{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CCXVZPOLHD34TWKTAXCEJHKXFK","short_pith_number":"pith:CCXVZPOL","schema_version":"1.0","canonical_sha256":"10af5cbdcb38f7c9d95305c4449d572aa7d4d2627f9cd7bfbed819a7cbec939f","source":{"kind":"arxiv","id":"2606.08471","version":1},"attestation_state":"computed","paper":{"title":"More Yap Less Meaning: Uncovering Self-Improvement Behavior in SLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Erik Arakelyan, Marina Igitkhanian","submitted_at":"2026-06-07T06:27:37Z","abstract_excerpt":"Recently, language models have made rapid progress across various domains and applications. However, their capability for self-improvement, i.e., whether they are adept at recognising and correcting flaws in their own reasoning, remains dubious. In this study, we address this question by constructing a sufficiency test to rigorously examine the self-correction capabilities of small language models (SLMs). We propose a minimal three-step self-correction pipeline that collects initial SLM answers, prompts the same model to generate hints for its incorrect responses given the ground truth, and fe"},"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":"2606.08471","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-07T06:27:37Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"18c0dae955de89151cfba0a610d27153290561962c97d8ac9cda9716ae38c89a","abstract_canon_sha256":"1637ca88c63613a7f587c2a968f0febd1a3c06ddb4276e70667f73c0e141118d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:05:37.561344Z","signature_b64":"Fcig2mpER1zrx1JmcLD2BG+2Fa3qexuStK3JxwFFVZAh4DojEyBge7U2TZVU772jWAUgKXgMpRvigflQV0RACw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"10af5cbdcb38f7c9d95305c4449d572aa7d4d2627f9cd7bfbed819a7cbec939f","last_reissued_at":"2026-06-09T01:05:37.560926Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:05:37.560926Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"More Yap Less Meaning: Uncovering Self-Improvement Behavior in SLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Erik Arakelyan, Marina Igitkhanian","submitted_at":"2026-06-07T06:27:37Z","abstract_excerpt":"Recently, language models have made rapid progress across various domains and applications. However, their capability for self-improvement, i.e., whether they are adept at recognising and correcting flaws in their own reasoning, remains dubious. In this study, we address this question by constructing a sufficiency test to rigorously examine the self-correction capabilities of small language models (SLMs). We propose a minimal three-step self-correction pipeline that collects initial SLM answers, prompts the same model to generate hints for its incorrect responses given the ground truth, and fe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.08471","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/2606.08471/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":"2606.08471","created_at":"2026-06-09T01:05:37.560988+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.08471v1","created_at":"2026-06-09T01:05:37.560988+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.08471","created_at":"2026-06-09T01:05:37.560988+00:00"},{"alias_kind":"pith_short_12","alias_value":"CCXVZPOLHD34","created_at":"2026-06-09T01:05:37.560988+00:00"},{"alias_kind":"pith_short_16","alias_value":"CCXVZPOLHD34TWKT","created_at":"2026-06-09T01:05:37.560988+00:00"},{"alias_kind":"pith_short_8","alias_value":"CCXVZPOL","created_at":"2026-06-09T01:05:37.560988+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/CCXVZPOLHD34TWKTAXCEJHKXFK","json":"https://pith.science/pith/CCXVZPOLHD34TWKTAXCEJHKXFK.json","graph_json":"https://pith.science/api/pith-number/CCXVZPOLHD34TWKTAXCEJHKXFK/graph.json","events_json":"https://pith.science/api/pith-number/CCXVZPOLHD34TWKTAXCEJHKXFK/events.json","paper":"https://pith.science/paper/CCXVZPOL"},"agent_actions":{"view_html":"https://pith.science/pith/CCXVZPOLHD34TWKTAXCEJHKXFK","download_json":"https://pith.science/pith/CCXVZPOLHD34TWKTAXCEJHKXFK.json","view_paper":"https://pith.science/paper/CCXVZPOL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.08471&json=true","fetch_graph":"https://pith.science/api/pith-number/CCXVZPOLHD34TWKTAXCEJHKXFK/graph.json","fetch_events":"https://pith.science/api/pith-number/CCXVZPOLHD34TWKTAXCEJHKXFK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CCXVZPOLHD34TWKTAXCEJHKXFK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CCXVZPOLHD34TWKTAXCEJHKXFK/action/storage_attestation","attest_author":"https://pith.science/pith/CCXVZPOLHD34TWKTAXCEJHKXFK/action/author_attestation","sign_citation":"https://pith.science/pith/CCXVZPOLHD34TWKTAXCEJHKXFK/action/citation_signature","submit_replication":"https://pith.science/pith/CCXVZPOLHD34TWKTAXCEJHKXFK/action/replication_record"}},"created_at":"2026-06-09T01:05:37.560988+00:00","updated_at":"2026-06-09T01:05:37.560988+00:00"}