{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:DMDYFWUIYIWCENDR3YGL54DJDM","short_pith_number":"pith:DMDYFWUI","schema_version":"1.0","canonical_sha256":"1b0782da88c22c223471de0cbef0691b13783db573a56f0883e4313087028ad3","source":{"kind":"arxiv","id":"1910.08210","version":6},"attestation_state":"computed","paper":{"title":"RTFM: Generalising to Novel Environment Dynamics via Reading","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Edward Grefenstette, Tim Rockt\\\"aschel, Victor Zhong","submitted_at":"2019-10-18T00:49:15Z","abstract_excerpt":"Obtaining policies that can generalise to new environments in reinforcement learning is challenging. In this work, we demonstrate that language understanding via a reading policy learner is a promising vehicle for generalisation to new environments. We propose a grounded policy learning problem, Read to Fight Monsters (RTFM), in which the agent must jointly reason over a language goal, relevant dynamics described in a document, and environment observations. We procedurally generate environment dynamics and corresponding language descriptions of the dynamics, such that agents must read to under"},"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":"1910.08210","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-10-18T00:49:15Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"a1275882bece149fb2276a83d42cdcc3cbc4588ab8d668c915b4419fe22eaa8d","abstract_canon_sha256":"8b3f3632d943d93ab7b256f4041ea3345f278856036d10452c011e58a5791d3a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:11:20.150607Z","signature_b64":"P95ZSprbpds8xCjpmi9PUbp2AEC2qzB6Y+wdLDxI+R0TDfSybkvdrK6zsahVTLk2tEXgsNA+4UlwkYsLms09Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1b0782da88c22c223471de0cbef0691b13783db573a56f0883e4313087028ad3","last_reissued_at":"2026-07-05T02:11:20.150198Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:11:20.150198Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RTFM: Generalising to Novel Environment Dynamics via Reading","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Edward Grefenstette, Tim Rockt\\\"aschel, Victor Zhong","submitted_at":"2019-10-18T00:49:15Z","abstract_excerpt":"Obtaining policies that can generalise to new environments in reinforcement learning is challenging. In this work, we demonstrate that language understanding via a reading policy learner is a promising vehicle for generalisation to new environments. We propose a grounded policy learning problem, Read to Fight Monsters (RTFM), in which the agent must jointly reason over a language goal, relevant dynamics described in a document, and environment observations. We procedurally generate environment dynamics and corresponding language descriptions of the dynamics, such that agents must read to under"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.08210","kind":"arxiv","version":6},"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/1910.08210/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":"1910.08210","created_at":"2026-07-05T02:11:20.150251+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.08210v6","created_at":"2026-07-05T02:11:20.150251+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.08210","created_at":"2026-07-05T02:11:20.150251+00:00"},{"alias_kind":"pith_short_12","alias_value":"DMDYFWUIYIWC","created_at":"2026-07-05T02:11:20.150251+00:00"},{"alias_kind":"pith_short_16","alias_value":"DMDYFWUIYIWCENDR","created_at":"2026-07-05T02:11:20.150251+00:00"},{"alias_kind":"pith_short_8","alias_value":"DMDYFWUI","created_at":"2026-07-05T02:11:20.150251+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.20097","citing_title":"PSALM-V: Automating Symbolic Planning in Interactive Visual Environments with Large Language Models","ref_index":5,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DMDYFWUIYIWCENDR3YGL54DJDM","json":"https://pith.science/pith/DMDYFWUIYIWCENDR3YGL54DJDM.json","graph_json":"https://pith.science/api/pith-number/DMDYFWUIYIWCENDR3YGL54DJDM/graph.json","events_json":"https://pith.science/api/pith-number/DMDYFWUIYIWCENDR3YGL54DJDM/events.json","paper":"https://pith.science/paper/DMDYFWUI"},"agent_actions":{"view_html":"https://pith.science/pith/DMDYFWUIYIWCENDR3YGL54DJDM","download_json":"https://pith.science/pith/DMDYFWUIYIWCENDR3YGL54DJDM.json","view_paper":"https://pith.science/paper/DMDYFWUI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.08210&json=true","fetch_graph":"https://pith.science/api/pith-number/DMDYFWUIYIWCENDR3YGL54DJDM/graph.json","fetch_events":"https://pith.science/api/pith-number/DMDYFWUIYIWCENDR3YGL54DJDM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DMDYFWUIYIWCENDR3YGL54DJDM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DMDYFWUIYIWCENDR3YGL54DJDM/action/storage_attestation","attest_author":"https://pith.science/pith/DMDYFWUIYIWCENDR3YGL54DJDM/action/author_attestation","sign_citation":"https://pith.science/pith/DMDYFWUIYIWCENDR3YGL54DJDM/action/citation_signature","submit_replication":"https://pith.science/pith/DMDYFWUIYIWCENDR3YGL54DJDM/action/replication_record"}},"created_at":"2026-07-05T02:11:20.150251+00:00","updated_at":"2026-07-05T02:11:20.150251+00:00"}