{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BQXNM7WFQQCKJADML5NTJKY6MG","short_pith_number":"pith:BQXNM7WF","schema_version":"1.0","canonical_sha256":"0c2ed67ec58404a4806c5f5b34ab1e619c22bdc59317888736469f175603d18f","source":{"kind":"arxiv","id":"2608.04243","version":1},"attestation_state":"computed","paper":{"title":"Attention-based representations for multi-task computation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Daniel Hsu, Mingyue Xu","submitted_at":"2026-08-04T21:52:04Z","abstract_excerpt":"Multi-head attention layers produce vector representations that support multiple downstream tasks. We establish bounds on the number of heads required in two simple and concrete multi-task scenarios. In the first scenario, a vector representation is sought so that linear predictors can compute both the smallest and largest numbers in a given list. In this case, it is known two attention heads with small embedding dimension and bit precision level suffice. We prove that a single attention head requires exponentially higher embedding dimension or precision level. In the second scenario, a vector"},"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":"2608.04243","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-08-04T21:52:04Z","cross_cats_sorted":[],"title_canon_sha256":"88b7394c81c0f36fc83abda920bc520f97e5d4995594fa5110373722b57943a5","abstract_canon_sha256":"cbc4140e76079835d03078df4ea3d7b765eab79e3f8de31c9c8cc726ce11f7dc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-06T01:31:44.215416Z","signature_b64":"a1ZZeDQFOXmjqrfb4rQRh7BVElemmruRheBEtaZyBIF2dElyuRCMER16A45HgyPbKZCmTa+8t9FAZD6xplvBBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0c2ed67ec58404a4806c5f5b34ab1e619c22bdc59317888736469f175603d18f","last_reissued_at":"2026-08-06T01:31:44.213926Z","signature_status":"signed_v1","first_computed_at":"2026-08-06T01:31:44.213926Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Attention-based representations for multi-task computation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Daniel Hsu, Mingyue Xu","submitted_at":"2026-08-04T21:52:04Z","abstract_excerpt":"Multi-head attention layers produce vector representations that support multiple downstream tasks. We establish bounds on the number of heads required in two simple and concrete multi-task scenarios. In the first scenario, a vector representation is sought so that linear predictors can compute both the smallest and largest numbers in a given list. In this case, it is known two attention heads with small embedding dimension and bit precision level suffice. We prove that a single attention head requires exponentially higher embedding dimension or precision level. In the second scenario, a vector"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.04243","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/2608.04243/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":"2608.04243","created_at":"2026-08-06T01:31:44.215736+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.04243v1","created_at":"2026-08-06T01:31:44.215736+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.04243","created_at":"2026-08-06T01:31:44.215736+00:00"},{"alias_kind":"pith_short_12","alias_value":"BQXNM7WFQQCK","created_at":"2026-08-06T01:31:44.215736+00:00"},{"alias_kind":"pith_short_16","alias_value":"BQXNM7WFQQCKJADM","created_at":"2026-08-06T01:31:44.215736+00:00"},{"alias_kind":"pith_short_8","alias_value":"BQXNM7WF","created_at":"2026-08-06T01:31:44.215736+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/BQXNM7WFQQCKJADML5NTJKY6MG","json":"https://pith.science/pith/BQXNM7WFQQCKJADML5NTJKY6MG.json","graph_json":"https://pith.science/api/pith-number/BQXNM7WFQQCKJADML5NTJKY6MG/graph.json","events_json":"https://pith.science/api/pith-number/BQXNM7WFQQCKJADML5NTJKY6MG/events.json","paper":"https://pith.science/paper/BQXNM7WF"},"agent_actions":{"view_html":"https://pith.science/pith/BQXNM7WFQQCKJADML5NTJKY6MG","download_json":"https://pith.science/pith/BQXNM7WFQQCKJADML5NTJKY6MG.json","view_paper":"https://pith.science/paper/BQXNM7WF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.04243&json=true","fetch_graph":"https://pith.science/api/pith-number/BQXNM7WFQQCKJADML5NTJKY6MG/graph.json","fetch_events":"https://pith.science/api/pith-number/BQXNM7WFQQCKJADML5NTJKY6MG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BQXNM7WFQQCKJADML5NTJKY6MG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BQXNM7WFQQCKJADML5NTJKY6MG/action/storage_attestation","attest_author":"https://pith.science/pith/BQXNM7WFQQCKJADML5NTJKY6MG/action/author_attestation","sign_citation":"https://pith.science/pith/BQXNM7WFQQCKJADML5NTJKY6MG/action/citation_signature","submit_replication":"https://pith.science/pith/BQXNM7WFQQCKJADML5NTJKY6MG/action/replication_record"}},"created_at":"2026-08-06T01:31:44.215736+00:00","updated_at":"2026-08-06T01:31:44.215736+00:00"}