{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SGBBLAWZYZ4724XP2FM2Z4QNNY","short_pith_number":"pith:SGBBLAWZ","schema_version":"1.0","canonical_sha256":"91821582d9c679fd72efd159acf20d6e359accc592e17e1a9e4fcc9c9e133bce","source":{"kind":"arxiv","id":"2605.16787","version":1},"attestation_state":"computed","paper":{"title":"The Unlearnability Phenomenon in RLVR for Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A substantial subset of hard examples remains unlearnable in RLVR even when correct rollouts are available.","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Chen Zhao, He He, Yulin Chen","submitted_at":"2026-05-16T03:43:19Z","abstract_excerpt":"Reinforcement Learning with Verifiable Reward (RLVR) has proven effective in improving Large Language Model's (LLM) reasoning ability. However, the learning dynamics of RLVR remain underexplored. In this paper, we reveal a counterintuitive phenomenon: among hard examples that the model initially struggles with, a substantial subset remains unlearnable even when correct rollouts are present. To understand the phenomenon, we first demonstrate that existing optimization and sampling techniques fail to resolve unlearnability. With cross-example gradient analysis, we show that unlearnable examples "},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.16787","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-16T03:43:19Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"ec1a0c61f014ba0d0ab86b2b250cf1910d0c67d264cd9b26ef7f759e9fc17930","abstract_canon_sha256":"d17927674a6936d454676c062b0206446384d52b5f939868107fff7be642c183"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:03:21.984979Z","signature_b64":"uUeunQJQ+y/+mdUTXlUQZ9KmucuVN1wbE+Um0OTeH4VR1TUQ91cKSHQP7v5ZtIwn386QCkqeuxJcPGvfL2orAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"91821582d9c679fd72efd159acf20d6e359accc592e17e1a9e4fcc9c9e133bce","last_reissued_at":"2026-05-20T00:03:21.984021Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:03:21.984021Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Unlearnability Phenomenon in RLVR for Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A substantial subset of hard examples remains unlearnable in RLVR even when correct rollouts are available.","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Chen Zhao, He He, Yulin Chen","submitted_at":"2026-05-16T03:43:19Z","abstract_excerpt":"Reinforcement Learning with Verifiable Reward (RLVR) has proven effective in improving Large Language Model's (LLM) reasoning ability. However, the learning dynamics of RLVR remain underexplored. In this paper, we reveal a counterintuitive phenomenon: among hard examples that the model initially struggles with, a substantial subset remains unlearnable even when correct rollouts are present. To understand the phenomenon, we first demonstrate that existing optimization and sampling techniques fail to resolve unlearnability. With cross-example gradient analysis, we show that unlearnable examples "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"among hard examples that the model initially struggles with, a substantial subset remains unlearnable even when correct rollouts are present, characterized by low gradient similarity with the rest of the examples and ungeneralizable reasoning patterns.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That cross-example gradient analysis reliably detects fundamental representation issues causing unlearnability, and that failure of data augmentation to improve gradient similarity demonstrates inherent limitations of RL approaches.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"RLVR training for language models exhibits an unlearnability phenomenon where certain hard examples stay unlearnable due to low gradient similarity and ungeneralizable reasoning patterns.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A substantial subset of hard examples remains unlearnable in RLVR even when correct rollouts are available.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"0442aa8e0c1996074c647de33d9a9a84f2aa157696221ffb85e432a94d5cab82"},"source":{"id":"2605.16787","kind":"arxiv","version":1},"verdict":{"id":"53416e6d-e9ed-4f2f-b547-ca99d0b58c26","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T21:34:46.572541Z","strongest_claim":"among hard examples that the model initially struggles with, a substantial subset remains unlearnable even when correct rollouts are present, characterized by low gradient similarity with the rest of the examples and ungeneralizable reasoning patterns.","one_line_summary":"RLVR training for language models exhibits an unlearnability phenomenon where certain hard examples stay unlearnable due to low gradient similarity and ungeneralizable reasoning patterns.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That cross-example gradient analysis reliably detects fundamental representation issues causing unlearnability, and that failure of data augmentation to improve gradient similarity demonstrates inherent limitations of RL approaches.","pith_extraction_headline":"A substantial subset of hard examples remains unlearnable in RLVR even when correct rollouts are available."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.16787/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T22:01:19.637991Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T21:40:53.320904Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T19:01:56.297811Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.433407Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"344db38d5eef81d2f33d7b949bd665284a8db74a5051cbf37cebe9e2c5cd664e"},"references":{"count":22,"sample":[{"doi":"","year":2025,"title":"arXiv preprint arXiv:2512.01775 , year=","work_id":"52575a72-367d-476d-89e4-a3641cc0cecd","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Qwen2.5 Technical Report","work_id":"d8432992-4980-4a81-85c7-9fa2c2b87f85","ref_index":2,"cited_arxiv_id":"2412.15115","is_internal_anchor":true},{"doi":"","year":null,"title":"We have also tried different sampling batch size and gradient update batch size to vary the maximum number of off-policy update","work_id":"f41dc38a-ebeb-4a7d-ad1b-a86cae1a975e","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"A reader should be able to solve any single subproblem without seeing the others","work_id":"497a4852-4ee4-4ca8-89d5-c29e360162a1","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Clarity: Each subproblem must be unambiguous and have a unique, well-defined answer","work_id":"925fbace-7d7c-4f4c-ac17-57b13af878b6","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":22,"snapshot_sha256":"aab469435f40f1eb4fdcef417f388936a7b532241aba15ae80bd6bd80f847046","internal_anchors":1},"formal_canon":{"evidence_count":2,"snapshot_sha256":"b417023e8fbf76db7c06b26d33aabe58271aee8772119b671c6a85aa140b5ef4"},"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":"2605.16787","created_at":"2026-05-20T00:03:21.984169+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.16787v1","created_at":"2026-05-20T00:03:21.984169+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16787","created_at":"2026-05-20T00:03:21.984169+00:00"},{"alias_kind":"pith_short_12","alias_value":"SGBBLAWZYZ47","created_at":"2026-05-20T00:03:21.984169+00:00"},{"alias_kind":"pith_short_16","alias_value":"SGBBLAWZYZ4724XP","created_at":"2026-05-20T00:03:21.984169+00:00"},{"alias_kind":"pith_short_8","alias_value":"SGBBLAWZ","created_at":"2026-05-20T00:03:21.984169+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SGBBLAWZYZ4724XP2FM2Z4QNNY","json":"https://pith.science/pith/SGBBLAWZYZ4724XP2FM2Z4QNNY.json","graph_json":"https://pith.science/api/pith-number/SGBBLAWZYZ4724XP2FM2Z4QNNY/graph.json","events_json":"https://pith.science/api/pith-number/SGBBLAWZYZ4724XP2FM2Z4QNNY/events.json","paper":"https://pith.science/paper/SGBBLAWZ"},"agent_actions":{"view_html":"https://pith.science/pith/SGBBLAWZYZ4724XP2FM2Z4QNNY","download_json":"https://pith.science/pith/SGBBLAWZYZ4724XP2FM2Z4QNNY.json","view_paper":"https://pith.science/paper/SGBBLAWZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.16787&json=true","fetch_graph":"https://pith.science/api/pith-number/SGBBLAWZYZ4724XP2FM2Z4QNNY/graph.json","fetch_events":"https://pith.science/api/pith-number/SGBBLAWZYZ4724XP2FM2Z4QNNY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SGBBLAWZYZ4724XP2FM2Z4QNNY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SGBBLAWZYZ4724XP2FM2Z4QNNY/action/storage_attestation","attest_author":"https://pith.science/pith/SGBBLAWZYZ4724XP2FM2Z4QNNY/action/author_attestation","sign_citation":"https://pith.science/pith/SGBBLAWZYZ4724XP2FM2Z4QNNY/action/citation_signature","submit_replication":"https://pith.science/pith/SGBBLAWZYZ4724XP2FM2Z4QNNY/action/replication_record"}},"created_at":"2026-05-20T00:03:21.984169+00:00","updated_at":"2026-05-20T00:03:21.984169+00:00"}