{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RSYEZRO3CFLWGOZIDPE7POAPKY","short_pith_number":"pith:RSYEZRO3","schema_version":"1.0","canonical_sha256":"8cb04cc5db1157633b281bc9f7b80f5630fae0e5ecbd2437ed0ac23a536e8fc9","source":{"kind":"arxiv","id":"2505.07215","version":1},"attestation_state":"computed","paper":{"title":"Measuring General Intelligence with Generated Games","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Dan Klein, David Huang, Nicholas Tomlin, Vivek Verma, William Chen","submitted_at":"2025-05-12T04:01:03Z","abstract_excerpt":"We present gg-bench, a collection of game environments designed to evaluate general reasoning capabilities in language models. Unlike most static benchmarks, gg-bench is a data generating process where new evaluation instances can be generated at will. In particular, gg-bench is synthetically generated by (1) using a large language model (LLM) to generate natural language descriptions of novel games, (2) using the LLM to implement each game in code as a Gym environment, and (3) training reinforcement learning (RL) agents via self-play on the generated games. We evaluate language models by thei"},"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":"2505.07215","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-12T04:01:03Z","cross_cats_sorted":[],"title_canon_sha256":"616b497a8ec8cba82f099a2a2720b8c880e0e0e6a99c5b06c33cf22ab7dae392","abstract_canon_sha256":"cb2ca01159d589cdd882d7cf458fd2bb820386780ba5c40fd6c2c189d4d6e4fe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:01:44.443692Z","signature_b64":"mYRTBYpD0Q+QW0vKd9tjYQ2cVwtGo1VwNiZm9xY7zph+IIg60MuPGH50ueUiPcqZCErqLvRJszQNLz8SmNpbCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8cb04cc5db1157633b281bc9f7b80f5630fae0e5ecbd2437ed0ac23a536e8fc9","last_reissued_at":"2026-07-05T11:01:44.443206Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:01:44.443206Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Measuring General Intelligence with Generated Games","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Dan Klein, David Huang, Nicholas Tomlin, Vivek Verma, William Chen","submitted_at":"2025-05-12T04:01:03Z","abstract_excerpt":"We present gg-bench, a collection of game environments designed to evaluate general reasoning capabilities in language models. Unlike most static benchmarks, gg-bench is a data generating process where new evaluation instances can be generated at will. In particular, gg-bench is synthetically generated by (1) using a large language model (LLM) to generate natural language descriptions of novel games, (2) using the LLM to implement each game in code as a Gym environment, and (3) training reinforcement learning (RL) agents via self-play on the generated games. We evaluate language models by thei"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.07215","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/2505.07215/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":"2505.07215","created_at":"2026-07-05T11:01:44.443264+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.07215v1","created_at":"2026-07-05T11:01:44.443264+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.07215","created_at":"2026-07-05T11:01:44.443264+00:00"},{"alias_kind":"pith_short_12","alias_value":"RSYEZRO3CFLW","created_at":"2026-07-05T11:01:44.443264+00:00"},{"alias_kind":"pith_short_16","alias_value":"RSYEZRO3CFLWGOZI","created_at":"2026-07-05T11:01:44.443264+00:00"},{"alias_kind":"pith_short_8","alias_value":"RSYEZRO3","created_at":"2026-07-05T11:01:44.443264+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17657","citing_title":"Using Cognitive Models to Improve Language Model Simulation of Human Persuasion Games","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24375","citing_title":"Distilling Game Code World Model Generation into Lightweight Large Language Models","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23238","citing_title":"GENSTRAT: Toward a Science of Strategic Reasoning in Large Language Models","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18181","citing_title":"Scalable Environments Drive Generalizable Agents","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17928","citing_title":"HEALing Entropy Collapse: Enhancing Exploration in Few-Shot RLVR via Hybrid-Domain Entropy Dynamics Alignment","ref_index":56,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RSYEZRO3CFLWGOZIDPE7POAPKY","json":"https://pith.science/pith/RSYEZRO3CFLWGOZIDPE7POAPKY.json","graph_json":"https://pith.science/api/pith-number/RSYEZRO3CFLWGOZIDPE7POAPKY/graph.json","events_json":"https://pith.science/api/pith-number/RSYEZRO3CFLWGOZIDPE7POAPKY/events.json","paper":"https://pith.science/paper/RSYEZRO3"},"agent_actions":{"view_html":"https://pith.science/pith/RSYEZRO3CFLWGOZIDPE7POAPKY","download_json":"https://pith.science/pith/RSYEZRO3CFLWGOZIDPE7POAPKY.json","view_paper":"https://pith.science/paper/RSYEZRO3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.07215&json=true","fetch_graph":"https://pith.science/api/pith-number/RSYEZRO3CFLWGOZIDPE7POAPKY/graph.json","fetch_events":"https://pith.science/api/pith-number/RSYEZRO3CFLWGOZIDPE7POAPKY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RSYEZRO3CFLWGOZIDPE7POAPKY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RSYEZRO3CFLWGOZIDPE7POAPKY/action/storage_attestation","attest_author":"https://pith.science/pith/RSYEZRO3CFLWGOZIDPE7POAPKY/action/author_attestation","sign_citation":"https://pith.science/pith/RSYEZRO3CFLWGOZIDPE7POAPKY/action/citation_signature","submit_replication":"https://pith.science/pith/RSYEZRO3CFLWGOZIDPE7POAPKY/action/replication_record"}},"created_at":"2026-07-05T11:01:44.443264+00:00","updated_at":"2026-07-05T11:01:44.443264+00:00"}