{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:XQU2KGED6YLCC35QX2TNQHGXFG","short_pith_number":"pith:XQU2KGED","schema_version":"1.0","canonical_sha256":"bc29a51883f616216fb0bea6d81cd729b1cfa2c4428fd8d39b7a7c36430be89a","source":{"kind":"arxiv","id":"2309.11489","version":3},"attestation_state":"computed","paper":{"title":"Text2Reward: Reward Shaping with Language Models for Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.RO"],"primary_cat":"cs.LG","authors_text":"Chen Henry Wu, Qian Luo, Siheng Zhao, Tao Yu, Tianbao Xie, Victor Zhong, Yanchao Yang, Yitao Liu","submitted_at":"2023-09-20T17:39:13Z","abstract_excerpt":"Designing reward functions is a longstanding challenge in reinforcement learning (RL); it requires specialized knowledge or domain data, leading to high costs for development. To address this, we introduce Text2Reward, a data-free framework that automates the generation and shaping of dense reward functions based on large language models (LLMs). Given a goal described in natural language, Text2Reward generates shaped dense reward functions as an executable program grounded in a compact representation of the environment. Unlike inverse RL and recent work that uses LLMs to write sparse reward co"},"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":"2309.11489","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-09-20T17:39:13Z","cross_cats_sorted":["cs.AI","cs.CL","cs.RO"],"title_canon_sha256":"af16c92cf8c86dde30d5c540836063fc23dd44d85a8966f5b62c1b7b019b1582","abstract_canon_sha256":"a78eb14deb35082812d65bf62980abddadce2e6c2507732eac256d48b834923f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:23:12.144110Z","signature_b64":"CxXupDUTGQNiacsHLZTjjTTTMYPhhtCBsXoZQwle/SD2QVQ+/2zQ7z/wLsBPTmUnfBOy2dDkCrlB+3NkfafNDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bc29a51883f616216fb0bea6d81cd729b1cfa2c4428fd8d39b7a7c36430be89a","last_reissued_at":"2026-07-05T08:23:12.143612Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:23:12.143612Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Text2Reward: Reward Shaping with Language Models for Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.RO"],"primary_cat":"cs.LG","authors_text":"Chen Henry Wu, Qian Luo, Siheng Zhao, Tao Yu, Tianbao Xie, Victor Zhong, Yanchao Yang, Yitao Liu","submitted_at":"2023-09-20T17:39:13Z","abstract_excerpt":"Designing reward functions is a longstanding challenge in reinforcement learning (RL); it requires specialized knowledge or domain data, leading to high costs for development. To address this, we introduce Text2Reward, a data-free framework that automates the generation and shaping of dense reward functions based on large language models (LLMs). Given a goal described in natural language, Text2Reward generates shaped dense reward functions as an executable program grounded in a compact representation of the environment. Unlike inverse RL and recent work that uses LLMs to write sparse reward co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.11489","kind":"arxiv","version":3},"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/2309.11489/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":"2309.11489","created_at":"2026-07-05T08:23:12.143681+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.11489v3","created_at":"2026-07-05T08:23:12.143681+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.11489","created_at":"2026-07-05T08:23:12.143681+00:00"},{"alias_kind":"pith_short_12","alias_value":"XQU2KGED6YLC","created_at":"2026-07-05T08:23:12.143681+00:00"},{"alias_kind":"pith_short_16","alias_value":"XQU2KGED6YLCC35Q","created_at":"2026-07-05T08:23:12.143681+00:00"},{"alias_kind":"pith_short_8","alias_value":"XQU2KGED","created_at":"2026-07-05T08:23:12.143681+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":14,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05391","citing_title":"LLM-as-a-Verifier: A General-Purpose Verification Framework","ref_index":82,"is_internal_anchor":true},{"citing_arxiv_id":"2606.21943","citing_title":"Modularized Reinforcement Learning on LLMs: From MDP Creation to Exploration and Learning","ref_index":235,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17362","citing_title":"DriveJudge: Rethinking Autonomous Driving Evaluation with Vision-Language Models","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00083","citing_title":"From Demonstrations to Rewards: Test-Time Prompt Optimization for VLM Reward Models","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2504.01990","citing_title":"Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems","ref_index":133,"is_internal_anchor":false},{"citing_arxiv_id":"2603.12145","citing_title":"Automatic Generation of High-Performance RL Environments","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2509.16615","citing_title":"LLM-Guided Task- and Affordance-Level Exploration in Reinforcement Learning","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19773","citing_title":"PR-CAD: Progressive Refinement for Unified Controllable and Faithful Text-to-CAD Generation with Large Language Models","ref_index":19,"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":91,"is_internal_anchor":false},{"citing_arxiv_id":"2307.05973","citing_title":"VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11859","citing_title":"EvoNav: Evolutionary Reward Function Design for Robot Navigation with Large Language Models","ref_index":45,"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":91,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02073","citing_title":"Enhanced LLM Reasoning by Optimizing Reward Functions with Search-Driven Reinforcement Learning","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02073","citing_title":"Enhanced LLM Reasoning by Optimizing Reward Functions with Search-Driven Reinforcement Learning","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XQU2KGED6YLCC35QX2TNQHGXFG","json":"https://pith.science/pith/XQU2KGED6YLCC35QX2TNQHGXFG.json","graph_json":"https://pith.science/api/pith-number/XQU2KGED6YLCC35QX2TNQHGXFG/graph.json","events_json":"https://pith.science/api/pith-number/XQU2KGED6YLCC35QX2TNQHGXFG/events.json","paper":"https://pith.science/paper/XQU2KGED"},"agent_actions":{"view_html":"https://pith.science/pith/XQU2KGED6YLCC35QX2TNQHGXFG","download_json":"https://pith.science/pith/XQU2KGED6YLCC35QX2TNQHGXFG.json","view_paper":"https://pith.science/paper/XQU2KGED","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.11489&json=true","fetch_graph":"https://pith.science/api/pith-number/XQU2KGED6YLCC35QX2TNQHGXFG/graph.json","fetch_events":"https://pith.science/api/pith-number/XQU2KGED6YLCC35QX2TNQHGXFG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XQU2KGED6YLCC35QX2TNQHGXFG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XQU2KGED6YLCC35QX2TNQHGXFG/action/storage_attestation","attest_author":"https://pith.science/pith/XQU2KGED6YLCC35QX2TNQHGXFG/action/author_attestation","sign_citation":"https://pith.science/pith/XQU2KGED6YLCC35QX2TNQHGXFG/action/citation_signature","submit_replication":"https://pith.science/pith/XQU2KGED6YLCC35QX2TNQHGXFG/action/replication_record"}},"created_at":"2026-07-05T08:23:12.143681+00:00","updated_at":"2026-07-05T08:23:12.143681+00:00"}