{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZC3B3HAIECW4RYMUZERALVXROU","short_pith_number":"pith:ZC3B3HAI","schema_version":"1.0","canonical_sha256":"c8b61d9c0820adc8e194c92205d6f175160d4a5c9854df2fa4afd3a142a5ba20","source":{"kind":"arxiv","id":"2608.02087","version":1},"attestation_state":"computed","paper":{"title":"Instruction-Conditioned Exploration with Asymmetric Reinforcement Learning and Self-Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Jim Dilkes, Sebastian Stein, Vahid Yazdanpanah","submitted_at":"2026-08-03T11:47:52Z","abstract_excerpt":"Post-training Large Language Models (LLMs) with Reinforcement Learning (RL) has become an important tool for improving model capabilities, but the LLM action-space structure introduces challenges distinct from classical RL, with implications for inducing exploration. New methods are required that leverage the broad knowledge and flexibility of pre-trained LLMs to deliberately generate diverse experience at training time. We propose Instruction-Conditioned Exploration (ICE), which supplements task prompts during training with one of several distinct instructions, increasing the coverage of beha"},"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":"2608.02087","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-03T11:47:52Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"7678631682ac339080bf9ec0b0f7766eb34a257dafe5171e4cb889754b586255","abstract_canon_sha256":"5988acd1e0394fbd160de248d2988184b4b6b29abb5b6581999e9b26f2e98afd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T02:10:49.289282Z","signature_b64":"Uct9MDeCqF6LifMVRnW0/nl9dcMyjWqKCSvuO6pua9YBHtmAe1QBk8HQm8ppvdg/90TV7z6yoqhWppEhqPzsCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c8b61d9c0820adc8e194c92205d6f175160d4a5c9854df2fa4afd3a142a5ba20","last_reissued_at":"2026-08-04T02:10:49.287801Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T02:10:49.287801Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Instruction-Conditioned Exploration with Asymmetric Reinforcement Learning and Self-Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Jim Dilkes, Sebastian Stein, Vahid Yazdanpanah","submitted_at":"2026-08-03T11:47:52Z","abstract_excerpt":"Post-training Large Language Models (LLMs) with Reinforcement Learning (RL) has become an important tool for improving model capabilities, but the LLM action-space structure introduces challenges distinct from classical RL, with implications for inducing exploration. New methods are required that leverage the broad knowledge and flexibility of pre-trained LLMs to deliberately generate diverse experience at training time. We propose Instruction-Conditioned Exploration (ICE), which supplements task prompts during training with one of several distinct instructions, increasing the coverage of beha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.02087","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/2608.02087/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":"2608.02087","created_at":"2026-08-04T02:10:49.289329+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.02087v1","created_at":"2026-08-04T02:10:49.289329+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.02087","created_at":"2026-08-04T02:10:49.289329+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZC3B3HAIECW4","created_at":"2026-08-04T02:10:49.289329+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZC3B3HAIECW4RYMU","created_at":"2026-08-04T02:10:49.289329+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZC3B3HAI","created_at":"2026-08-04T02:10:49.289329+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/ZC3B3HAIECW4RYMUZERALVXROU","json":"https://pith.science/pith/ZC3B3HAIECW4RYMUZERALVXROU.json","graph_json":"https://pith.science/api/pith-number/ZC3B3HAIECW4RYMUZERALVXROU/graph.json","events_json":"https://pith.science/api/pith-number/ZC3B3HAIECW4RYMUZERALVXROU/events.json","paper":"https://pith.science/paper/ZC3B3HAI"},"agent_actions":{"view_html":"https://pith.science/pith/ZC3B3HAIECW4RYMUZERALVXROU","download_json":"https://pith.science/pith/ZC3B3HAIECW4RYMUZERALVXROU.json","view_paper":"https://pith.science/paper/ZC3B3HAI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.02087&json=true","fetch_graph":"https://pith.science/api/pith-number/ZC3B3HAIECW4RYMUZERALVXROU/graph.json","fetch_events":"https://pith.science/api/pith-number/ZC3B3HAIECW4RYMUZERALVXROU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZC3B3HAIECW4RYMUZERALVXROU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZC3B3HAIECW4RYMUZERALVXROU/action/storage_attestation","attest_author":"https://pith.science/pith/ZC3B3HAIECW4RYMUZERALVXROU/action/author_attestation","sign_citation":"https://pith.science/pith/ZC3B3HAIECW4RYMUZERALVXROU/action/citation_signature","submit_replication":"https://pith.science/pith/ZC3B3HAIECW4RYMUZERALVXROU/action/replication_record"}},"created_at":"2026-08-04T02:10:49.289329+00:00","updated_at":"2026-08-04T02:10:49.289329+00:00"}