{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CH74DGY3WYQGKIIB6OANKZ775V","short_pith_number":"pith:CH74DGY3","schema_version":"1.0","canonical_sha256":"11ffc19b1bb620652101f380d567ffed4f2f3212c2aa7364288da448baca13ef","source":{"kind":"arxiv","id":"2508.10428","version":1},"attestation_state":"computed","paper":{"title":"SC2Arena and StarEvolve: Benchmark and Self-Improvement Framework for LLMs in Complex Decision-Making Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Xu, Hao Hu, Lexiang Wang, Ni Mu, Pengbo Shen, Runpeng Xie, Senhao Yang, Shuang Xu, Yao Luan, Yaqing Wang, Yiqin Yang","submitted_at":"2025-08-14T07:58:01Z","abstract_excerpt":"Evaluating large language models (LLMs) in complex decision-making is essential for advancing AI's ability for strategic planning and real-time adaptation. However, existing benchmarks for tasks like StarCraft II fail to capture the game's full complexity, such as its complete game context, diverse action spaces, and all playable races. To address this gap, we present SC2Arena, a benchmark that fully supports all playable races, low-level action spaces, and optimizes text-based observations to tackle spatial reasoning challenges. Complementing this, we introduce StarEvolve, a hierarchical fram"},"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":"2508.10428","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-14T07:58:01Z","cross_cats_sorted":[],"title_canon_sha256":"23ac22c3baf3c77e68cadb237d556f40e0a7df836cf2325ef2a73fdf3c48f66d","abstract_canon_sha256":"7400e65682f2aa1ce087d43240adbf6170c93fd1072f8032685a73ec60d0e4f6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:56.049803Z","signature_b64":"qsAxDoulzsHV9qrMG7owMYCqtECMZVaVXpLUP5T+1E1Ik03aV4Yv/vMcDrmchD2d3Gpes9QqUqYfR8kx2T8+BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"11ffc19b1bb620652101f380d567ffed4f2f3212c2aa7364288da448baca13ef","last_reissued_at":"2026-07-05T11:53:56.049339Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:56.049339Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SC2Arena and StarEvolve: Benchmark and Self-Improvement Framework for LLMs in Complex Decision-Making Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Xu, Hao Hu, Lexiang Wang, Ni Mu, Pengbo Shen, Runpeng Xie, Senhao Yang, Shuang Xu, Yao Luan, Yaqing Wang, Yiqin Yang","submitted_at":"2025-08-14T07:58:01Z","abstract_excerpt":"Evaluating large language models (LLMs) in complex decision-making is essential for advancing AI's ability for strategic planning and real-time adaptation. However, existing benchmarks for tasks like StarCraft II fail to capture the game's full complexity, such as its complete game context, diverse action spaces, and all playable races. To address this gap, we present SC2Arena, a benchmark that fully supports all playable races, low-level action spaces, and optimizes text-based observations to tackle spatial reasoning challenges. Complementing this, we introduce StarEvolve, a hierarchical fram"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.10428","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/2508.10428/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":"2508.10428","created_at":"2026-07-05T11:53:56.049397+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.10428v1","created_at":"2026-07-05T11:53:56.049397+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.10428","created_at":"2026-07-05T11:53:56.049397+00:00"},{"alias_kind":"pith_short_12","alias_value":"CH74DGY3WYQG","created_at":"2026-07-05T11:53:56.049397+00:00"},{"alias_kind":"pith_short_16","alias_value":"CH74DGY3WYQGKIIB","created_at":"2026-07-05T11:53:56.049397+00:00"},{"alias_kind":"pith_short_8","alias_value":"CH74DGY3","created_at":"2026-07-05T11:53:56.049397+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18950","citing_title":"RTSGameBench: An RTS Benchmark for Strategic Reasoning by Vision-Language Models","ref_index":45,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CH74DGY3WYQGKIIB6OANKZ775V","json":"https://pith.science/pith/CH74DGY3WYQGKIIB6OANKZ775V.json","graph_json":"https://pith.science/api/pith-number/CH74DGY3WYQGKIIB6OANKZ775V/graph.json","events_json":"https://pith.science/api/pith-number/CH74DGY3WYQGKIIB6OANKZ775V/events.json","paper":"https://pith.science/paper/CH74DGY3"},"agent_actions":{"view_html":"https://pith.science/pith/CH74DGY3WYQGKIIB6OANKZ775V","download_json":"https://pith.science/pith/CH74DGY3WYQGKIIB6OANKZ775V.json","view_paper":"https://pith.science/paper/CH74DGY3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.10428&json=true","fetch_graph":"https://pith.science/api/pith-number/CH74DGY3WYQGKIIB6OANKZ775V/graph.json","fetch_events":"https://pith.science/api/pith-number/CH74DGY3WYQGKIIB6OANKZ775V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CH74DGY3WYQGKIIB6OANKZ775V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CH74DGY3WYQGKIIB6OANKZ775V/action/storage_attestation","attest_author":"https://pith.science/pith/CH74DGY3WYQGKIIB6OANKZ775V/action/author_attestation","sign_citation":"https://pith.science/pith/CH74DGY3WYQGKIIB6OANKZ775V/action/citation_signature","submit_replication":"https://pith.science/pith/CH74DGY3WYQGKIIB6OANKZ775V/action/replication_record"}},"created_at":"2026-07-05T11:53:56.049397+00:00","updated_at":"2026-07-05T11:53:56.049397+00:00"}