{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TAYZGEGC6M5AQYRXJW4DMZ4VAY","short_pith_number":"pith:TAYZGEGC","schema_version":"1.0","canonical_sha256":"98319310c2f33a0862374db83667950627694c29e75b7eca5c1ec27a2f4a02c6","source":{"kind":"arxiv","id":"2608.04160","version":1},"attestation_state":"computed","paper":{"title":"Mind the Cap: Output-Budget Regimes Change the Measured Multilingual Reasoning Gap","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Ankit Goyal, Jaideep Ray","submitted_at":"2026-08-04T19:18:41Z","abstract_excerpt":"Multilingual evaluations report accuracy at a single output-token cap, but languages need different numbers of tokens to express the same content, so the cap is a hidden experimental variable. We test whether the native-vs-translate gap on MGSM (German, Thai, Swahili) is a token-budget artifact for Qwen3-8B and Llama-3.1-8B-Instruct under four prompting strategies. The measured gap swings by up to 57 points across budgets, length normalization moves it by up to 38.9 points where the cap binds, and at tight caps normalization can reverse which strategy scores higher. We prospectively froze the "},"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":"2608.04160","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-04T19:18:41Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c10b3bddd8e3771f063cbba80ddbf2e56fb48872810b619a01b63c98ac8050e3","abstract_canon_sha256":"5642bb93349590de82f5260e4959be34fb8c140571afae64ed8f6746d676d27d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-06T01:30:44.423037Z","signature_b64":"qHgONi7TwZl5QkmolFOO7wGCaLSa4jNWjSZjydXtKQsLh3vMr/r555aJ7mrJ3k8qd4ZX+K6M9MMNmJm9S9bXBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"98319310c2f33a0862374db83667950627694c29e75b7eca5c1ec27a2f4a02c6","last_reissued_at":"2026-08-06T01:30:44.421638Z","signature_status":"signed_v1","first_computed_at":"2026-08-06T01:30:44.421638Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mind the Cap: Output-Budget Regimes Change the Measured Multilingual Reasoning Gap","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Ankit Goyal, Jaideep Ray","submitted_at":"2026-08-04T19:18:41Z","abstract_excerpt":"Multilingual evaluations report accuracy at a single output-token cap, but languages need different numbers of tokens to express the same content, so the cap is a hidden experimental variable. We test whether the native-vs-translate gap on MGSM (German, Thai, Swahili) is a token-budget artifact for Qwen3-8B and Llama-3.1-8B-Instruct under four prompting strategies. The measured gap swings by up to 57 points across budgets, length normalization moves it by up to 38.9 points where the cap binds, and at tight caps normalization can reverse which strategy scores higher. We prospectively froze the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.04160","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/2608.04160/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":"2608.04160","created_at":"2026-08-06T01:30:44.423431+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.04160v1","created_at":"2026-08-06T01:30:44.423431+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.04160","created_at":"2026-08-06T01:30:44.423431+00:00"},{"alias_kind":"pith_short_12","alias_value":"TAYZGEGC6M5A","created_at":"2026-08-06T01:30:44.423431+00:00"},{"alias_kind":"pith_short_16","alias_value":"TAYZGEGC6M5AQYRX","created_at":"2026-08-06T01:30:44.423431+00:00"},{"alias_kind":"pith_short_8","alias_value":"TAYZGEGC","created_at":"2026-08-06T01:30:44.423431+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/TAYZGEGC6M5AQYRXJW4DMZ4VAY","json":"https://pith.science/pith/TAYZGEGC6M5AQYRXJW4DMZ4VAY.json","graph_json":"https://pith.science/api/pith-number/TAYZGEGC6M5AQYRXJW4DMZ4VAY/graph.json","events_json":"https://pith.science/api/pith-number/TAYZGEGC6M5AQYRXJW4DMZ4VAY/events.json","paper":"https://pith.science/paper/TAYZGEGC"},"agent_actions":{"view_html":"https://pith.science/pith/TAYZGEGC6M5AQYRXJW4DMZ4VAY","download_json":"https://pith.science/pith/TAYZGEGC6M5AQYRXJW4DMZ4VAY.json","view_paper":"https://pith.science/paper/TAYZGEGC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.04160&json=true","fetch_graph":"https://pith.science/api/pith-number/TAYZGEGC6M5AQYRXJW4DMZ4VAY/graph.json","fetch_events":"https://pith.science/api/pith-number/TAYZGEGC6M5AQYRXJW4DMZ4VAY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TAYZGEGC6M5AQYRXJW4DMZ4VAY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TAYZGEGC6M5AQYRXJW4DMZ4VAY/action/storage_attestation","attest_author":"https://pith.science/pith/TAYZGEGC6M5AQYRXJW4DMZ4VAY/action/author_attestation","sign_citation":"https://pith.science/pith/TAYZGEGC6M5AQYRXJW4DMZ4VAY/action/citation_signature","submit_replication":"https://pith.science/pith/TAYZGEGC6M5AQYRXJW4DMZ4VAY/action/replication_record"}},"created_at":"2026-08-06T01:30:44.423431+00:00","updated_at":"2026-08-06T01:30:44.423431+00:00"}