{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DQBHY7OX3O5OIMICAQ766E64RA","short_pith_number":"pith:DQBHY7OX","schema_version":"1.0","canonical_sha256":"1c027c7dd7dbbae43102043fef13dc882c71bd4d7319cefc4f6b8094bbbef2f3","source":{"kind":"arxiv","id":"2606.31201","version":1},"attestation_state":"computed","paper":{"title":"ExPLoRe: Expert Patch-Level Loss Routing for Multi-Objective Masked Image Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hairong Qi, Konstantinos Georgiou, Maofeng Tang","submitted_at":"2026-06-30T06:31:18Z","abstract_excerpt":"Multi-objective masked image modeling (MIM) combines complementary learning signals (token distillation, CLS alignment, and pixel reconstruction) but existing methods weight these objectives with global scalars, ignoring spatial heterogeneity across patches. We present ExPLoRe (Expert Patch-Level Loss Routing), which repurposes Soft Mixture of Experts (MoE) dispatch weights as learned, per-patch loss coefficients. The key mechanism is loss-coupling: allowing loss gradients to flow through dispatch weights to the router enables content-dependent specialization, where different patches receive d"},"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":"2606.31201","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-30T06:31:18Z","cross_cats_sorted":[],"title_canon_sha256":"27a7a73cac3d11393fa7587b01716fd4a9df8194e19ddb2fa2a2d2600298fa42","abstract_canon_sha256":"cd4fb846834174d67cd886d94455e1ceda5ce9a2a69b5fa00f3141a30f9a6360"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-01T01:17:32.079234Z","signature_b64":"Y1BvY000Mswsru1qcenEqsIOFMLT4SB062wihSmRY7FYrBr06OYp4siolNXuEvhlKb+antJpppJIZvrtwxhMBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1c027c7dd7dbbae43102043fef13dc882c71bd4d7319cefc4f6b8094bbbef2f3","last_reissued_at":"2026-07-01T01:17:32.078792Z","signature_status":"signed_v1","first_computed_at":"2026-07-01T01:17:32.078792Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ExPLoRe: Expert Patch-Level Loss Routing for Multi-Objective Masked Image Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hairong Qi, Konstantinos Georgiou, Maofeng Tang","submitted_at":"2026-06-30T06:31:18Z","abstract_excerpt":"Multi-objective masked image modeling (MIM) combines complementary learning signals (token distillation, CLS alignment, and pixel reconstruction) but existing methods weight these objectives with global scalars, ignoring spatial heterogeneity across patches. We present ExPLoRe (Expert Patch-Level Loss Routing), which repurposes Soft Mixture of Experts (MoE) dispatch weights as learned, per-patch loss coefficients. The key mechanism is loss-coupling: allowing loss gradients to flow through dispatch weights to the router enables content-dependent specialization, where different patches receive d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.31201","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/2606.31201/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":"2606.31201","created_at":"2026-07-01T01:17:32.078861+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.31201v1","created_at":"2026-07-01T01:17:32.078861+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.31201","created_at":"2026-07-01T01:17:32.078861+00:00"},{"alias_kind":"pith_short_12","alias_value":"DQBHY7OX3O5O","created_at":"2026-07-01T01:17:32.078861+00:00"},{"alias_kind":"pith_short_16","alias_value":"DQBHY7OX3O5OIMIC","created_at":"2026-07-01T01:17:32.078861+00:00"},{"alias_kind":"pith_short_8","alias_value":"DQBHY7OX","created_at":"2026-07-01T01:17:32.078861+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/DQBHY7OX3O5OIMICAQ766E64RA","json":"https://pith.science/pith/DQBHY7OX3O5OIMICAQ766E64RA.json","graph_json":"https://pith.science/api/pith-number/DQBHY7OX3O5OIMICAQ766E64RA/graph.json","events_json":"https://pith.science/api/pith-number/DQBHY7OX3O5OIMICAQ766E64RA/events.json","paper":"https://pith.science/paper/DQBHY7OX"},"agent_actions":{"view_html":"https://pith.science/pith/DQBHY7OX3O5OIMICAQ766E64RA","download_json":"https://pith.science/pith/DQBHY7OX3O5OIMICAQ766E64RA.json","view_paper":"https://pith.science/paper/DQBHY7OX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.31201&json=true","fetch_graph":"https://pith.science/api/pith-number/DQBHY7OX3O5OIMICAQ766E64RA/graph.json","fetch_events":"https://pith.science/api/pith-number/DQBHY7OX3O5OIMICAQ766E64RA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DQBHY7OX3O5OIMICAQ766E64RA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DQBHY7OX3O5OIMICAQ766E64RA/action/storage_attestation","attest_author":"https://pith.science/pith/DQBHY7OX3O5OIMICAQ766E64RA/action/author_attestation","sign_citation":"https://pith.science/pith/DQBHY7OX3O5OIMICAQ766E64RA/action/citation_signature","submit_replication":"https://pith.science/pith/DQBHY7OX3O5OIMICAQ766E64RA/action/replication_record"}},"created_at":"2026-07-01T01:17:32.078861+00:00","updated_at":"2026-07-01T01:17:32.078861+00:00"}