{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YMVSPOFEH7CB6YNJU3LIJUV3WD","short_pith_number":"pith:YMVSPOFE","schema_version":"1.0","canonical_sha256":"c32b27b8a43fc41f61a9a6d684d2bbb0f4312e532f73f126b7528b7b635b4e19","source":{"kind":"arxiv","id":"2607.08837","version":1},"attestation_state":"computed","paper":{"title":"Prompt-Driven Exploration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"David Zhang, John Marangola, Nitish Dashora, Pulkit Agrawal, Raghuram Kowdeed, Richard Li, Ruiyang Luo, Sunshine Jiang, Zhang-Wei Hong","submitted_at":"2026-07-09T18:00:24Z","abstract_excerpt":"Exploration is essential to RL since a policy cannot improve by repeatedly sampling the behaviors it already prefers. Standard methods inject stochasticity in the action space, but such jitter only yields rollouts close to the original. Escaping a weak policy often requires global perturbations that action noise cannot produce. Large language models (LLMs) and vision-language-action (VLA) models offer a pathway: they condition the policy on a natural language prompt, and since the rollout follows from it, modifying the prompt induces global changes. The challenge is finding prompts that induce"},"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":"2607.08837","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-09T18:00:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1bec916f5978849e271a14adeff0af1b0bbd00a079925e415249c7a2fd122a75","abstract_canon_sha256":"729204454bbcb795f0f7d7db2d0edbaaaac282821b4ac6f689207f3be42444e3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-13T00:17:21.051982Z","signature_b64":"iU8tmpfg41+H0GfxyumYbZvysY0uxoXXHsf41CqFX63Fk5+TXUC6cASM48Chd/wj3yOG3U4zyTCHIbmT5ayBBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c32b27b8a43fc41f61a9a6d684d2bbb0f4312e532f73f126b7528b7b635b4e19","last_reissued_at":"2026-07-13T00:17:21.050818Z","signature_status":"signed_v1","first_computed_at":"2026-07-13T00:17:21.050818Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Prompt-Driven Exploration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"David Zhang, John Marangola, Nitish Dashora, Pulkit Agrawal, Raghuram Kowdeed, Richard Li, Ruiyang Luo, Sunshine Jiang, Zhang-Wei Hong","submitted_at":"2026-07-09T18:00:24Z","abstract_excerpt":"Exploration is essential to RL since a policy cannot improve by repeatedly sampling the behaviors it already prefers. Standard methods inject stochasticity in the action space, but such jitter only yields rollouts close to the original. Escaping a weak policy often requires global perturbations that action noise cannot produce. Large language models (LLMs) and vision-language-action (VLA) models offer a pathway: they condition the policy on a natural language prompt, and since the rollout follows from it, modifying the prompt induces global changes. The challenge is finding prompts that induce"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.08837","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/2607.08837/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":"2607.08837","created_at":"2026-07-13T00:17:21.051414+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.08837v1","created_at":"2026-07-13T00:17:21.051414+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.08837","created_at":"2026-07-13T00:17:21.051414+00:00"},{"alias_kind":"pith_short_12","alias_value":"YMVSPOFEH7CB","created_at":"2026-07-13T00:17:21.051414+00:00"},{"alias_kind":"pith_short_16","alias_value":"YMVSPOFEH7CB6YNJ","created_at":"2026-07-13T00:17:21.051414+00:00"},{"alias_kind":"pith_short_8","alias_value":"YMVSPOFE","created_at":"2026-07-13T00:17:21.051414+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YMVSPOFEH7CB6YNJU3LIJUV3WD","json":"https://pith.science/pith/YMVSPOFEH7CB6YNJU3LIJUV3WD.json","graph_json":"https://pith.science/api/pith-number/YMVSPOFEH7CB6YNJU3LIJUV3WD/graph.json","events_json":"https://pith.science/api/pith-number/YMVSPOFEH7CB6YNJU3LIJUV3WD/events.json","paper":"https://pith.science/paper/YMVSPOFE"},"agent_actions":{"view_html":"https://pith.science/pith/YMVSPOFEH7CB6YNJU3LIJUV3WD","download_json":"https://pith.science/pith/YMVSPOFEH7CB6YNJU3LIJUV3WD.json","view_paper":"https://pith.science/paper/YMVSPOFE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.08837&json=true","fetch_graph":"https://pith.science/api/pith-number/YMVSPOFEH7CB6YNJU3LIJUV3WD/graph.json","fetch_events":"https://pith.science/api/pith-number/YMVSPOFEH7CB6YNJU3LIJUV3WD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YMVSPOFEH7CB6YNJU3LIJUV3WD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YMVSPOFEH7CB6YNJU3LIJUV3WD/action/storage_attestation","attest_author":"https://pith.science/pith/YMVSPOFEH7CB6YNJU3LIJUV3WD/action/author_attestation","sign_citation":"https://pith.science/pith/YMVSPOFEH7CB6YNJU3LIJUV3WD/action/citation_signature","submit_replication":"https://pith.science/pith/YMVSPOFEH7CB6YNJU3LIJUV3WD/action/replication_record"}},"created_at":"2026-07-13T00:17:21.051414+00:00","updated_at":"2026-07-13T00:17:21.051414+00:00"}