{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:DUXT2OPL72AJ5EBL46VPYEK7SA","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"c1e91c12841991675f5031397acde6993e72b8b3823e498e0742058b7c74b7ca","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-08-26T09:59:44Z","title_canon_sha256":"d803f3281c303e3afa086b8e61e0f5bd53b136466cf92721fb587e5233297849"},"schema_version":"1.0","source":{"id":"2508.18884","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.18884","created_at":"2026-07-05T11:59:32Z"},{"alias_kind":"arxiv_version","alias_value":"2508.18884v1","created_at":"2026-07-05T11:59:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.18884","created_at":"2026-07-05T11:59:32Z"},{"alias_kind":"pith_short_12","alias_value":"DUXT2OPL72AJ","created_at":"2026-07-05T11:59:32Z"},{"alias_kind":"pith_short_16","alias_value":"DUXT2OPL72AJ5EBL","created_at":"2026-07-05T11:59:32Z"},{"alias_kind":"pith_short_8","alias_value":"DUXT2OPL","created_at":"2026-07-05T11:59:32Z"}],"graph_snapshots":[{"event_id":"sha256:f30f1d7d498a2b2de7c91f955dbfa467b0e9c117ee580aa0ca2cf024f33c2a82","target":"graph","created_at":"2026-07-05T11:59:32Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2508.18884/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Exploration is essential in modern learning, from reinforcement learning environments with small neural policies to large language models (LLMs). Existing work, such as DPO, leverages full sequence log-likelihoods to capture an entire trajectory of the model's decisions, while methods like GRPO aggregate per-token ratios into a trajectory-level update. However, both often limit exploration on long-horizon tasks. We introduce History-Aggregated Exploratory Policy Optimization (HAEPO), a history-aware exploratory loss to combat these shortcomings. HAEPO compresses each trajectory into the sum of","authors_text":"Alakh Sharma, Dhruv Kumar, Gaurish Trivedi, Jagat Sesh Challa, Kartikey Singh Bhandari, Pratik Narang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-08-26T09:59:44Z","title":"HAEPO: History-Aggregated Exploratory Policy Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.18884","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:29407dc51b86b5333b6bd4836e540b41c10bde0cef9b4bde688b9f7843da0c96","target":"record","created_at":"2026-07-05T11:59:32Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"c1e91c12841991675f5031397acde6993e72b8b3823e498e0742058b7c74b7ca","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-08-26T09:59:44Z","title_canon_sha256":"d803f3281c303e3afa086b8e61e0f5bd53b136466cf92721fb587e5233297849"},"schema_version":"1.0","source":{"id":"2508.18884","kind":"arxiv","version":1}},"canonical_sha256":"1d2f3d39ebfe809e902be7aafc115f902376e640100553161038b9e9682ccf59","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1d2f3d39ebfe809e902be7aafc115f902376e640100553161038b9e9682ccf59","first_computed_at":"2026-07-05T11:59:32.987937Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:59:32.987937Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ETi9ERqp8JV+gMPnA3XTxASb8L77IbI8sw7+tCmoBa80wqw5KsNMX5JN6EOdHwbsg0dqu8AfaTAN2qCGhvEdCg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:59:32.988414Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.18884","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:29407dc51b86b5333b6bd4836e540b41c10bde0cef9b4bde688b9f7843da0c96","sha256:f30f1d7d498a2b2de7c91f955dbfa467b0e9c117ee580aa0ca2cf024f33c2a82"],"state_sha256":"ed69fee623ee3d6fe7f4675bf7e5422bb0b91da29af2cead09646bd545f94fbd"}