{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:AIZJIPD3JYSQEOTOCBKQPKFHZP","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"6260e9cce2e63d75ad2c4f9d56ddcfa76c7fcfeec2fdb0396f88d7028c63b497","cross_cats_sorted":["cs.AI","cs.LG","cs.PF"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-04T03:36:44Z","title_canon_sha256":"1c3aa3ff9eb891fdaac3d49597a110f75796db3f99d8b158a263ea7c7fb2f35a"},"schema_version":"1.0","source":{"id":"2502.01976","kind":"arxiv","version":6}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.01976","created_at":"2026-07-05T12:07:58Z"},{"alias_kind":"arxiv_version","alias_value":"2502.01976v6","created_at":"2026-07-05T12:07:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.01976","created_at":"2026-07-05T12:07:58Z"},{"alias_kind":"pith_short_12","alias_value":"AIZJIPD3JYSQ","created_at":"2026-07-05T12:07:58Z"},{"alias_kind":"pith_short_16","alias_value":"AIZJIPD3JYSQEOTO","created_at":"2026-07-05T12:07:58Z"},{"alias_kind":"pith_short_8","alias_value":"AIZJIPD3","created_at":"2026-07-05T12:07:58Z"}],"graph_snapshots":[{"event_id":"sha256:a95a4229eafc59131f25ea288a38819ab9a73c9ff8b3e7c4c2db00ef8aab4337","target":"graph","created_at":"2026-07-05T12:07:58Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2502.01976/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models have achieved remarkable success in various tasks but suffer from high computational costs during inference, limiting their deployment in resource-constrained applications. To address this issue, we propose a novel Collaborative Inference with Token-lEvel Routing (CITER) framework that enables efficient collaboration between small and large language models (SLMs \\& LLMs) through a token-level routing strategy. Specifically, CITER routes non-critical tokens to an SLM for efficiency and routes critical tokens to an LLM for generalization quality. We formulate router trainin","authors_text":"Eric P. Xing, Hongyi Wang, Huaxiu Yao, Souvik Kundu, Weitong Zhang, Wenhao Zheng, Yixiao Chen, Yun Li, Zhengzhong Liu","cross_cats":["cs.AI","cs.LG","cs.PF"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-04T03:36:44Z","title":"CITER: Collaborative Inference for Efficient Large Language Model Decoding with Token-Level Routing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.01976","kind":"arxiv","version":6},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:213079e558bdbd5d321e9f96e97e5ff6a9c6c98176a6ce37b02f012831a3caf1","target":"record","created_at":"2026-07-05T12:07:58Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"6260e9cce2e63d75ad2c4f9d56ddcfa76c7fcfeec2fdb0396f88d7028c63b497","cross_cats_sorted":["cs.AI","cs.LG","cs.PF"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-04T03:36:44Z","title_canon_sha256":"1c3aa3ff9eb891fdaac3d49597a110f75796db3f99d8b158a263ea7c7fb2f35a"},"schema_version":"1.0","source":{"id":"2502.01976","kind":"arxiv","version":6}},"canonical_sha256":"0232943c7b4e25023a6e105507a8a7cbc25ed4eda80142fb625579aec76d14c1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0232943c7b4e25023a6e105507a8a7cbc25ed4eda80142fb625579aec76d14c1","first_computed_at":"2026-07-05T12:07:58.040451Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:07:58.040451Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nTRLU1ZGraJlERRDRhSrqmsj2FwavNIwLtTDbGIogng3DQrrHKwezdU12mebi4rjYGOJCFSTN0gaZ6hbATqlCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T12:07:58.041249Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.01976","source_kind":"arxiv","source_version":6}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:213079e558bdbd5d321e9f96e97e5ff6a9c6c98176a6ce37b02f012831a3caf1","sha256:a95a4229eafc59131f25ea288a38819ab9a73c9ff8b3e7c4c2db00ef8aab4337"],"state_sha256":"c22833329ade794459edbe767f346e5478cf8c43094b701a842fb62124f8017d"}