{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2J6365YO7ONFABOU7WGW7QOLOU","short_pith_number":"pith:2J6365YO","schema_version":"1.0","canonical_sha256":"d27dbf770efb9a5005d4fd8d6fc1cb750533b86356b59f8c278305882741a9c4","source":{"kind":"arxiv","id":"2502.12468","version":2},"attestation_state":"computed","paper":{"title":"MCTS-Judge: Test-Time Scaling in LLM-as-a-Judge for Code Correctness Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chaoqun Yang, Guillaume Sartoretti, Jiaoyang Li, Kaixin Li, Ming Hu, Pengliang Ji, Yutong Wang","submitted_at":"2025-02-18T02:55:48Z","abstract_excerpt":"The LLM-as-a-Judge paradigm shows promise for evaluating generative content but lacks reliability in reasoning-intensive scenarios, such as programming. Inspired by recent advances in reasoning models and shifts in scaling laws, we pioneer bringing test-time computation into LLM-as-a-Judge, proposing MCTS-Judge, a resource-efficient, System-2 thinking framework for code correctness evaluation. MCTS-Judge leverages Monte Carlo Tree Search (MCTS) to decompose problems into simpler, multi-perspective evaluations. Through a node-selection strategy that combines self-assessment based on historical "},"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":"2502.12468","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-18T02:55:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5427eb9c6693adb461e729b4e881182c29caa215e2cd73bc5d2a7401fee5ee26","abstract_canon_sha256":"b0e05323ce9d6eb80d8abb629bf90e46c013d168207eb98903b3acd8bceb50d6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:26.763364Z","signature_b64":"+oLclM0vzqnslyBWDU1pOVEcQvkZ2jTo94Uy8107F/leRgOR1Ysvi5LqYtm8FuzEdH+IwdtTxEKoisLsGjTNBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d27dbf770efb9a5005d4fd8d6fc1cb750533b86356b59f8c278305882741a9c4","last_reissued_at":"2026-05-28T01:04:26.762842Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:26.762842Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MCTS-Judge: Test-Time Scaling in LLM-as-a-Judge for Code Correctness Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chaoqun Yang, Guillaume Sartoretti, Jiaoyang Li, Kaixin Li, Ming Hu, Pengliang Ji, Yutong Wang","submitted_at":"2025-02-18T02:55:48Z","abstract_excerpt":"The LLM-as-a-Judge paradigm shows promise for evaluating generative content but lacks reliability in reasoning-intensive scenarios, such as programming. Inspired by recent advances in reasoning models and shifts in scaling laws, we pioneer bringing test-time computation into LLM-as-a-Judge, proposing MCTS-Judge, a resource-efficient, System-2 thinking framework for code correctness evaluation. MCTS-Judge leverages Monte Carlo Tree Search (MCTS) to decompose problems into simpler, multi-perspective evaluations. Through a node-selection strategy that combines self-assessment based on historical "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.12468","kind":"arxiv","version":2},"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/2502.12468/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":"2502.12468","created_at":"2026-05-28T01:04:26.762897+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.12468v2","created_at":"2026-05-28T01:04:26.762897+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.12468","created_at":"2026-05-28T01:04:26.762897+00:00"},{"alias_kind":"pith_short_12","alias_value":"2J6365YO7ONF","created_at":"2026-05-28T01:04:26.762897+00:00"},{"alias_kind":"pith_short_16","alias_value":"2J6365YO7ONFABOU","created_at":"2026-05-28T01:04:26.762897+00:00"},{"alias_kind":"pith_short_8","alias_value":"2J6365YO","created_at":"2026-05-28T01:04:26.762897+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2605.21748","citing_title":"RankJudge: A Multi-Turn LLM-as-a-Judge Synthetic Benchmark Generator","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2604.27727","citing_title":"LLM-as-a-Judge for Human-AI Co-Creation: A Reliability-Aware Evaluation Framework for Coding","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2J6365YO7ONFABOU7WGW7QOLOU","json":"https://pith.science/pith/2J6365YO7ONFABOU7WGW7QOLOU.json","graph_json":"https://pith.science/api/pith-number/2J6365YO7ONFABOU7WGW7QOLOU/graph.json","events_json":"https://pith.science/api/pith-number/2J6365YO7ONFABOU7WGW7QOLOU/events.json","paper":"https://pith.science/paper/2J6365YO"},"agent_actions":{"view_html":"https://pith.science/pith/2J6365YO7ONFABOU7WGW7QOLOU","download_json":"https://pith.science/pith/2J6365YO7ONFABOU7WGW7QOLOU.json","view_paper":"https://pith.science/paper/2J6365YO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.12468&json=true","fetch_graph":"https://pith.science/api/pith-number/2J6365YO7ONFABOU7WGW7QOLOU/graph.json","fetch_events":"https://pith.science/api/pith-number/2J6365YO7ONFABOU7WGW7QOLOU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2J6365YO7ONFABOU7WGW7QOLOU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2J6365YO7ONFABOU7WGW7QOLOU/action/storage_attestation","attest_author":"https://pith.science/pith/2J6365YO7ONFABOU7WGW7QOLOU/action/author_attestation","sign_citation":"https://pith.science/pith/2J6365YO7ONFABOU7WGW7QOLOU/action/citation_signature","submit_replication":"https://pith.science/pith/2J6365YO7ONFABOU7WGW7QOLOU/action/replication_record"}},"created_at":"2026-05-28T01:04:26.762897+00:00","updated_at":"2026-05-28T01:04:26.762897+00:00"}