{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PW2A7SD56RIEIMYNDRUP754JL5","short_pith_number":"pith:PW2A7SD5","schema_version":"1.0","canonical_sha256":"7db40fc87df45044330d1c68fff7895f5a2d5076cd59ac94cb670d4682af8c4e","source":{"kind":"arxiv","id":"2507.12666","version":1},"attestation_state":"computed","paper":{"title":"Fly, Fail, Fix: Iterative Game Repair with Reinforcement Learning and Large Multimodal Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Alex Zook, Jonathan Tremblay, Josef Spjut","submitted_at":"2025-07-16T22:45:40Z","abstract_excerpt":"Game design hinges on understanding how static rules and content translate into dynamic player behavior - something modern generative systems that inspect only a game's code or assets struggle to capture. We present an automated design iteration framework that closes this gap by pairing a reinforcement learning (RL) agent, which playtests the game, with a large multimodal model (LMM), which revises the game based on what the agent does. In each loop the RL player completes several episodes, producing (i) numerical play metrics and/or (ii) a compact image strip summarising recent video frames. "},"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":"2507.12666","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-07-16T22:45:40Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8b6d5d92f47229d3bc3469577e028100da688138fd210606fff87c836423a871","abstract_canon_sha256":"8793f59fe8fb0f1a8d38dd9281f96f45c0dd79a4f99ef70a0fcde79798881ee3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:36.875259Z","signature_b64":"P13by4WLRgzKrDpIj0cohjtlm7Lt09XxOA722ifBBLgu3BLtxcYShQpqoTq6PmCeqbIMVfDDqzgOHq2iZqR8DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7db40fc87df45044330d1c68fff7895f5a2d5076cd59ac94cb670d4682af8c4e","last_reissued_at":"2026-07-05T11:38:36.874722Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:36.874722Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fly, Fail, Fix: Iterative Game Repair with Reinforcement Learning and Large Multimodal Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Alex Zook, Jonathan Tremblay, Josef Spjut","submitted_at":"2025-07-16T22:45:40Z","abstract_excerpt":"Game design hinges on understanding how static rules and content translate into dynamic player behavior - something modern generative systems that inspect only a game's code or assets struggle to capture. We present an automated design iteration framework that closes this gap by pairing a reinforcement learning (RL) agent, which playtests the game, with a large multimodal model (LMM), which revises the game based on what the agent does. In each loop the RL player completes several episodes, producing (i) numerical play metrics and/or (ii) a compact image strip summarising recent video frames. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.12666","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/2507.12666/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":"2507.12666","created_at":"2026-07-05T11:38:36.874786+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.12666v1","created_at":"2026-07-05T11:38:36.874786+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.12666","created_at":"2026-07-05T11:38:36.874786+00:00"},{"alias_kind":"pith_short_12","alias_value":"PW2A7SD56RIE","created_at":"2026-07-05T11:38:36.874786+00:00"},{"alias_kind":"pith_short_16","alias_value":"PW2A7SD56RIEIMYN","created_at":"2026-07-05T11:38:36.874786+00:00"},{"alias_kind":"pith_short_8","alias_value":"PW2A7SD5","created_at":"2026-07-05T11:38:36.874786+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/PW2A7SD56RIEIMYNDRUP754JL5","json":"https://pith.science/pith/PW2A7SD56RIEIMYNDRUP754JL5.json","graph_json":"https://pith.science/api/pith-number/PW2A7SD56RIEIMYNDRUP754JL5/graph.json","events_json":"https://pith.science/api/pith-number/PW2A7SD56RIEIMYNDRUP754JL5/events.json","paper":"https://pith.science/paper/PW2A7SD5"},"agent_actions":{"view_html":"https://pith.science/pith/PW2A7SD56RIEIMYNDRUP754JL5","download_json":"https://pith.science/pith/PW2A7SD56RIEIMYNDRUP754JL5.json","view_paper":"https://pith.science/paper/PW2A7SD5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.12666&json=true","fetch_graph":"https://pith.science/api/pith-number/PW2A7SD56RIEIMYNDRUP754JL5/graph.json","fetch_events":"https://pith.science/api/pith-number/PW2A7SD56RIEIMYNDRUP754JL5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PW2A7SD56RIEIMYNDRUP754JL5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PW2A7SD56RIEIMYNDRUP754JL5/action/storage_attestation","attest_author":"https://pith.science/pith/PW2A7SD56RIEIMYNDRUP754JL5/action/author_attestation","sign_citation":"https://pith.science/pith/PW2A7SD56RIEIMYNDRUP754JL5/action/citation_signature","submit_replication":"https://pith.science/pith/PW2A7SD56RIEIMYNDRUP754JL5/action/replication_record"}},"created_at":"2026-07-05T11:38:36.874786+00:00","updated_at":"2026-07-05T11:38:36.874786+00:00"}