{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:4Q7TU3WKCTPCDJQZ3ZWG2MK3R3","short_pith_number":"pith:4Q7TU3WK","canonical_record":{"source":{"id":"2607.26643","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-29T09:05:40Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6d01e62a60cfa4773997e1b49229fb9a8227e841a9eebae69d956719046939cf","abstract_canon_sha256":"5ef5292c3f812261d123cea119dcbb519c292c11732e828fa418c7d63680fa62"},"schema_version":"1.0"},"canonical_sha256":"e43f3a6eca14de21a619de6c6d315b8ec3420e4dbfa4dd61bc3758c99d944144","source":{"kind":"arxiv","id":"2607.26643","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.26643","created_at":"2026-07-30T01:21:40Z"},{"alias_kind":"arxiv_version","alias_value":"2607.26643v1","created_at":"2026-07-30T01:21:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.26643","created_at":"2026-07-30T01:21:40Z"},{"alias_kind":"pith_short_12","alias_value":"4Q7TU3WKCTPC","created_at":"2026-07-30T01:21:40Z"},{"alias_kind":"pith_short_16","alias_value":"4Q7TU3WKCTPCDJQZ","created_at":"2026-07-30T01:21:40Z"},{"alias_kind":"pith_short_8","alias_value":"4Q7TU3WK","created_at":"2026-07-30T01:21:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:4Q7TU3WKCTPCDJQZ3ZWG2MK3R3","target":"record","payload":{"canonical_record":{"source":{"id":"2607.26643","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-29T09:05:40Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6d01e62a60cfa4773997e1b49229fb9a8227e841a9eebae69d956719046939cf","abstract_canon_sha256":"5ef5292c3f812261d123cea119dcbb519c292c11732e828fa418c7d63680fa62"},"schema_version":"1.0"},"canonical_sha256":"e43f3a6eca14de21a619de6c6d315b8ec3420e4dbfa4dd61bc3758c99d944144","receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e43f3a6eca14de21a619de6c6d315b8ec3420e4dbfa4dd61bc3758c99d944144","last_reissued_at":"2026-07-30T01:21:40.469653Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-30T01:21:40.469653Z"},"source_kind":"arxiv","source_id":"2607.26643","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-30T01:21:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"inCQfYPXB90FnAnJGQhyvEro1Swz27lFtDnZ6gYBZsm4SPhfYYxFmobSW4NvsRR1v3szF16h21aHdlBQGKiABg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:43:20.539922Z"},"content_sha256":"791b189af947a2857626626c1f5ef4594448114910843b251ddfbb0c6eb5f1f9","schema_version":"1.0","event_id":"sha256:791b189af947a2857626626c1f5ef4594448114910843b251ddfbb0c6eb5f1f9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:4Q7TU3WKCTPCDJQZ3ZWG2MK3R3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Rethinking Self-Evolution: A Constrained Exploration-Exploitation Process for Mitigating Skill Overfitting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Chao Liu, Hongqiang Lin, Nenggan Zheng, Xiaofan Bai, Xipeng Cao, Xuan Jin, Yuhong Li","submitted_at":"2026-07-29T09:05:40Z","abstract_excerpt":"Enabling large language model (LLM) agents to accumulate and reuse experience from past interactions remains a central challenge in real-world applications. A promising solution is to treat skills as trainable states and optimize them in the same way as model parameters in neural network training. However, data-driven skill optimization is prone to overfitting to the limited trajectories collected from real environments. Overexploiting these trajectories overfits the current batch, while unconstrained exploration causes regression on previously solved cases. This tension motivates a constraine"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.26643","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/2607.26643/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-30T01:21:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"m51lNgpDEgmVyue9sGrOtr8jSdrwFnhfwuraK+eK0NCTQ/nHthCXpCEUTiJiGkMqpKIbnes/ADNSHutCpvbgBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:43:20.540419Z"},"content_sha256":"148584d05aab9fcf389c7d2f73edf4831cfb9e774c88d575960ed401f606aebd","schema_version":"1.0","event_id":"sha256:148584d05aab9fcf389c7d2f73edf4831cfb9e774c88d575960ed401f606aebd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4Q7TU3WKCTPCDJQZ3ZWG2MK3R3/bundle.json","state_url":"https://pith.science/pith/4Q7TU3WKCTPCDJQZ3ZWG2MK3R3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4Q7TU3WKCTPCDJQZ3ZWG2MK3R3/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-03T16:43:20Z","links":{"resolver":"https://pith.science/pith/4Q7TU3WKCTPCDJQZ3ZWG2MK3R3","bundle":"https://pith.science/pith/4Q7TU3WKCTPCDJQZ3ZWG2MK3R3/bundle.json","state":"https://pith.science/pith/4Q7TU3WKCTPCDJQZ3ZWG2MK3R3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4Q7TU3WKCTPCDJQZ3ZWG2MK3R3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4Q7TU3WKCTPCDJQZ3ZWG2MK3R3","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":"5ef5292c3f812261d123cea119dcbb519c292c11732e828fa418c7d63680fa62","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-29T09:05:40Z","title_canon_sha256":"6d01e62a60cfa4773997e1b49229fb9a8227e841a9eebae69d956719046939cf"},"schema_version":"1.0","source":{"id":"2607.26643","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.26643","created_at":"2026-07-30T01:21:40Z"},{"alias_kind":"arxiv_version","alias_value":"2607.26643v1","created_at":"2026-07-30T01:21:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.26643","created_at":"2026-07-30T01:21:40Z"},{"alias_kind":"pith_short_12","alias_value":"4Q7TU3WKCTPC","created_at":"2026-07-30T01:21:40Z"},{"alias_kind":"pith_short_16","alias_value":"4Q7TU3WKCTPCDJQZ","created_at":"2026-07-30T01:21:40Z"},{"alias_kind":"pith_short_8","alias_value":"4Q7TU3WK","created_at":"2026-07-30T01:21:40Z"}],"graph_snapshots":[{"event_id":"sha256:148584d05aab9fcf389c7d2f73edf4831cfb9e774c88d575960ed401f606aebd","target":"graph","created_at":"2026-07-30T01:21:40Z","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/2607.26643/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Enabling large language model (LLM) agents to accumulate and reuse experience from past interactions remains a central challenge in real-world applications. A promising solution is to treat skills as trainable states and optimize them in the same way as model parameters in neural network training. However, data-driven skill optimization is prone to overfitting to the limited trajectories collected from real environments. Overexploiting these trajectories overfits the current batch, while unconstrained exploration causes regression on previously solved cases. This tension motivates a constraine","authors_text":"Chao Liu, Hongqiang Lin, Nenggan Zheng, Xiaofan Bai, Xipeng Cao, Xuan Jin, Yuhong Li","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-29T09:05:40Z","title":"Rethinking Self-Evolution: A Constrained Exploration-Exploitation Process for Mitigating Skill Overfitting"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.26643","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":"sha256:791b189af947a2857626626c1f5ef4594448114910843b251ddfbb0c6eb5f1f9","target":"record","created_at":"2026-07-30T01:21:40Z","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":"5ef5292c3f812261d123cea119dcbb519c292c11732e828fa418c7d63680fa62","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-29T09:05:40Z","title_canon_sha256":"6d01e62a60cfa4773997e1b49229fb9a8227e841a9eebae69d956719046939cf"},"schema_version":"1.0","source":{"id":"2607.26643","kind":"arxiv","version":1}},"canonical_sha256":"e43f3a6eca14de21a619de6c6d315b8ec3420e4dbfa4dd61bc3758c99d944144","receipt":{"builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e43f3a6eca14de21a619de6c6d315b8ec3420e4dbfa4dd61bc3758c99d944144","first_computed_at":"2026-07-30T01:21:40.469653Z","kind":"pith_receipt","last_reissued_at":"2026-07-30T01:21:40.469653Z","receipt_version":"0.3","signature_status":"unsigned_v0"},"source_id":"2607.26643","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:791b189af947a2857626626c1f5ef4594448114910843b251ddfbb0c6eb5f1f9","sha256:148584d05aab9fcf389c7d2f73edf4831cfb9e774c88d575960ed401f606aebd"],"state_sha256":"ecf6cd44a481fe91ecf4d43bda69bb8125988df6a4ef0285b6ac9df1ba8c27f2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dzkL9+hbK4F+5vUyYs4vbKsaOHKSle58dLjmVJqxtjNr9NMgxhIos5Zo44EA6h2VlzMNCc9JqUu+Ngssr9vdAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T16:43:20.544098Z","bundle_sha256":"19c06b1e349d4806691e94294759d2feb3c8b4143a81e2f435dbec48b6f36d15"}}