{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2ICTSO677GBWF4TFK6U5CJ6DA7","short_pith_number":"pith:2ICTSO67","schema_version":"1.0","canonical_sha256":"d205393bdff98362f26557a9d127c307f574267cd377dfcb23fc8b6cda20b3ec","source":{"kind":"arxiv","id":"2403.12014","version":2},"attestation_state":"computed","paper":{"title":"EnvGen: Generating and Adapting Environments via LLMs for Training Embodied Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Abhay Zala, Han Lin, Jaehong Yoon, Jaemin Cho, Mohit Bansal","submitted_at":"2024-03-18T17:51:16Z","abstract_excerpt":"Recent SOTA approaches for embodied learning via interaction directly employ large language models (LLMs) as agents to determine the next steps in an environment. Due to their world knowledge and reasoning capabilities, LLM agents achieve stronger performance than previous smaller agents based on reinforcement learning (RL); however, frequently calling LLMs is slow and expensive. Instead of directly employing LLMs as agents, can we use LLMs' reasoning capabilities to adaptively create training environments to help smaller RL agents learn useful skills that they are weak at? We propose EnvGen, "},"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":"2403.12014","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-18T17:51:16Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"c8569152b1d6380824b36b03b3be2d1f4b40212bac443763956bc2b214ce5229","abstract_canon_sha256":"8b236684f12e5107495f58eec1be5a4c1f4ead6bd6c7571a6c4bf1c9d11a3f61"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:43:09.356844Z","signature_b64":"hFpFDMVjW4sSrK0lzr9h+HiKKiSpaZxBmjeAap8zTGH6qyuiIwc9/anMMLHdTjhEn80GTLHRorolQVzue0RnBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d205393bdff98362f26557a9d127c307f574267cd377dfcb23fc8b6cda20b3ec","last_reissued_at":"2026-07-05T08:43:09.356342Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:43:09.356342Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EnvGen: Generating and Adapting Environments via LLMs for Training Embodied Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Abhay Zala, Han Lin, Jaehong Yoon, Jaemin Cho, Mohit Bansal","submitted_at":"2024-03-18T17:51:16Z","abstract_excerpt":"Recent SOTA approaches for embodied learning via interaction directly employ large language models (LLMs) as agents to determine the next steps in an environment. Due to their world knowledge and reasoning capabilities, LLM agents achieve stronger performance than previous smaller agents based on reinforcement learning (RL); however, frequently calling LLMs is slow and expensive. Instead of directly employing LLMs as agents, can we use LLMs' reasoning capabilities to adaptively create training environments to help smaller RL agents learn useful skills that they are weak at? We propose EnvGen, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.12014","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/2403.12014/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":"2403.12014","created_at":"2026-07-05T08:43:09.356411+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.12014v2","created_at":"2026-07-05T08:43:09.356411+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.12014","created_at":"2026-07-05T08:43:09.356411+00:00"},{"alias_kind":"pith_short_12","alias_value":"2ICTSO677GBW","created_at":"2026-07-05T08:43:09.356411+00:00"},{"alias_kind":"pith_short_16","alias_value":"2ICTSO677GBWF4TF","created_at":"2026-07-05T08:43:09.356411+00:00"},{"alias_kind":"pith_short_8","alias_value":"2ICTSO67","created_at":"2026-07-05T08:43:09.356411+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23049","citing_title":"PhoneBuddy: Training Open Models for Agentic Phone Use","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29486","citing_title":"PhoneWorld: Scaling Phone-Use Agent Environments","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12191","citing_title":"Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Application","ref_index":175,"is_internal_anchor":false},{"citing_arxiv_id":"2504.02605","citing_title":"Multi-SWE-bench: A Multilingual Benchmark for Issue Resolving","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09423","citing_title":"SimWorld Studio: Automatic Environment Generation with Evolving Coding Agent for Embodied Agent Learning","ref_index":101,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09423","citing_title":"SimWorld Studio: Automatic Environment Generation with Evolving Coding Agent for Embodied Agent Learning","ref_index":101,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18292","citing_title":"Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence","ref_index":125,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2ICTSO677GBWF4TFK6U5CJ6DA7","json":"https://pith.science/pith/2ICTSO677GBWF4TFK6U5CJ6DA7.json","graph_json":"https://pith.science/api/pith-number/2ICTSO677GBWF4TFK6U5CJ6DA7/graph.json","events_json":"https://pith.science/api/pith-number/2ICTSO677GBWF4TFK6U5CJ6DA7/events.json","paper":"https://pith.science/paper/2ICTSO67"},"agent_actions":{"view_html":"https://pith.science/pith/2ICTSO677GBWF4TFK6U5CJ6DA7","download_json":"https://pith.science/pith/2ICTSO677GBWF4TFK6U5CJ6DA7.json","view_paper":"https://pith.science/paper/2ICTSO67","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.12014&json=true","fetch_graph":"https://pith.science/api/pith-number/2ICTSO677GBWF4TFK6U5CJ6DA7/graph.json","fetch_events":"https://pith.science/api/pith-number/2ICTSO677GBWF4TFK6U5CJ6DA7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2ICTSO677GBWF4TFK6U5CJ6DA7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2ICTSO677GBWF4TFK6U5CJ6DA7/action/storage_attestation","attest_author":"https://pith.science/pith/2ICTSO677GBWF4TFK6U5CJ6DA7/action/author_attestation","sign_citation":"https://pith.science/pith/2ICTSO677GBWF4TFK6U5CJ6DA7/action/citation_signature","submit_replication":"https://pith.science/pith/2ICTSO677GBWF4TFK6U5CJ6DA7/action/replication_record"}},"created_at":"2026-07-05T08:43:09.356411+00:00","updated_at":"2026-07-05T08:43:09.356411+00:00"}