{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RDF6FFPHSTALWJO3LKBKZXWKG7","short_pith_number":"pith:RDF6FFPH","schema_version":"1.0","canonical_sha256":"88cbe295e794c0bb25db5a82acdeca37c06101d82dca15c645ce297ec25ab3e2","source":{"kind":"arxiv","id":"2505.10088","version":1},"attestation_state":"computed","paper":{"title":"MMRL++: Parameter-Efficient and Interaction-Aware Representation Learning for Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Xiaodong Gu, Yuncheng Guo","submitted_at":"2025-05-15T08:43:53Z","abstract_excerpt":"Large-scale pre-trained Vision-Language Models (VLMs) have significantly advanced transfer learning across diverse tasks. However, adapting these models with limited few-shot data often leads to overfitting, undermining their ability to generalize to new tasks. To address this, we propose Multi-Modal Representation Learning (MMRL), which introduces a shared, learnable, modality-agnostic representation space. MMRL generates space tokens projected into both text and image encoders as representation tokens, enabling more effective cross-modal interactions. Unlike prior methods that mainly optimiz"},"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":"2505.10088","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-15T08:43:53Z","cross_cats_sorted":[],"title_canon_sha256":"7bb711fffeef273151b2c3234abc52517ff0581f11efe5a996454462b3c98fcb","abstract_canon_sha256":"74e7a283ee78eca02cc049929e255e429dcc23b6704c8e7d5daa4617a4b0941f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:03:33.010547Z","signature_b64":"BgquagnYz347sVDDBa+NntaGr47K9FphR4Uo6jfq6ev4IQm3B0E6GoWvdzpi3AjhDyT7Z5y+K4THGXznj68+Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"88cbe295e794c0bb25db5a82acdeca37c06101d82dca15c645ce297ec25ab3e2","last_reissued_at":"2026-07-05T11:03:33.010077Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:03:33.010077Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MMRL++: Parameter-Efficient and Interaction-Aware Representation Learning for Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Xiaodong Gu, Yuncheng Guo","submitted_at":"2025-05-15T08:43:53Z","abstract_excerpt":"Large-scale pre-trained Vision-Language Models (VLMs) have significantly advanced transfer learning across diverse tasks. However, adapting these models with limited few-shot data often leads to overfitting, undermining their ability to generalize to new tasks. To address this, we propose Multi-Modal Representation Learning (MMRL), which introduces a shared, learnable, modality-agnostic representation space. MMRL generates space tokens projected into both text and image encoders as representation tokens, enabling more effective cross-modal interactions. Unlike prior methods that mainly optimiz"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.10088","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/2505.10088/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":"2505.10088","created_at":"2026-07-05T11:03:33.010131+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.10088v1","created_at":"2026-07-05T11:03:33.010131+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.10088","created_at":"2026-07-05T11:03:33.010131+00:00"},{"alias_kind":"pith_short_12","alias_value":"RDF6FFPHSTAL","created_at":"2026-07-05T11:03:33.010131+00:00"},{"alias_kind":"pith_short_16","alias_value":"RDF6FFPHSTALWJO3","created_at":"2026-07-05T11:03:33.010131+00:00"},{"alias_kind":"pith_short_8","alias_value":"RDF6FFPH","created_at":"2026-07-05T11:03:33.010131+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.13161","citing_title":"A$_3$B$_2$: Adaptive Asymmetric Adapter for Alleviating Branch Bias in Vision-Language Image Classification with Few-Shot Learning","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13161","citing_title":"A$_3$B$_2$: Adaptive Asymmetric Adapter for Alleviating Branch Bias in Vision-Language Image Classification with Few-Shot Learning","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RDF6FFPHSTALWJO3LKBKZXWKG7","json":"https://pith.science/pith/RDF6FFPHSTALWJO3LKBKZXWKG7.json","graph_json":"https://pith.science/api/pith-number/RDF6FFPHSTALWJO3LKBKZXWKG7/graph.json","events_json":"https://pith.science/api/pith-number/RDF6FFPHSTALWJO3LKBKZXWKG7/events.json","paper":"https://pith.science/paper/RDF6FFPH"},"agent_actions":{"view_html":"https://pith.science/pith/RDF6FFPHSTALWJO3LKBKZXWKG7","download_json":"https://pith.science/pith/RDF6FFPHSTALWJO3LKBKZXWKG7.json","view_paper":"https://pith.science/paper/RDF6FFPH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.10088&json=true","fetch_graph":"https://pith.science/api/pith-number/RDF6FFPHSTALWJO3LKBKZXWKG7/graph.json","fetch_events":"https://pith.science/api/pith-number/RDF6FFPHSTALWJO3LKBKZXWKG7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RDF6FFPHSTALWJO3LKBKZXWKG7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RDF6FFPHSTALWJO3LKBKZXWKG7/action/storage_attestation","attest_author":"https://pith.science/pith/RDF6FFPHSTALWJO3LKBKZXWKG7/action/author_attestation","sign_citation":"https://pith.science/pith/RDF6FFPHSTALWJO3LKBKZXWKG7/action/citation_signature","submit_replication":"https://pith.science/pith/RDF6FFPHSTALWJO3LKBKZXWKG7/action/replication_record"}},"created_at":"2026-07-05T11:03:33.010131+00:00","updated_at":"2026-07-05T11:03:33.010131+00:00"}