{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JU6VNNPPO2XRJ7RC6QCRWODCPK","short_pith_number":"pith:JU6VNNPP","schema_version":"1.0","canonical_sha256":"4d3d56b5ef76af14fe22f4051b38627a9debdba1a2b902eab1971ad2e56ef082","source":{"kind":"arxiv","id":"2507.17699","version":1},"attestation_state":"computed","paper":{"title":"Thinking Isn't an Illusion: Overcoming the Limitations of Reasoning Models via Tool Augmentations","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jiahao Zhang, Song Yue, Zhao Song","submitted_at":"2025-07-23T17:04:20Z","abstract_excerpt":"Large Reasoning Models (LRMs) have become a central focus in today's large language model (LLM) research, where models are designed to output a step-by-step thinking process before arriving at a final answer to handle complex reasoning tasks. Despite their promise, recent empirical studies (e.g., [Shojaee et al., 2025] from Apple) suggest that this thinking process may not actually enhance reasoning ability, where LLMs without explicit reasoning actually outperform LRMs on tasks with low or high complexity. In this work, we revisit these findings and investigate whether the limitations of LRMs"},"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":"2507.17699","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-23T17:04:20Z","cross_cats_sorted":[],"title_canon_sha256":"e68b7aa9e5bcd3cbd733db9e88449f06296efc279d47cf2b329d6bb9c1c5767f","abstract_canon_sha256":"838d86a89c8d8593b5f47a8218232a7710a6cb8bca64907bd0c2359fbd18eda5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:12.002243Z","signature_b64":"ZOfuqLBDlHO/Iw2jNcBkHqWx5CLnts/M6n4sqN6gxb/8WMUV9B8qZeU2FuXTiEI34R3D2K9qY28L+mpUL1TdAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4d3d56b5ef76af14fe22f4051b38627a9debdba1a2b902eab1971ad2e56ef082","last_reissued_at":"2026-07-05T11:42:12.001774Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:12.001774Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Thinking Isn't an Illusion: Overcoming the Limitations of Reasoning Models via Tool Augmentations","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jiahao Zhang, Song Yue, Zhao Song","submitted_at":"2025-07-23T17:04:20Z","abstract_excerpt":"Large Reasoning Models (LRMs) have become a central focus in today's large language model (LLM) research, where models are designed to output a step-by-step thinking process before arriving at a final answer to handle complex reasoning tasks. Despite their promise, recent empirical studies (e.g., [Shojaee et al., 2025] from Apple) suggest that this thinking process may not actually enhance reasoning ability, where LLMs without explicit reasoning actually outperform LRMs on tasks with low or high complexity. In this work, we revisit these findings and investigate whether the limitations of LRMs"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.17699","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/2507.17699/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":"2507.17699","created_at":"2026-07-05T11:42:12.001830+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.17699v1","created_at":"2026-07-05T11:42:12.001830+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.17699","created_at":"2026-07-05T11:42:12.001830+00:00"},{"alias_kind":"pith_short_12","alias_value":"JU6VNNPPO2XR","created_at":"2026-07-05T11:42:12.001830+00:00"},{"alias_kind":"pith_short_16","alias_value":"JU6VNNPPO2XRJ7RC","created_at":"2026-07-05T11:42:12.001830+00:00"},{"alias_kind":"pith_short_8","alias_value":"JU6VNNPP","created_at":"2026-07-05T11:42:12.001830+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.30052","citing_title":"REPOT: Recoverable Program-of-Thought via Checkpoint Repair","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2509.02547","citing_title":"The Landscape of Agentic Reinforcement Learning for LLMs: A Survey","ref_index":117,"is_internal_anchor":false},{"citing_arxiv_id":"2601.19924","citing_title":"OPT-Engine: Benchmarking the Limits of LLMs in Optimization Modeling via Complexity Scaling","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JU6VNNPPO2XRJ7RC6QCRWODCPK","json":"https://pith.science/pith/JU6VNNPPO2XRJ7RC6QCRWODCPK.json","graph_json":"https://pith.science/api/pith-number/JU6VNNPPO2XRJ7RC6QCRWODCPK/graph.json","events_json":"https://pith.science/api/pith-number/JU6VNNPPO2XRJ7RC6QCRWODCPK/events.json","paper":"https://pith.science/paper/JU6VNNPP"},"agent_actions":{"view_html":"https://pith.science/pith/JU6VNNPPO2XRJ7RC6QCRWODCPK","download_json":"https://pith.science/pith/JU6VNNPPO2XRJ7RC6QCRWODCPK.json","view_paper":"https://pith.science/paper/JU6VNNPP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.17699&json=true","fetch_graph":"https://pith.science/api/pith-number/JU6VNNPPO2XRJ7RC6QCRWODCPK/graph.json","fetch_events":"https://pith.science/api/pith-number/JU6VNNPPO2XRJ7RC6QCRWODCPK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JU6VNNPPO2XRJ7RC6QCRWODCPK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JU6VNNPPO2XRJ7RC6QCRWODCPK/action/storage_attestation","attest_author":"https://pith.science/pith/JU6VNNPPO2XRJ7RC6QCRWODCPK/action/author_attestation","sign_citation":"https://pith.science/pith/JU6VNNPPO2XRJ7RC6QCRWODCPK/action/citation_signature","submit_replication":"https://pith.science/pith/JU6VNNPPO2XRJ7RC6QCRWODCPK/action/replication_record"}},"created_at":"2026-07-05T11:42:12.001830+00:00","updated_at":"2026-07-05T11:42:12.001830+00:00"}