{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6F5NIZGL4E2TVXFBTQ2DRE5V63","short_pith_number":"pith:6F5NIZGL","schema_version":"1.0","canonical_sha256":"f17ad464cbe1353adca19c343893b5f6ed3190018ffc251f8f47c26b2516dc36","source":{"kind":"arxiv","id":"2606.06546","version":1},"attestation_state":"computed","paper":{"title":"Elmes*: Automated Construction of Fine-Grained Evaluation Rubrics for Large Language Models in Long-Tail Educational Scenarios","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aimin Zhou, Hao Hao, Ruohua Zhang, Siyu Song, Tao Liu, Wentao Liu, Ye Lu","submitted_at":"2026-06-04T07:40:12Z","abstract_excerpt":"Evaluating large language models (LLMs) for education requires measuring how models teach, not only what they know. Existing benchmarks emphasize domain-general correctness or depend on manually designed rubrics that scale poorly to long-tail pedagogical scenarios. We introduce Elmes*, an end-to-end framework for constructing, refining, and applying fine-grained scenario-specific rubrics. Elmes* combines a declarative multi-agent engine for teacher--student--judge interactions with SceneGen, a self-evolving module that co-optimizes evaluation criteria and test data from expert-defined pedagogi"},"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":"2606.06546","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-04T07:40:12Z","cross_cats_sorted":[],"title_canon_sha256":"54d92965b4beb2342fdb461554dcdc1a59799cfac7916a178364cdb4d51bf63b","abstract_canon_sha256":"5c15efcd71fad02d22c2cb3865e2dd36ad39f05a5cf0f9ab443d8b42362c2f03"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-08T00:03:41.880251Z","signature_b64":"In4KwiIuf5WW5I+StMJ42k88SRHIgQw4BsvPBFGxXHha6yy+4qYDdQuaLJuXHd1gDd0ALVAeHS3ghF8VHf7kAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f17ad464cbe1353adca19c343893b5f6ed3190018ffc251f8f47c26b2516dc36","last_reissued_at":"2026-06-08T00:03:41.879358Z","signature_status":"signed_v1","first_computed_at":"2026-06-08T00:03:41.879358Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Elmes*: Automated Construction of Fine-Grained Evaluation Rubrics for Large Language Models in Long-Tail Educational Scenarios","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aimin Zhou, Hao Hao, Ruohua Zhang, Siyu Song, Tao Liu, Wentao Liu, Ye Lu","submitted_at":"2026-06-04T07:40:12Z","abstract_excerpt":"Evaluating large language models (LLMs) for education requires measuring how models teach, not only what they know. Existing benchmarks emphasize domain-general correctness or depend on manually designed rubrics that scale poorly to long-tail pedagogical scenarios. We introduce Elmes*, an end-to-end framework for constructing, refining, and applying fine-grained scenario-specific rubrics. Elmes* combines a declarative multi-agent engine for teacher--student--judge interactions with SceneGen, a self-evolving module that co-optimizes evaluation criteria and test data from expert-defined pedagogi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.06546","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/2606.06546/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":"2606.06546","created_at":"2026-06-08T00:03:41.879513+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.06546v1","created_at":"2026-06-08T00:03:41.879513+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.06546","created_at":"2026-06-08T00:03:41.879513+00:00"},{"alias_kind":"pith_short_12","alias_value":"6F5NIZGL4E2T","created_at":"2026-06-08T00:03:41.879513+00:00"},{"alias_kind":"pith_short_16","alias_value":"6F5NIZGL4E2TVXFB","created_at":"2026-06-08T00:03:41.879513+00:00"},{"alias_kind":"pith_short_8","alias_value":"6F5NIZGL","created_at":"2026-06-08T00:03:41.879513+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/6F5NIZGL4E2TVXFBTQ2DRE5V63","json":"https://pith.science/pith/6F5NIZGL4E2TVXFBTQ2DRE5V63.json","graph_json":"https://pith.science/api/pith-number/6F5NIZGL4E2TVXFBTQ2DRE5V63/graph.json","events_json":"https://pith.science/api/pith-number/6F5NIZGL4E2TVXFBTQ2DRE5V63/events.json","paper":"https://pith.science/paper/6F5NIZGL"},"agent_actions":{"view_html":"https://pith.science/pith/6F5NIZGL4E2TVXFBTQ2DRE5V63","download_json":"https://pith.science/pith/6F5NIZGL4E2TVXFBTQ2DRE5V63.json","view_paper":"https://pith.science/paper/6F5NIZGL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.06546&json=true","fetch_graph":"https://pith.science/api/pith-number/6F5NIZGL4E2TVXFBTQ2DRE5V63/graph.json","fetch_events":"https://pith.science/api/pith-number/6F5NIZGL4E2TVXFBTQ2DRE5V63/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6F5NIZGL4E2TVXFBTQ2DRE5V63/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6F5NIZGL4E2TVXFBTQ2DRE5V63/action/storage_attestation","attest_author":"https://pith.science/pith/6F5NIZGL4E2TVXFBTQ2DRE5V63/action/author_attestation","sign_citation":"https://pith.science/pith/6F5NIZGL4E2TVXFBTQ2DRE5V63/action/citation_signature","submit_replication":"https://pith.science/pith/6F5NIZGL4E2TVXFBTQ2DRE5V63/action/replication_record"}},"created_at":"2026-06-08T00:03:41.879513+00:00","updated_at":"2026-06-08T00:03:41.879513+00:00"}