.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A graph-attention policy network trained by reinforcement learning learns selective merge decisions that cut packet expiration rates by 40.9 percent in deadline-constrained coded caching.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"7b89698be64007f821bfd3d2712fd725e060e347cf41084f4881048f4d2f8d50"},"source":{"id":"2605.15236","kind":"arxiv","version":1},"verdict":{"id":"9d60ff35-d911-40dd-ac33-f8b8ab68ad67","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T17:21:41.096037Z","strongest_claim":"The policy network reduces the broadcast-packet expiration ratio ρ by 40.9% (0.208 vs. 0.352) with respect to the best coded multi-casting baseline (SACM++) on the uniform-demand benchmark, while also attaining the best broadcast-efficiency score σ across the Track A battery among the coded multi-casting methods.","one_line_summary":"A DRL policy with graph attention learns selective merging for deadline-constrained coded caching, cutting packet expiration ratio by 40.9% versus SACM++ while merging only about 32% of the time.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The simulator used for training and evaluation accurately captures the real-world trade-offs between current multicast opportunities and future deadline violations in coded caching systems (abstract states the formulation as a masked discrete-action queue-state control problem without providing validation against live networks).","pith_extraction_headline":"A graph-attention policy network trained by reinforcement learning learns selective merge decisions that cut packet expiration rates by 40.9 percent in deadline-constrained coded caching."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.15236/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T17:31:18.472992Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T17:26:19.765610Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T16:21:55.596454Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.826169Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"bae4ab9a4867b3848f93493dc63c05f5e7804e600fd5e12319e6430c3ac622f6"},"references":{"count":32,"sample":[{"doi":"","year":2024,"title":"Ericsson mobility report, June 2024,","work_id":"6119169d-71e3-4588-9083-261bd44a963e","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"Mobile edge caching: A survey,","work_id":"ede55c89-1424-4805-8c8e-d7230dda08ad","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2014,"title":"Fundamental limits of caching,","work_id":"b322afe2-dcaa-40c2-85e6-d38de54229d6","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2015,"title":"Decentralized coded caching attains order-optimal memory-rate tradeoff,","work_id":"8d2f1fae-26ba-4fcf-93b7-71b8f54894d5","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2014,"title":"Coded Caching for Delay-Sensitive Content","work_id":"2930a5d7-b063-4a48-aff0-526d724b1279","ref_index":5,"cited_arxiv_id":"1407.4489","is_internal_anchor":true}],"resolved_work":32,"snapshot_sha256":"ca5d0765b8123e14412c753d1409c262aea6e90ecf9144768fa830f4e9eb2389","internal_anchors":5},"formal_canon":{"evidence_count":2,"snapshot_sha256":"1093cfd7f6989cf6c797f9ba0b8b51ec5a8a3192f7d3ae3740296b9d8ffa4245"},"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":"2605.15236","created_at":"2026-05-20T00:00:47.756571+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.15236v1","created_at":"2026-05-20T00:00:47.756571+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15236","created_at":"2026-05-20T00:00:47.756571+00:00"},{"alias_kind":"pith_short_12","alias_value":"VT6CMHEOWGHK","created_at":"2026-05-20T00:00:47.756571+00:00"},{"alias_kind":"pith_short_16","alias_value":"VT6CMHEOWGHKIVKR","created_at":"2026-05-20T00:00:47.756571+00:00"},{"alias_kind":"pith_short_8","alias_value":"VT6CMHEO","created_at":"2026-05-20T00:00:47.756571+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VT6CMHEOWGHKIVKRUZL4J73HC2","json":"https://pith.science/pith/VT6CMHEOWGHKIVKRUZL4J73HC2.json","graph_json":"https://pith.science/api/pith-number/VT6CMHEOWGHKIVKRUZL4J73HC2/graph.json","events_json":"https://pith.science/api/pith-number/VT6CMHEOWGHKIVKRUZL4J73HC2/events.json","paper":"https://pith.science/paper/VT6CMHEO"},"agent_actions":{"view_html":"https://pith.science/pith/VT6CMHEOWGHKIVKRUZL4J73HC2","download_json":"https://pith.science/pith/VT6CMHEOWGHKIVKRUZL4J73HC2.json","view_paper":"https://pith.science/paper/VT6CMHEO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.15236&json=true","fetch_graph":"https://pith.science/api/pith-number/VT6CMHEOWGHKIVKRUZL4J73HC2/graph.json","fetch_events":"https://pith.science/api/pith-number/VT6CMHEOWGHKIVKRUZL4J73HC2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VT6CMHEOWGHKIVKRUZL4J73HC2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VT6CMHEOWGHKIVKRUZL4J73HC2/action/storage_attestation","attest_author":"https://pith.science/pith/VT6CMHEOWGHKIVKRUZL4J73HC2/action/author_attestation","sign_citation":"https://pith.science/pith/VT6CMHEOWGHKIVKRUZL4J73HC2/action/citation_signature","submit_replication":"https://pith.science/pith/VT6CMHEOWGHKIVKRUZL4J73HC2/action/replication_record"}},"created_at":"2026-05-20T00:00:47.756571+00:00","updated_at":"2026-05-20T00:00:47.756571+00:00"}