{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:AHLYPIB676LZ7FPBW42KWQX6QJ","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":"a26ad1ed1c403a87ece0a993f16bb7fcfa2f2cd47b79106611db98be4126430c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-02T19:24:04Z","title_canon_sha256":"9e76903305cc455099b14a0575d466670ccb1769da4f3de4950c4d9498cd9c12"},"schema_version":"1.0","source":{"id":"2602.02711","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.02711","created_at":"2026-05-17T23:39:00Z"},{"alias_kind":"arxiv_version","alias_value":"2602.02711v2","created_at":"2026-05-17T23:39:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.02711","created_at":"2026-05-17T23:39:00Z"},{"alias_kind":"pith_short_12","alias_value":"AHLYPIB676LZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"AHLYPIB676LZ7FPB","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"AHLYPIB6","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:a02ba61f771ee0ac357165b9eaea0e0d79791ea32b045a570792c601fc2f8bb4","target":"graph","created_at":"2026-05-17T23:39:00Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Experiments on ALFWorld and WebShop demonstrate that our approach achieves a strong accuracy-cost trade-off over single-precision baselines."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The observation that interaction steps have diverse sensitivities to precision allows a router to be trained that reliably selects the right precision level without harming overall task success."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"DMR uses a router trained in two stages (KL-divergence supervision then GRPO) to pick high or low precision LLMs per step, delivering better accuracy-cost trade-offs than fixed-precision baselines on ALFWorld and WebShop."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Dynamic mixed-precision routing lets LLMs switch between high and low precision at each step to cut costs while keeping task success high."}],"snapshot_sha256":"0cab96d6ac0b07af14e1e5a4908ba364d2a65dbde47c058901cbabd74cf20275"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"72ddd5294c4e2f5fc1657334bc660c5bca43a59e53677028fc119b6f052bad21"},"paper":{"abstract_excerpt":"Large language models (LLMs) achieve strong performance in long-horizon decision-making tasks through multi-step interaction and reasoning at test time. While practitioners commonly believe a higher task success rate necessitates the use of a larger and stronger LLM model, multi-step interaction with a large LLM incurs prohibitive inference cost.\n  To address this problem, we explore the use of low-precision quantized LLMs in the long-horizon decision-making process. Based on the observation of diverse sensitivities among interaction steps, we propose Dynamic Mixed-Precision Routing (DMR), a f","authors_text":"Huanrui Yang, Jianing Deng, Jingtong Hu, Song Wang, Tianlong Chen, Yuanzhe Li","cross_cats":[],"headline":"Dynamic mixed-precision routing lets LLMs switch between high and low precision at each step to cut costs while keeping task success high.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-02T19:24:04Z","title":"Dynamic Mixed-Precision Routing for Efficient Multi-step LLM Interaction"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.02711","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-16T07:52:44.077711Z","id":"e67e2e07-e25d-4b79-8827-dadc1b5f6d3d","model_set":{"reader":"grok-4.3"},"one_line_summary":"DMR uses a router trained in two stages (KL-divergence supervision then GRPO) to pick high or low precision LLMs per step, delivering better accuracy-cost trade-offs than fixed-precision baselines on ALFWorld and WebShop.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Dynamic mixed-precision routing lets LLMs switch between high and low precision at each step to cut costs while keeping task success high.","strongest_claim":"Experiments on ALFWorld and WebShop demonstrate that our approach achieves a strong accuracy-cost trade-off over single-precision baselines.","weakest_assumption":"The observation that interaction steps have diverse sensitivities to precision allows a router to be trained that reliably selects the right precision level without harming overall task success."}},"verdict_id":"e67e2e07-e25d-4b79-8827-dadc1b5f6d3d"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:be67b2a2a04f31fc10f064d57cb295acca0247b6dde5c75525a2c0f150029c95","target":"record","created_at":"2026-05-17T23:39:00Z","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":"a26ad1ed1c403a87ece0a993f16bb7fcfa2f2cd47b79106611db98be4126430c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-02T19:24:04Z","title_canon_sha256":"9e76903305cc455099b14a0575d466670ccb1769da4f3de4950c4d9498cd9c12"},"schema_version":"1.0","source":{"id":"2602.02711","kind":"arxiv","version":2}},"canonical_sha256":"01d787a03eff979f95e1b734ab42fe824c2843c35348f9ab525f918ee0b25b56","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"01d787a03eff979f95e1b734ab42fe824c2843c35348f9ab525f918ee0b25b56","first_computed_at":"2026-05-17T23:39:00.088347Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:00.088347Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bDFaGQ0XWDTYwwmdcpMv3BXrFIucVNcBR38d/qmBrZqX7870qkeJe5bZSDFjUufACKKdMx7f1TQrU7PVgoKsAg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:00.088944Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.02711","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:be67b2a2a04f31fc10f064d57cb295acca0247b6dde5c75525a2c0f150029c95","sha256:a02ba61f771ee0ac357165b9eaea0e0d79791ea32b045a570792c601fc2f8bb4"],"state_sha256":"5fd78f8de24fec499e6a0253d430c208a56b08d1de8104be67948bb19fa9e161"}