{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FLJD5XYA4URSMS7BGRT62EYM4D","short_pith_number":"pith:FLJD5XYA","schema_version":"1.0","canonical_sha256":"2ad23edf00e523264be13467ed130ce0e1bc2048f752e252037f0a89fdf8159b","source":{"kind":"arxiv","id":"2503.13356","version":1},"attestation_state":"computed","paper":{"title":"Agents Play Thousands of 3D Video Games","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Liang Wang, Qiang Fu, Siyi Li, Tao Yu, Wei Yang, Xianliang Wang, Zhongwen Xu","submitted_at":"2025-03-17T16:42:34Z","abstract_excerpt":"We present PORTAL, a novel framework for developing artificial intelligence agents capable of playing thousands of 3D video games through language-guided policy generation. By transforming decision-making problems into language modeling tasks, our approach leverages large language models (LLMs) to generate behavior trees represented in domain-specific language (DSL). This method eliminates the computational burden associated with traditional reinforcement learning approaches while preserving strategic depth and rapid adaptability. Our framework introduces a hybrid policy structure that combine"},"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":"2503.13356","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-17T16:42:34Z","cross_cats_sorted":[],"title_canon_sha256":"be7acb52d30ea4095f44c823960daf500d49ea832e083de9d4b1b6a140c7e8b3","abstract_canon_sha256":"fe40e1d25f542ceac30c5f362cb29d153679d402588fe344778ea1aa5c95f9f3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:32:58.990923Z","signature_b64":"i7Xaju2LZdMLmoHoHeLUEkeKCvIQROkj191S9cCAHd95O7KRzxle2eeTh1ovUN69MyHiE/zKwJyxQAQfEzmlBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2ad23edf00e523264be13467ed130ce0e1bc2048f752e252037f0a89fdf8159b","last_reissued_at":"2026-07-05T10:32:58.990089Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:32:58.990089Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Agents Play Thousands of 3D Video Games","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Liang Wang, Qiang Fu, Siyi Li, Tao Yu, Wei Yang, Xianliang Wang, Zhongwen Xu","submitted_at":"2025-03-17T16:42:34Z","abstract_excerpt":"We present PORTAL, a novel framework for developing artificial intelligence agents capable of playing thousands of 3D video games through language-guided policy generation. By transforming decision-making problems into language modeling tasks, our approach leverages large language models (LLMs) to generate behavior trees represented in domain-specific language (DSL). This method eliminates the computational burden associated with traditional reinforcement learning approaches while preserving strategic depth and rapid adaptability. Our framework introduces a hybrid policy structure that combine"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.13356","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/2503.13356/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":"2503.13356","created_at":"2026-07-05T10:32:58.990195+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.13356v1","created_at":"2026-07-05T10:32:58.990195+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.13356","created_at":"2026-07-05T10:32:58.990195+00:00"},{"alias_kind":"pith_short_12","alias_value":"FLJD5XYA4URS","created_at":"2026-07-05T10:32:58.990195+00:00"},{"alias_kind":"pith_short_16","alias_value":"FLJD5XYA4URSMS7B","created_at":"2026-07-05T10:32:58.990195+00:00"},{"alias_kind":"pith_short_8","alias_value":"FLJD5XYA","created_at":"2026-07-05T10:32:58.990195+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01709","citing_title":"COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20987","citing_title":"Co-Evolving LLM Decision and Skill Bank Agents for Long-Horizon Tasks","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FLJD5XYA4URSMS7BGRT62EYM4D","json":"https://pith.science/pith/FLJD5XYA4URSMS7BGRT62EYM4D.json","graph_json":"https://pith.science/api/pith-number/FLJD5XYA4URSMS7BGRT62EYM4D/graph.json","events_json":"https://pith.science/api/pith-number/FLJD5XYA4URSMS7BGRT62EYM4D/events.json","paper":"https://pith.science/paper/FLJD5XYA"},"agent_actions":{"view_html":"https://pith.science/pith/FLJD5XYA4URSMS7BGRT62EYM4D","download_json":"https://pith.science/pith/FLJD5XYA4URSMS7BGRT62EYM4D.json","view_paper":"https://pith.science/paper/FLJD5XYA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.13356&json=true","fetch_graph":"https://pith.science/api/pith-number/FLJD5XYA4URSMS7BGRT62EYM4D/graph.json","fetch_events":"https://pith.science/api/pith-number/FLJD5XYA4URSMS7BGRT62EYM4D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FLJD5XYA4URSMS7BGRT62EYM4D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FLJD5XYA4URSMS7BGRT62EYM4D/action/storage_attestation","attest_author":"https://pith.science/pith/FLJD5XYA4URSMS7BGRT62EYM4D/action/author_attestation","sign_citation":"https://pith.science/pith/FLJD5XYA4URSMS7BGRT62EYM4D/action/citation_signature","submit_replication":"https://pith.science/pith/FLJD5XYA4URSMS7BGRT62EYM4D/action/replication_record"}},"created_at":"2026-07-05T10:32:58.990195+00:00","updated_at":"2026-07-05T10:32:58.990195+00:00"}