{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:GHZAQT3GPWSPY6MZRMHCBVH2K2","short_pith_number":"pith:GHZAQT3G","schema_version":"1.0","canonical_sha256":"31f2084f667da4fc79998b0e20d4fa56abec7dc1fd48e4c8c4c6ed4eacf8da27","source":{"kind":"arxiv","id":"2111.07783","version":1},"attestation_state":"computed","paper":{"title":"FILIP: Fine-grained Interactive Language-Image Pre-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Chunjing Xu, Guansong Lu, Hang Xu, Lewei Yao, Lu Hou, Minzhe Niu, Runhui Huang, Xiaodan Liang, Xin Jiang, Zhenguo Li","submitted_at":"2021-11-09T17:15:38Z","abstract_excerpt":"Unsupervised large-scale vision-language pre-training has shown promising advances on various downstream tasks. Existing methods often model the cross-modal interaction either via the similarity of the global feature of each modality which misses sufficient information, or finer-grained interactions using cross/self-attention upon visual and textual tokens. However, cross/self-attention suffers from inferior efficiency in both training and inference. In this paper, we introduce a large-scale Fine-grained Interactive Language-Image Pre-training (FILIP) to achieve finer-level alignment through a"},"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":"2111.07783","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-11-09T17:15:38Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3ee72395c2fd37503d0a9c4ba92eb27d3631c43d89b22c5d1b011cebcbee9339","abstract_canon_sha256":"fc8a65690cfa2cd214e763f3fc0f3909c206d018f2a81dd4af55747f7e89addc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:31:43.705798Z","signature_b64":"XS/HvJM0q1y/rINZKmRkROxr6TzTAwDJJ5TbS2bCj1hZviKclPqFI42RsD4Obe37UewNtkaH9kW5nfpD/RaqAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"31f2084f667da4fc79998b0e20d4fa56abec7dc1fd48e4c8c4c6ed4eacf8da27","last_reissued_at":"2026-07-05T03:31:43.705306Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:31:43.705306Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FILIP: Fine-grained Interactive Language-Image Pre-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Chunjing Xu, Guansong Lu, Hang Xu, Lewei Yao, Lu Hou, Minzhe Niu, Runhui Huang, Xiaodan Liang, Xin Jiang, Zhenguo Li","submitted_at":"2021-11-09T17:15:38Z","abstract_excerpt":"Unsupervised large-scale vision-language pre-training has shown promising advances on various downstream tasks. Existing methods often model the cross-modal interaction either via the similarity of the global feature of each modality which misses sufficient information, or finer-grained interactions using cross/self-attention upon visual and textual tokens. However, cross/self-attention suffers from inferior efficiency in both training and inference. In this paper, we introduce a large-scale Fine-grained Interactive Language-Image Pre-training (FILIP) to achieve finer-level alignment through a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2111.07783","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/2111.07783/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":"2111.07783","created_at":"2026-07-05T03:31:43.705376+00:00"},{"alias_kind":"arxiv_version","alias_value":"2111.07783v1","created_at":"2026-07-05T03:31:43.705376+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.07783","created_at":"2026-07-05T03:31:43.705376+00:00"},{"alias_kind":"pith_short_12","alias_value":"GHZAQT3GPWSP","created_at":"2026-07-05T03:31:43.705376+00:00"},{"alias_kind":"pith_short_16","alias_value":"GHZAQT3GPWSPY6MZ","created_at":"2026-07-05T03:31:43.705376+00:00"},{"alias_kind":"pith_short_8","alias_value":"GHZAQT3G","created_at":"2026-07-05T03:31:43.705376+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":28,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2607.05727","citing_title":"SAMPLe: SAM-based Optimizer for Prompt Learning in VLMs","ref_index":42,"is_internal_anchor":true},{"citing_arxiv_id":"2607.07907","citing_title":"Multimodal Unlearning Across Vision, Language, Video, and Audio: Survey of Methods, Datasets, and Benchmarks","ref_index":44,"is_internal_anchor":true},{"citing_arxiv_id":"2606.23475","citing_title":"Multi-Vector Embeddings are Provably More Expressive than Single Vector Embeddings","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18885","citing_title":"LARE: Low-Attention Region Encoding for Text-Image Retrieval","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2607.02484","citing_title":"Combating Textual Noise and Redundancy: Entropy-Aware Dense Visual Token Pruning","ref_index":63,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25922","citing_title":"Closed-Loop Bidirectional Prompting for Adversarial Robustness of Vision Language Models","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2510.16335","citing_title":"On the Provable Importance of Gradients for Language-Assisted Image Clustering","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2208.14649","citing_title":"DetailCLIP: Injecting Image Details into CLIP's Feature Space","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2410.10247","citing_title":"LPT: Less-overfitting Prompt Tuning for Vision-Language Model","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2505.13777","citing_title":"Sat2Sound: A Unified Framework for Zero-Shot Soundscape Mapping","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15615","citing_title":"Neutral-Reference Prompting for Vision-Language Models","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18018","citing_title":"See What I Mean: Aligning Vision and Language Representations for Video Fine-grained Object Understanding","ref_index":83,"is_internal_anchor":false},{"citing_arxiv_id":"2511.13415","citing_title":"Attention Grounded Enhancement for Visual Document Retrieval","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2212.03191","citing_title":"InternVideo: General Video Foundation Models via Generative and Discriminative Learning","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2111.11432","citing_title":"Florence: A New Foundation Model for Computer Vision","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2603.09921","citing_title":"WikiCLIP: An Efficient Contrastive Baseline for Open-domain Visual Entity Recognition","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2205.01917","citing_title":"CoCa: Contrastive Captioners are Image-Text Foundation Models","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2307.06942","citing_title":"InternVid: A Large-scale Video-Text Dataset for Multimodal Understanding and Generation","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2305.06355","citing_title":"VideoChat: Chat-Centric Video Understanding","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11939","citing_title":"Cluster-Aware Neural Collapse Prompt Tuning for Long-Tailed Generalization of Vision-Language Models","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08814","citing_title":"Zero-Shot Chinese Character Recognition via Global-Local Dual-Branch Alignment and Hierarchical Inference","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2204.14198","citing_title":"Flamingo: a Visual Language Model for Few-Shot Learning","ref_index":139,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10130","citing_title":"Thermal-Det: Language-Guided Cross-Modal Distillation for Open-Vocabulary Thermal Object Detection","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06229","citing_title":"Look Beyond Saliency: Low-Attention Guided Dual Encoding for Video Semantic Search","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04425","citing_title":"Joint Semantic Token Selection and Prompt Optimization for Interpretable Prompt Learning","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GHZAQT3GPWSPY6MZRMHCBVH2K2","json":"https://pith.science/pith/GHZAQT3GPWSPY6MZRMHCBVH2K2.json","graph_json":"https://pith.science/api/pith-number/GHZAQT3GPWSPY6MZRMHCBVH2K2/graph.json","events_json":"https://pith.science/api/pith-number/GHZAQT3GPWSPY6MZRMHCBVH2K2/events.json","paper":"https://pith.science/paper/GHZAQT3G"},"agent_actions":{"view_html":"https://pith.science/pith/GHZAQT3GPWSPY6MZRMHCBVH2K2","download_json":"https://pith.science/pith/GHZAQT3GPWSPY6MZRMHCBVH2K2.json","view_paper":"https://pith.science/paper/GHZAQT3G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2111.07783&json=true","fetch_graph":"https://pith.science/api/pith-number/GHZAQT3GPWSPY6MZRMHCBVH2K2/graph.json","fetch_events":"https://pith.science/api/pith-number/GHZAQT3GPWSPY6MZRMHCBVH2K2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GHZAQT3GPWSPY6MZRMHCBVH2K2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GHZAQT3GPWSPY6MZRMHCBVH2K2/action/storage_attestation","attest_author":"https://pith.science/pith/GHZAQT3GPWSPY6MZRMHCBVH2K2/action/author_attestation","sign_citation":"https://pith.science/pith/GHZAQT3GPWSPY6MZRMHCBVH2K2/action/citation_signature","submit_replication":"https://pith.science/pith/GHZAQT3GPWSPY6MZRMHCBVH2K2/action/replication_record"}},"created_at":"2026-07-05T03:31:43.705376+00:00","updated_at":"2026-07-05T03:31:43.705376+00:00"}