{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:CKX65LSV3ZFEKSZTZ3Z2T6D3FF","short_pith_number":"pith:CKX65LSV","schema_version":"1.0","canonical_sha256":"12afeeae55de4a454b33cef3a9f87b2944dfa22e0f0adffe54b7c496234cd4dd","source":{"kind":"arxiv","id":"2409.13757","version":1},"attestation_state":"computed","paper":{"title":"Efficient Hybrid Inference for LLMs: Reward-Based Token Modelling with Selective Cloud Assistance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Adarsh MS, Ditto PS, Jithin VG","submitted_at":"2024-09-15T15:12:45Z","abstract_excerpt":"Large language models (LLMs) are known for their exceptional performance across a range of natural language processing tasks, but their deployment comes at a high computational and financial cost. On the other hand, smaller language models (SLMs), which can be deployed on lower-cost edge devices, struggle to match the performance of their larger counterparts. This paper presents a novel hybrid inference approach that leverages the strengths of both model types while minimizing reliance on costly cloud-based LLMs. Unlike existing methods that route entire queries to either an SLM or a cloud LLM"},"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":"2409.13757","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-15T15:12:45Z","cross_cats_sorted":[],"title_canon_sha256":"ae8502a56a82683f4ec9d52d2c7678c5a5c7d2bb308f5382924e9ce58d10ed85","abstract_canon_sha256":"75d18a4932984a1559f7f8b5a8b5fef364089172a164710824f61bc4c3401514"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:10:06.373389Z","signature_b64":"dltYjzPyTK5VYgvXdNUnrmrXvRkpaZUg8JalY6uKBJXmFj0V8KEaLpYoKFeldughPI7mixotfaNPExzaavQwBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"12afeeae55de4a454b33cef3a9f87b2944dfa22e0f0adffe54b7c496234cd4dd","last_reissued_at":"2026-07-05T09:10:06.372898Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:10:06.372898Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Hybrid Inference for LLMs: Reward-Based Token Modelling with Selective Cloud Assistance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Adarsh MS, Ditto PS, Jithin VG","submitted_at":"2024-09-15T15:12:45Z","abstract_excerpt":"Large language models (LLMs) are known for their exceptional performance across a range of natural language processing tasks, but their deployment comes at a high computational and financial cost. On the other hand, smaller language models (SLMs), which can be deployed on lower-cost edge devices, struggle to match the performance of their larger counterparts. This paper presents a novel hybrid inference approach that leverages the strengths of both model types while minimizing reliance on costly cloud-based LLMs. Unlike existing methods that route entire queries to either an SLM or a cloud LLM"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.13757","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/2409.13757/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":"2409.13757","created_at":"2026-07-05T09:10:06.372956+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.13757v1","created_at":"2026-07-05T09:10:06.372956+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.13757","created_at":"2026-07-05T09:10:06.372956+00:00"},{"alias_kind":"pith_short_12","alias_value":"CKX65LSV3ZFE","created_at":"2026-07-05T09:10:06.372956+00:00"},{"alias_kind":"pith_short_16","alias_value":"CKX65LSV3ZFEKSZT","created_at":"2026-07-05T09:10:06.372956+00:00"},{"alias_kind":"pith_short_8","alias_value":"CKX65LSV","created_at":"2026-07-05T09:10:06.372956+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.06924","citing_title":"From Sampled Outcomes to Capability Distributions: Rethinking Supervision for LLM Routing","ref_index":150,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CKX65LSV3ZFEKSZTZ3Z2T6D3FF","json":"https://pith.science/pith/CKX65LSV3ZFEKSZTZ3Z2T6D3FF.json","graph_json":"https://pith.science/api/pith-number/CKX65LSV3ZFEKSZTZ3Z2T6D3FF/graph.json","events_json":"https://pith.science/api/pith-number/CKX65LSV3ZFEKSZTZ3Z2T6D3FF/events.json","paper":"https://pith.science/paper/CKX65LSV"},"agent_actions":{"view_html":"https://pith.science/pith/CKX65LSV3ZFEKSZTZ3Z2T6D3FF","download_json":"https://pith.science/pith/CKX65LSV3ZFEKSZTZ3Z2T6D3FF.json","view_paper":"https://pith.science/paper/CKX65LSV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.13757&json=true","fetch_graph":"https://pith.science/api/pith-number/CKX65LSV3ZFEKSZTZ3Z2T6D3FF/graph.json","fetch_events":"https://pith.science/api/pith-number/CKX65LSV3ZFEKSZTZ3Z2T6D3FF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CKX65LSV3ZFEKSZTZ3Z2T6D3FF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CKX65LSV3ZFEKSZTZ3Z2T6D3FF/action/storage_attestation","attest_author":"https://pith.science/pith/CKX65LSV3ZFEKSZTZ3Z2T6D3FF/action/author_attestation","sign_citation":"https://pith.science/pith/CKX65LSV3ZFEKSZTZ3Z2T6D3FF/action/citation_signature","submit_replication":"https://pith.science/pith/CKX65LSV3ZFEKSZTZ3Z2T6D3FF/action/replication_record"}},"created_at":"2026-07-05T09:10:06.372956+00:00","updated_at":"2026-07-05T09:10:06.372956+00:00"}