{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:75WZHTFZEN4XB7R32ZEYYMIDOK","short_pith_number":"pith:75WZHTFZ","schema_version":"1.0","canonical_sha256":"ff6d93ccb9237970fe3bd6498c310372bff798d0cae24870b9097e3136d87316","source":{"kind":"arxiv","id":"2508.21046","version":3},"attestation_state":"computed","paper":{"title":"CogVLA: Cognition-Aligned Vision-Language-Action Model via Instruction-Driven Routing & Sparsification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.CV","authors_text":"Jie He, Liqiang Nie, Renshan Zhang, Rui Shao, Wei Li","submitted_at":"2025-08-28T17:50:58Z","abstract_excerpt":"Recent Vision-Language-Action (VLA) models built on pre-trained Vision-Language Models (VLMs) require extensive post-training, resulting in high computational overhead that limits scalability and deployment.We propose CogVLA, a Cognition-Aligned Vision-Language-Action framework that leverages instruction-driven routing and sparsification to improve both efficiency and performance. CogVLA draws inspiration from human multimodal coordination and introduces a 3-stage progressive architecture. 1) Encoder-FiLM based Aggregation Routing (EFA-Routing) injects instruction information into the vision e"},"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":"2508.21046","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-28T17:50:58Z","cross_cats_sorted":["cs.RO"],"title_canon_sha256":"f450bfefbe3cf088e6fbce69b47b503c5c73cceef41d26bea0125871759a78cb","abstract_canon_sha256":"a6e0f01a98c3179002aaf8782f4f44029e483ef363942b6f5c6f9c610e528767"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:30.091663Z","signature_b64":"4/+T/i/RzYfrb83cetA66kdtwfprOcwhT4R1r+7PxxO4IkF2+nODaLi99xpFiIwu8rg/oaM3vlMaJzhPtAd+AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff6d93ccb9237970fe3bd6498c310372bff798d0cae24870b9097e3136d87316","last_reissued_at":"2026-05-28T01:04:30.090927Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:30.090927Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CogVLA: Cognition-Aligned Vision-Language-Action Model via Instruction-Driven Routing & Sparsification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.CV","authors_text":"Jie He, Liqiang Nie, Renshan Zhang, Rui Shao, Wei Li","submitted_at":"2025-08-28T17:50:58Z","abstract_excerpt":"Recent Vision-Language-Action (VLA) models built on pre-trained Vision-Language Models (VLMs) require extensive post-training, resulting in high computational overhead that limits scalability and deployment.We propose CogVLA, a Cognition-Aligned Vision-Language-Action framework that leverages instruction-driven routing and sparsification to improve both efficiency and performance. CogVLA draws inspiration from human multimodal coordination and introduces a 3-stage progressive architecture. 1) Encoder-FiLM based Aggregation Routing (EFA-Routing) injects instruction information into the vision e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.21046","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/2508.21046/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":"2508.21046","created_at":"2026-05-28T01:04:30.091043+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.21046v3","created_at":"2026-05-28T01:04:30.091043+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.21046","created_at":"2026-05-28T01:04:30.091043+00:00"},{"alias_kind":"pith_short_12","alias_value":"75WZHTFZEN4X","created_at":"2026-05-28T01:04:30.091043+00:00"},{"alias_kind":"pith_short_16","alias_value":"75WZHTFZEN4XB7R3","created_at":"2026-05-28T01:04:30.091043+00:00"},{"alias_kind":"pith_short_8","alias_value":"75WZHTFZ","created_at":"2026-05-28T01:04:30.091043+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":10,"sample":[{"citing_arxiv_id":"2605.23856","citing_title":"Point Tracking Improves World Action Models","ref_index":79,"is_internal_anchor":true},{"citing_arxiv_id":"2602.20200","citing_title":"Global Prior Meets Local Consistency: Dual-Memory Augmented Vision-Language-Action Model for Efficient Robotic Manipulation","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2508.13073","citing_title":"Large VLM-based Vision-Language-Action Models for Robotic Manipulation: A Survey","ref_index":118,"is_internal_anchor":true},{"citing_arxiv_id":"2511.18960","citing_title":"AVA-VLA: Improving Vision-Language-Action models with Active Visual Attention","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05714","citing_title":"TriRelVLA: Triadic Relational Structure for Generalizable Embodied Manipulation","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05126","citing_title":"ConsisVLA-4D: Advancing Spatiotemporal Consistency in Efficient 3D-Perception and 4D-Reasoning for Robotic Manipulation","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2604.08168","citing_title":"ViVa: A Video-Generative Value Model for Robot Reinforcement Learning","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2604.09330","citing_title":"VAG: Dual-Stream Video-Action Generation for Embodied Data Synthesis","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2605.07794","citing_title":"NoiseGate: Learning Per-Latent Timestep Schedules as Information Gating in World Action Models","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2604.06067","citing_title":"HiPolicy: Hierarchical Multi-Frequency Action Chunking for Policy Learning","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/75WZHTFZEN4XB7R32ZEYYMIDOK","json":"https://pith.science/pith/75WZHTFZEN4XB7R32ZEYYMIDOK.json","graph_json":"https://pith.science/api/pith-number/75WZHTFZEN4XB7R32ZEYYMIDOK/graph.json","events_json":"https://pith.science/api/pith-number/75WZHTFZEN4XB7R32ZEYYMIDOK/events.json","paper":"https://pith.science/paper/75WZHTFZ"},"agent_actions":{"view_html":"https://pith.science/pith/75WZHTFZEN4XB7R32ZEYYMIDOK","download_json":"https://pith.science/pith/75WZHTFZEN4XB7R32ZEYYMIDOK.json","view_paper":"https://pith.science/paper/75WZHTFZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.21046&json=true","fetch_graph":"https://pith.science/api/pith-number/75WZHTFZEN4XB7R32ZEYYMIDOK/graph.json","fetch_events":"https://pith.science/api/pith-number/75WZHTFZEN4XB7R32ZEYYMIDOK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/75WZHTFZEN4XB7R32ZEYYMIDOK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/75WZHTFZEN4XB7R32ZEYYMIDOK/action/storage_attestation","attest_author":"https://pith.science/pith/75WZHTFZEN4XB7R32ZEYYMIDOK/action/author_attestation","sign_citation":"https://pith.science/pith/75WZHTFZEN4XB7R32ZEYYMIDOK/action/citation_signature","submit_replication":"https://pith.science/pith/75WZHTFZEN4XB7R32ZEYYMIDOK/action/replication_record"}},"created_at":"2026-05-28T01:04:30.091043+00:00","updated_at":"2026-05-28T01:04:30.091043+00:00"}