{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Z7J2DJGHM3PFYXSJNMT2TDQYLF","short_pith_number":"pith:Z7J2DJGH","schema_version":"1.0","canonical_sha256":"cfd3a1a4c766de5c5e496b27a98e185947ca7d91310a45fa75da2f4fd3c73b76","source":{"kind":"arxiv","id":"2502.07393","version":1},"attestation_state":"computed","paper":{"title":"FinRL-DeepSeek: LLM-Infused Risk-Sensitive Reinforcement Learning for Trading Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"q-fin.TR","authors_text":"Mostapha Benhenda (LAGA)","submitted_at":"2025-02-11T09:23:14Z","abstract_excerpt":"This paper presents a novel risk-sensitive trading agent combining reinforcement learning and large language models (LLMs). We extend the Conditional Value-at-Risk Proximal Policy Optimization (CPPO) algorithm, by adding risk assessment and trading recommendation signals generated by a LLM from financial news. Our approach is backtested on the Nasdaq-100 index benchmark, using financial news data from the FNSPID dataset and the DeepSeek V3, Qwen 2.5 and Llama 3.3 language models. The code, data, and trading agents are available at: https://github.com/benstaf/FinRL_DeepSeek"},"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":"2502.07393","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"q-fin.TR","submitted_at":"2025-02-11T09:23:14Z","cross_cats_sorted":[],"title_canon_sha256":"81c0a392de51026d438d9a8dfb968da091a8bfd02e4da59a9821ad0fa431ddde","abstract_canon_sha256":"9ccf521f2962095768b7e71995ab2f92137d5f42b7c6e557f0dfcac5be054c89"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:12:36.799897Z","signature_b64":"gKg2222Vki+ROjz/LMsTu4RqKxyAvBelDwTXw6Rue4RBjCr4+uESCzWopZe/G8X4UwSt79JmIRm8SiooPcb2AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cfd3a1a4c766de5c5e496b27a98e185947ca7d91310a45fa75da2f4fd3c73b76","last_reissued_at":"2026-07-05T10:12:36.799423Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:12:36.799423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FinRL-DeepSeek: LLM-Infused Risk-Sensitive Reinforcement Learning for Trading Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"q-fin.TR","authors_text":"Mostapha Benhenda (LAGA)","submitted_at":"2025-02-11T09:23:14Z","abstract_excerpt":"This paper presents a novel risk-sensitive trading agent combining reinforcement learning and large language models (LLMs). We extend the Conditional Value-at-Risk Proximal Policy Optimization (CPPO) algorithm, by adding risk assessment and trading recommendation signals generated by a LLM from financial news. Our approach is backtested on the Nasdaq-100 index benchmark, using financial news data from the FNSPID dataset and the DeepSeek V3, Qwen 2.5 and Llama 3.3 language models. The code, data, and trading agents are available at: https://github.com/benstaf/FinRL_DeepSeek"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.07393","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/2502.07393/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":"2502.07393","created_at":"2026-07-05T10:12:36.799496+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.07393v1","created_at":"2026-07-05T10:12:36.799496+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.07393","created_at":"2026-07-05T10:12:36.799496+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z7J2DJGHM3PF","created_at":"2026-07-05T10:12:36.799496+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z7J2DJGHM3PFYXSJ","created_at":"2026-07-05T10:12:36.799496+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z7J2DJGH","created_at":"2026-07-05T10:12:36.799496+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08285","citing_title":"Beyond Agent Architecture: Execution Assumptions and Reproducibility in LLM-Based Trading Systems","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06730","citing_title":"Semantic State Abstraction Interfaces for LLM-Augmented Portfolio Decisions: Multi-Axis News Decomposition and RL Diagnostics","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z7J2DJGHM3PFYXSJNMT2TDQYLF","json":"https://pith.science/pith/Z7J2DJGHM3PFYXSJNMT2TDQYLF.json","graph_json":"https://pith.science/api/pith-number/Z7J2DJGHM3PFYXSJNMT2TDQYLF/graph.json","events_json":"https://pith.science/api/pith-number/Z7J2DJGHM3PFYXSJNMT2TDQYLF/events.json","paper":"https://pith.science/paper/Z7J2DJGH"},"agent_actions":{"view_html":"https://pith.science/pith/Z7J2DJGHM3PFYXSJNMT2TDQYLF","download_json":"https://pith.science/pith/Z7J2DJGHM3PFYXSJNMT2TDQYLF.json","view_paper":"https://pith.science/paper/Z7J2DJGH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.07393&json=true","fetch_graph":"https://pith.science/api/pith-number/Z7J2DJGHM3PFYXSJNMT2TDQYLF/graph.json","fetch_events":"https://pith.science/api/pith-number/Z7J2DJGHM3PFYXSJNMT2TDQYLF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z7J2DJGHM3PFYXSJNMT2TDQYLF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z7J2DJGHM3PFYXSJNMT2TDQYLF/action/storage_attestation","attest_author":"https://pith.science/pith/Z7J2DJGHM3PFYXSJNMT2TDQYLF/action/author_attestation","sign_citation":"https://pith.science/pith/Z7J2DJGHM3PFYXSJNMT2TDQYLF/action/citation_signature","submit_replication":"https://pith.science/pith/Z7J2DJGHM3PFYXSJNMT2TDQYLF/action/replication_record"}},"created_at":"2026-07-05T10:12:36.799496+00:00","updated_at":"2026-07-05T10:12:36.799496+00:00"}