{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TQLCBNYLDVDBILXDGTAD5AVTHT","short_pith_number":"pith:TQLCBNYL","schema_version":"1.0","canonical_sha256":"9c1620b70b1d46142ee334c03e82b33cf1f66a1a69a9a3c7e08a19cb63d891f7","source":{"kind":"arxiv","id":"2606.18521","version":1},"attestation_state":"computed","paper":{"title":"Sparsity Curse: Understanding RLVR Model Parameter Space from Model Merging","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chenrui Wu, Haishuai Wang, Jiajun Bu, Jiangchuan Liu, Zexi Li","submitted_at":"2026-06-16T22:22:40Z","abstract_excerpt":"Reinforcement Learning with Verifiable Reward (RLVR) has emerged as a powerful post-training paradigm that surpasses Supervised Fine-Tuning (SFT) in eliciting reasoning intelligence and resisting catastrophic forgetting. Recent studies further reveal that RLVR induces highly sparse and off-principal parameter updates compared to SFT. This naturally raises the question: does such sparsity make RLVR models more amenable to model merging? If so, model merging would offer a scalable, training-free path to aggregate diverse reasoning capabilities from independently trained RLVR models. Surprisingly"},"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":"2606.18521","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-06-16T22:22:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"031e6f7009980e9a4e0a87ff03aeae43d0678992e58c436d4155c787e552e2c4","abstract_canon_sha256":"25a51ee8f8b3c7cbcca2f6e0faba7dacdf4532c27d3f0d3b76d502ca33cedbc7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:11:39.920120Z","signature_b64":"r3KcWm7G4UqveeeCIIJ+kL8vj4QRkpPFseP2wLyyDwiq3rLqrxXfwjRD+XALwomcOdZrIm4311mboLoze886CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c1620b70b1d46142ee334c03e82b33cf1f66a1a69a9a3c7e08a19cb63d891f7","last_reissued_at":"2026-06-19T16:11:39.919611Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:11:39.919611Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sparsity Curse: Understanding RLVR Model Parameter Space from Model Merging","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chenrui Wu, Haishuai Wang, Jiajun Bu, Jiangchuan Liu, Zexi Li","submitted_at":"2026-06-16T22:22:40Z","abstract_excerpt":"Reinforcement Learning with Verifiable Reward (RLVR) has emerged as a powerful post-training paradigm that surpasses Supervised Fine-Tuning (SFT) in eliciting reasoning intelligence and resisting catastrophic forgetting. Recent studies further reveal that RLVR induces highly sparse and off-principal parameter updates compared to SFT. This naturally raises the question: does such sparsity make RLVR models more amenable to model merging? If so, model merging would offer a scalable, training-free path to aggregate diverse reasoning capabilities from independently trained RLVR models. Surprisingly"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.18521","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/2606.18521/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":"2606.18521","created_at":"2026-06-19T16:11:39.919702+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.18521v1","created_at":"2026-06-19T16:11:39.919702+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.18521","created_at":"2026-06-19T16:11:39.919702+00:00"},{"alias_kind":"pith_short_12","alias_value":"TQLCBNYLDVDB","created_at":"2026-06-19T16:11:39.919702+00:00"},{"alias_kind":"pith_short_16","alias_value":"TQLCBNYLDVDBILXD","created_at":"2026-06-19T16:11:39.919702+00:00"},{"alias_kind":"pith_short_8","alias_value":"TQLCBNYL","created_at":"2026-06-19T16:11:39.919702+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/TQLCBNYLDVDBILXDGTAD5AVTHT","json":"https://pith.science/pith/TQLCBNYLDVDBILXDGTAD5AVTHT.json","graph_json":"https://pith.science/api/pith-number/TQLCBNYLDVDBILXDGTAD5AVTHT/graph.json","events_json":"https://pith.science/api/pith-number/TQLCBNYLDVDBILXDGTAD5AVTHT/events.json","paper":"https://pith.science/paper/TQLCBNYL"},"agent_actions":{"view_html":"https://pith.science/pith/TQLCBNYLDVDBILXDGTAD5AVTHT","download_json":"https://pith.science/pith/TQLCBNYLDVDBILXDGTAD5AVTHT.json","view_paper":"https://pith.science/paper/TQLCBNYL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.18521&json=true","fetch_graph":"https://pith.science/api/pith-number/TQLCBNYLDVDBILXDGTAD5AVTHT/graph.json","fetch_events":"https://pith.science/api/pith-number/TQLCBNYLDVDBILXDGTAD5AVTHT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TQLCBNYLDVDBILXDGTAD5AVTHT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TQLCBNYLDVDBILXDGTAD5AVTHT/action/storage_attestation","attest_author":"https://pith.science/pith/TQLCBNYLDVDBILXDGTAD5AVTHT/action/author_attestation","sign_citation":"https://pith.science/pith/TQLCBNYLDVDBILXDGTAD5AVTHT/action/citation_signature","submit_replication":"https://pith.science/pith/TQLCBNYLDVDBILXDGTAD5AVTHT/action/replication_record"}},"created_at":"2026-06-19T16:11:39.919702+00:00","updated_at":"2026-06-19T16:11:39.919702+00:00"}