{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:EYCLPWTQQ5PTFO6Z7RSF3LDBVJ","short_pith_number":"pith:EYCLPWTQ","canonical_record":{"source":{"id":"2603.11679","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-03-12T08:44:06Z","cross_cats_sorted":[],"title_canon_sha256":"0f866a8f167ce6d21c5a063a403c6d28bcc7da6dac5d841e1e684c89ae587b6b","abstract_canon_sha256":"2e96a02dccad3457b90e6c4e19e2d56c37a73aacb411ccd7947c8b3c133d8e1c"},"schema_version":"1.0"},"canonical_sha256":"2604b7da70875f32bbd9fc645dac61aa6046093b419087993cb9d5ced8e883d5","source":{"kind":"arxiv","id":"2603.11679","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.11679","created_at":"2026-05-22T01:03:18Z"},{"alias_kind":"arxiv_version","alias_value":"2603.11679v3","created_at":"2026-05-22T01:03:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.11679","created_at":"2026-05-22T01:03:18Z"},{"alias_kind":"pith_short_12","alias_value":"EYCLPWTQQ5PT","created_at":"2026-05-22T01:03:18Z"},{"alias_kind":"pith_short_16","alias_value":"EYCLPWTQQ5PTFO6Z","created_at":"2026-05-22T01:03:18Z"},{"alias_kind":"pith_short_8","alias_value":"EYCLPWTQ","created_at":"2026-05-22T01:03:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:EYCLPWTQQ5PTFO6Z7RSF3LDBVJ","target":"record","payload":{"canonical_record":{"source":{"id":"2603.11679","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-03-12T08:44:06Z","cross_cats_sorted":[],"title_canon_sha256":"0f866a8f167ce6d21c5a063a403c6d28bcc7da6dac5d841e1e684c89ae587b6b","abstract_canon_sha256":"2e96a02dccad3457b90e6c4e19e2d56c37a73aacb411ccd7947c8b3c133d8e1c"},"schema_version":"1.0"},"canonical_sha256":"2604b7da70875f32bbd9fc645dac61aa6046093b419087993cb9d5ced8e883d5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:03:18.703011Z","signature_b64":"eoyP2A6h/HZFcm3okbOIDCqsEBmpw+Tw0SmgmYpzInSIX12lRld+1AtM5f5Df/X702xRMhGMsS3KLIOGdKxvAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2604b7da70875f32bbd9fc645dac61aa6046093b419087993cb9d5ced8e883d5","last_reissued_at":"2026-05-22T01:03:18.702144Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:03:18.702144Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2603.11679","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T01:03:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GrObdfrxjbVLTHM+oRDBy+x9NG85waNqb8gXP+xXdgNmL/7XnF1akfE355TQ3kIk+juQ44ZUvTurpGvW6ESyBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T19:33:06.228750Z"},"content_sha256":"217313391377e763293b33048af5d82734f3ca68833e43f32d34c78e641d3720","schema_version":"1.0","event_id":"sha256:217313391377e763293b33048af5d82734f3ca68833e43f32d34c78e641d3720"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:EYCLPWTQQ5PTFO6Z7RSF3LDBVJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LLMs can construct powerful representations and streamline sample-efficient supervised learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"David Sontag, Ilker Demirel, Lawrence Shi, Zeshan Hussain","submitted_at":"2026-03-12T08:44:06Z","abstract_excerpt":"As real-world datasets become more complex and heterogeneous, supervised learning is often bottlenecked by input representation design. Modeling multimodal data, such as time-series, free text, and structured records, often requires non-trivial domain expertise. We propose an agentic pipeline to streamline this process. First, an LLM analyzes a small but diverse subset of text-serialized input examples in-context to synthesize a global rubric, which acts as a programmatic specification for extracting and organizing evidence. This rubric is then used to transform naive text-serializations of in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.11679","kind":"arxiv","version":3},"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/2603.11679/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T01:03:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IRCQHOeSeLyezdInBDccBNKIRJteDmXKL6NbeO3mq1rzseCojtHHteRGCkp9PA/JgwEzt2LqKy/lI+nBKJHWBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T19:33:06.229267Z"},"content_sha256":"61305d2f0cbe89afddaa49871237a83bf98347f62f5ebd7a1b99c1754a531895","schema_version":"1.0","event_id":"sha256:61305d2f0cbe89afddaa49871237a83bf98347f62f5ebd7a1b99c1754a531895"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EYCLPWTQQ5PTFO6Z7RSF3LDBVJ/bundle.json","state_url":"https://pith.science/pith/EYCLPWTQQ5PTFO6Z7RSF3LDBVJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EYCLPWTQQ5PTFO6Z7RSF3LDBVJ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-25T19:33:06Z","links":{"resolver":"https://pith.science/pith/EYCLPWTQQ5PTFO6Z7RSF3LDBVJ","bundle":"https://pith.science/pith/EYCLPWTQQ5PTFO6Z7RSF3LDBVJ/bundle.json","state":"https://pith.science/pith/EYCLPWTQQ5PTFO6Z7RSF3LDBVJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EYCLPWTQQ5PTFO6Z7RSF3LDBVJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:EYCLPWTQQ5PTFO6Z7RSF3LDBVJ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"2e96a02dccad3457b90e6c4e19e2d56c37a73aacb411ccd7947c8b3c133d8e1c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-03-12T08:44:06Z","title_canon_sha256":"0f866a8f167ce6d21c5a063a403c6d28bcc7da6dac5d841e1e684c89ae587b6b"},"schema_version":"1.0","source":{"id":"2603.11679","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.11679","created_at":"2026-05-22T01:03:18Z"},{"alias_kind":"arxiv_version","alias_value":"2603.11679v3","created_at":"2026-05-22T01:03:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.11679","created_at":"2026-05-22T01:03:18Z"},{"alias_kind":"pith_short_12","alias_value":"EYCLPWTQQ5PT","created_at":"2026-05-22T01:03:18Z"},{"alias_kind":"pith_short_16","alias_value":"EYCLPWTQQ5PTFO6Z","created_at":"2026-05-22T01:03:18Z"},{"alias_kind":"pith_short_8","alias_value":"EYCLPWTQ","created_at":"2026-05-22T01:03:18Z"}],"graph_snapshots":[{"event_id":"sha256:61305d2f0cbe89afddaa49871237a83bf98347f62f5ebd7a1b99c1754a531895","target":"graph","created_at":"2026-05-22T01:03:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2603.11679/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As real-world datasets become more complex and heterogeneous, supervised learning is often bottlenecked by input representation design. Modeling multimodal data, such as time-series, free text, and structured records, often requires non-trivial domain expertise. We propose an agentic pipeline to streamline this process. First, an LLM analyzes a small but diverse subset of text-serialized input examples in-context to synthesize a global rubric, which acts as a programmatic specification for extracting and organizing evidence. This rubric is then used to transform naive text-serializations of in","authors_text":"David Sontag, Ilker Demirel, Lawrence Shi, Zeshan Hussain","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-03-12T08:44:06Z","title":"LLMs can construct powerful representations and streamline sample-efficient supervised learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.11679","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:217313391377e763293b33048af5d82734f3ca68833e43f32d34c78e641d3720","target":"record","created_at":"2026-05-22T01:03:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"2e96a02dccad3457b90e6c4e19e2d56c37a73aacb411ccd7947c8b3c133d8e1c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-03-12T08:44:06Z","title_canon_sha256":"0f866a8f167ce6d21c5a063a403c6d28bcc7da6dac5d841e1e684c89ae587b6b"},"schema_version":"1.0","source":{"id":"2603.11679","kind":"arxiv","version":3}},"canonical_sha256":"2604b7da70875f32bbd9fc645dac61aa6046093b419087993cb9d5ced8e883d5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2604b7da70875f32bbd9fc645dac61aa6046093b419087993cb9d5ced8e883d5","first_computed_at":"2026-05-22T01:03:18.702144Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T01:03:18.702144Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eoyP2A6h/HZFcm3okbOIDCqsEBmpw+Tw0SmgmYpzInSIX12lRld+1AtM5f5Df/X702xRMhGMsS3KLIOGdKxvAw==","signature_status":"signed_v1","signed_at":"2026-05-22T01:03:18.703011Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.11679","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:217313391377e763293b33048af5d82734f3ca68833e43f32d34c78e641d3720","sha256:61305d2f0cbe89afddaa49871237a83bf98347f62f5ebd7a1b99c1754a531895"],"state_sha256":"89634c6259025b1539c3fbf850304edad20a9da97be035b23f457bcd008defb3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JhEWN4A+RZ00TPLUnQxoW8tmqcTVsM7n7Osr1Rgc2CJ3S6j1zOZ8mMocXP0snD8Y8lb+pRQfqns+1cFJGSowBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T19:33:06.232176Z","bundle_sha256":"5a17b4afa3f2cd912a7fd18968d9e8b02ee93317f883d755d11c3590b0d08a41"}}