{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ZWSV27YJO65AYQGJYHOFP2CMUT","short_pith_number":"pith:ZWSV27YJ","schema_version":"1.0","canonical_sha256":"cda55d7f0977ba0c40c9c1dc57e84ca4f6e2bf3b4ef0e26f5475877c223ac8e7","source":{"kind":"arxiv","id":"2310.08235","version":2},"attestation_state":"computed","paper":{"title":"GROOT: Learning to Follow Instructions by Watching Gameplay Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Anji Liu, Bowei Zhang, Shaofei Cai, Xiaojian Ma, Yitao Liang, Zihao Wang","submitted_at":"2023-10-12T11:31:01Z","abstract_excerpt":"We study the problem of building a controller that can follow open-ended instructions in open-world environments. We propose to follow reference videos as instructions, which offer expressive goal specifications while eliminating the need for expensive text-gameplay annotations. A new learning framework is derived to allow learning such instruction-following controllers from gameplay videos while producing a video instruction encoder that induces a structured goal space. We implement our agent GROOT in a simple yet effective encoder-decoder architecture based on causal transformers. We evaluat"},"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":"2310.08235","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-10-12T11:31:01Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"67b3d8007c145582229561287283fb92883fdbf1bcf5a839538ec35cf4aa2ebb","abstract_canon_sha256":"1121ffa0efd68818a6a5e1e837627903247456379f4fc3686c90d48fb249d7b6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:17:55.092551Z","signature_b64":"Cd1LRiGabImKMmvD94xkquW7HReqXLwPviS/P80O/U313dBReoSlJ+galjZ+QPWAJW0V5QuTZJjprP2JgKbUBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cda55d7f0977ba0c40c9c1dc57e84ca4f6e2bf3b4ef0e26f5475877c223ac8e7","last_reissued_at":"2026-07-05T07:17:55.092070Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:17:55.092070Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GROOT: Learning to Follow Instructions by Watching Gameplay Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Anji Liu, Bowei Zhang, Shaofei Cai, Xiaojian Ma, Yitao Liang, Zihao Wang","submitted_at":"2023-10-12T11:31:01Z","abstract_excerpt":"We study the problem of building a controller that can follow open-ended instructions in open-world environments. We propose to follow reference videos as instructions, which offer expressive goal specifications while eliminating the need for expensive text-gameplay annotations. A new learning framework is derived to allow learning such instruction-following controllers from gameplay videos while producing a video instruction encoder that induces a structured goal space. We implement our agent GROOT in a simple yet effective encoder-decoder architecture based on causal transformers. We evaluat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.08235","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/2310.08235/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":"2310.08235","created_at":"2026-07-05T07:17:55.092113+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.08235v2","created_at":"2026-07-05T07:17:55.092113+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.08235","created_at":"2026-07-05T07:17:55.092113+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZWSV27YJO65A","created_at":"2026-07-05T07:17:55.092113+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZWSV27YJO65AYQGJ","created_at":"2026-07-05T07:17:55.092113+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZWSV27YJ","created_at":"2026-07-05T07:17:55.092113+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2302.01560","citing_title":"Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2509.24527","citing_title":"Training Agents Inside of Scalable World Models","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27578","citing_title":"World2Minecraft: Occupancy-Driven Simulated Scenes Construction","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZWSV27YJO65AYQGJYHOFP2CMUT","json":"https://pith.science/pith/ZWSV27YJO65AYQGJYHOFP2CMUT.json","graph_json":"https://pith.science/api/pith-number/ZWSV27YJO65AYQGJYHOFP2CMUT/graph.json","events_json":"https://pith.science/api/pith-number/ZWSV27YJO65AYQGJYHOFP2CMUT/events.json","paper":"https://pith.science/paper/ZWSV27YJ"},"agent_actions":{"view_html":"https://pith.science/pith/ZWSV27YJO65AYQGJYHOFP2CMUT","download_json":"https://pith.science/pith/ZWSV27YJO65AYQGJYHOFP2CMUT.json","view_paper":"https://pith.science/paper/ZWSV27YJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.08235&json=true","fetch_graph":"https://pith.science/api/pith-number/ZWSV27YJO65AYQGJYHOFP2CMUT/graph.json","fetch_events":"https://pith.science/api/pith-number/ZWSV27YJO65AYQGJYHOFP2CMUT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZWSV27YJO65AYQGJYHOFP2CMUT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZWSV27YJO65AYQGJYHOFP2CMUT/action/storage_attestation","attest_author":"https://pith.science/pith/ZWSV27YJO65AYQGJYHOFP2CMUT/action/author_attestation","sign_citation":"https://pith.science/pith/ZWSV27YJO65AYQGJYHOFP2CMUT/action/citation_signature","submit_replication":"https://pith.science/pith/ZWSV27YJO65AYQGJYHOFP2CMUT/action/replication_record"}},"created_at":"2026-07-05T07:17:55.092113+00:00","updated_at":"2026-07-05T07:17:55.092113+00:00"}