{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OY5WHORFCSPTSECXAMWXG67UIA","short_pith_number":"pith:OY5WHORF","schema_version":"1.0","canonical_sha256":"763b63ba25149f391057032d737bf4401134f89de78b593567e3afbb56d1cb59","source":{"kind":"arxiv","id":"2606.31580","version":1},"attestation_state":"computed","paper":{"title":"LASER: Load-Aware Serving with Early-Exit for Reasoning LLMs at the Edge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Hanshuai Cui, Jianxiong Guo, Size Li, Tian Wang, Weijia Jia, Yuan Wu, Zhiqing Tang, Zilan Huang","submitted_at":"2026-06-30T12:35:48Z","abstract_excerpt":"Large reasoning models (LRMs) such as DeepSeek-R1 have achieved strong performance through extended chain-of-thought (CoT) generation. However, deploying them on edge devices raises a conflict between long CoT sequences and constrained resources. Recent confidence-based early exit methods reduce CoT length for individual requests, yet they apply fixed thresholds from a single-request perspective, ignoring multi-request concurrency and load fluctuation in edge serving. To bridge this gap, we propose \\underline{L}oad-\\underline{A}ware \\underline{S}erving with \\underline{E}arly-exit for \\underlin"},"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":"2606.31580","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-06-30T12:35:48Z","cross_cats_sorted":[],"title_canon_sha256":"4f7e2ae248363bb52890d8c5f24459d4b033668233f8cc5f357dc3e7f0b653fa","abstract_canon_sha256":"1b9b9ca90c93aeb3f809854d245fada44aea05a6dbb7629a5bc1c789af415d76"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-01T01:18:08.282496Z","signature_b64":"ehwvekaIJSiJk//tnFoJBR6k9m1bIkyFO0mLd6ZEHiToxDbHRZfYu1sR6rJIu5AQxRgvfSa6ftjAvOs7IybzBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"763b63ba25149f391057032d737bf4401134f89de78b593567e3afbb56d1cb59","last_reissued_at":"2026-07-01T01:18:08.282021Z","signature_status":"signed_v1","first_computed_at":"2026-07-01T01:18:08.282021Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LASER: Load-Aware Serving with Early-Exit for Reasoning LLMs at the Edge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Hanshuai Cui, Jianxiong Guo, Size Li, Tian Wang, Weijia Jia, Yuan Wu, Zhiqing Tang, Zilan Huang","submitted_at":"2026-06-30T12:35:48Z","abstract_excerpt":"Large reasoning models (LRMs) such as DeepSeek-R1 have achieved strong performance through extended chain-of-thought (CoT) generation. However, deploying them on edge devices raises a conflict between long CoT sequences and constrained resources. Recent confidence-based early exit methods reduce CoT length for individual requests, yet they apply fixed thresholds from a single-request perspective, ignoring multi-request concurrency and load fluctuation in edge serving. To bridge this gap, we propose \\underline{L}oad-\\underline{A}ware \\underline{S}erving with \\underline{E}arly-exit for \\underlin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.31580","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/2606.31580/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":"2606.31580","created_at":"2026-07-01T01:18:08.282096+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.31580v1","created_at":"2026-07-01T01:18:08.282096+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.31580","created_at":"2026-07-01T01:18:08.282096+00:00"},{"alias_kind":"pith_short_12","alias_value":"OY5WHORFCSPT","created_at":"2026-07-01T01:18:08.282096+00:00"},{"alias_kind":"pith_short_16","alias_value":"OY5WHORFCSPTSECX","created_at":"2026-07-01T01:18:08.282096+00:00"},{"alias_kind":"pith_short_8","alias_value":"OY5WHORF","created_at":"2026-07-01T01:18:08.282096+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OY5WHORFCSPTSECXAMWXG67UIA","json":"https://pith.science/pith/OY5WHORFCSPTSECXAMWXG67UIA.json","graph_json":"https://pith.science/api/pith-number/OY5WHORFCSPTSECXAMWXG67UIA/graph.json","events_json":"https://pith.science/api/pith-number/OY5WHORFCSPTSECXAMWXG67UIA/events.json","paper":"https://pith.science/paper/OY5WHORF"},"agent_actions":{"view_html":"https://pith.science/pith/OY5WHORFCSPTSECXAMWXG67UIA","download_json":"https://pith.science/pith/OY5WHORFCSPTSECXAMWXG67UIA.json","view_paper":"https://pith.science/paper/OY5WHORF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.31580&json=true","fetch_graph":"https://pith.science/api/pith-number/OY5WHORFCSPTSECXAMWXG67UIA/graph.json","fetch_events":"https://pith.science/api/pith-number/OY5WHORFCSPTSECXAMWXG67UIA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OY5WHORFCSPTSECXAMWXG67UIA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OY5WHORFCSPTSECXAMWXG67UIA/action/storage_attestation","attest_author":"https://pith.science/pith/OY5WHORFCSPTSECXAMWXG67UIA/action/author_attestation","sign_citation":"https://pith.science/pith/OY5WHORFCSPTSECXAMWXG67UIA/action/citation_signature","submit_replication":"https://pith.science/pith/OY5WHORFCSPTSECXAMWXG67UIA/action/replication_record"}},"created_at":"2026-07-01T01:18:08.282096+00:00","updated_at":"2026-07-01T01:18:08.282096+00:00"}