{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VEG4NNTJ3464BJ43XEWOQUHF2Q","short_pith_number":"pith:VEG4NNTJ","schema_version":"1.0","canonical_sha256":"a90dc6b669df3dc0a79bb92ce850e5d40cbc01bba7c8add3c9678ebcf0685c9b","source":{"kind":"arxiv","id":"2605.28398","version":1},"attestation_state":"computed","paper":{"title":"HRBench: Benchmarking and Understanding Thinking-Mode Switch Strategies in Hybrid-Reasoning LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Hao Liu, Jingwen Ye, Mianpeng Liu, Weidong Zhang, Yansong Ning","submitted_at":"2026-05-27T12:35:39Z","abstract_excerpt":"Hybrid-reasoning large language models (LLMs) expose explicit controls over reasoning effort, allowing users or systems to trade off answer quality against inference cost. However, existing methods for adaptive thinking-mode selection are typically evaluated under different models, datasets, and implementation assumptions, making it difficult to compare their practical behavior. We introduce HRBench, a unified evaluation framework for studying thinking-mode switching in hybrid-reasoning LLMs. HRBench organizes the design space along two axes: three switching strategy families, prompt-based sel"},"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":"2605.28398","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-27T12:35:39Z","cross_cats_sorted":[],"title_canon_sha256":"9b2c83cdef0fff9406702b0704365e7beea4dfdedb66a9c79a55cd1f3d6aeb00","abstract_canon_sha256":"4ec88e49adc9aeebf526180f38e5595a48f1d10764e52dda9e00966842c37cba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:05:17.065948Z","signature_b64":"9oZpmaCxO7/f9GL0eJIiT+DdxJ3Whdyye46jXVdZjqIIQQhm9xNmd0LgY4p/tMMZ2588o+sf1S4aulq+zTYbAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a90dc6b669df3dc0a79bb92ce850e5d40cbc01bba7c8add3c9678ebcf0685c9b","last_reissued_at":"2026-05-28T01:05:17.065505Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:05:17.065505Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HRBench: Benchmarking and Understanding Thinking-Mode Switch Strategies in Hybrid-Reasoning LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Hao Liu, Jingwen Ye, Mianpeng Liu, Weidong Zhang, Yansong Ning","submitted_at":"2026-05-27T12:35:39Z","abstract_excerpt":"Hybrid-reasoning large language models (LLMs) expose explicit controls over reasoning effort, allowing users or systems to trade off answer quality against inference cost. However, existing methods for adaptive thinking-mode selection are typically evaluated under different models, datasets, and implementation assumptions, making it difficult to compare their practical behavior. We introduce HRBench, a unified evaluation framework for studying thinking-mode switching in hybrid-reasoning LLMs. HRBench organizes the design space along two axes: three switching strategy families, prompt-based sel"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28398","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/2605.28398/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":"2605.28398","created_at":"2026-05-28T01:05:17.065565+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.28398v1","created_at":"2026-05-28T01:05:17.065565+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28398","created_at":"2026-05-28T01:05:17.065565+00:00"},{"alias_kind":"pith_short_12","alias_value":"VEG4NNTJ3464","created_at":"2026-05-28T01:05:17.065565+00:00"},{"alias_kind":"pith_short_16","alias_value":"VEG4NNTJ3464BJ43","created_at":"2026-05-28T01:05:17.065565+00:00"},{"alias_kind":"pith_short_8","alias_value":"VEG4NNTJ","created_at":"2026-05-28T01:05:17.065565+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/VEG4NNTJ3464BJ43XEWOQUHF2Q","json":"https://pith.science/pith/VEG4NNTJ3464BJ43XEWOQUHF2Q.json","graph_json":"https://pith.science/api/pith-number/VEG4NNTJ3464BJ43XEWOQUHF2Q/graph.json","events_json":"https://pith.science/api/pith-number/VEG4NNTJ3464BJ43XEWOQUHF2Q/events.json","paper":"https://pith.science/paper/VEG4NNTJ"},"agent_actions":{"view_html":"https://pith.science/pith/VEG4NNTJ3464BJ43XEWOQUHF2Q","download_json":"https://pith.science/pith/VEG4NNTJ3464BJ43XEWOQUHF2Q.json","view_paper":"https://pith.science/paper/VEG4NNTJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.28398&json=true","fetch_graph":"https://pith.science/api/pith-number/VEG4NNTJ3464BJ43XEWOQUHF2Q/graph.json","fetch_events":"https://pith.science/api/pith-number/VEG4NNTJ3464BJ43XEWOQUHF2Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VEG4NNTJ3464BJ43XEWOQUHF2Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VEG4NNTJ3464BJ43XEWOQUHF2Q/action/storage_attestation","attest_author":"https://pith.science/pith/VEG4NNTJ3464BJ43XEWOQUHF2Q/action/author_attestation","sign_citation":"https://pith.science/pith/VEG4NNTJ3464BJ43XEWOQUHF2Q/action/citation_signature","submit_replication":"https://pith.science/pith/VEG4NNTJ3464BJ43XEWOQUHF2Q/action/replication_record"}},"created_at":"2026-05-28T01:05:17.065565+00:00","updated_at":"2026-05-28T01:05:17.065565+00:00"}