{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:LJBFNZFCVT5YQTXBFDOARYXPBG","short_pith_number":"pith:LJBFNZFC","schema_version":"1.0","canonical_sha256":"5a4256e4a2acfb884ee128dc08e2ef09a6b4d0a4a1d7ab03a537a39d96ab4239","source":{"kind":"arxiv","id":"2509.04996","version":1},"attestation_state":"computed","paper":{"title":"FLOWER: Democratizing Generalist Robot Policies with Efficient Vision-Language-Action Flow Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Fabian Otto, Hongyi Zhou, Marcel R\\\"uhle, Moritz Reuss, \\\"Omer Erdin\\c{c} Ya\\u{g}murlu, Rudolf Lioutikov","submitted_at":"2025-09-05T10:43:12Z","abstract_excerpt":"Developing efficient Vision-Language-Action (VLA) policies is crucial for practical robotics deployment, yet current approaches face prohibitive computational costs and resource requirements. Existing diffusion-based VLA policies require multi-billion-parameter models and massive datasets to achieve strong performance. We tackle this efficiency challenge with two contributions: intermediate-modality fusion, which reallocates capacity to the diffusion head by pruning up to $50\\%$ of LLM layers, and action-specific Global-AdaLN conditioning, which cuts parameters by $20\\%$ through modular adapta"},"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":"2509.04996","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-09-05T10:43:12Z","cross_cats_sorted":[],"title_canon_sha256":"81775262f973727f132d85cab3adc12d15eab9f6c7599fcc5b803340411c99a9","abstract_canon_sha256":"8532249e7299100afa75c5b52c7133710f70a216fa8446cd494209e959985496"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:05:34.177449Z","signature_b64":"FVXl4ovRba6FG+dvaQP1ZuQZpFJLrX/8Pe9ba6XKW6WMTEhFb2ryZSZxFIcgpejSLGIUUOjwaPQ3qwReX3jDCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a4256e4a2acfb884ee128dc08e2ef09a6b4d0a4a1d7ab03a537a39d96ab4239","last_reissued_at":"2026-07-05T12:05:34.176952Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:05:34.176952Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FLOWER: Democratizing Generalist Robot Policies with Efficient Vision-Language-Action Flow Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Fabian Otto, Hongyi Zhou, Marcel R\\\"uhle, Moritz Reuss, \\\"Omer Erdin\\c{c} Ya\\u{g}murlu, Rudolf Lioutikov","submitted_at":"2025-09-05T10:43:12Z","abstract_excerpt":"Developing efficient Vision-Language-Action (VLA) policies is crucial for practical robotics deployment, yet current approaches face prohibitive computational costs and resource requirements. Existing diffusion-based VLA policies require multi-billion-parameter models and massive datasets to achieve strong performance. We tackle this efficiency challenge with two contributions: intermediate-modality fusion, which reallocates capacity to the diffusion head by pruning up to $50\\%$ of LLM layers, and action-specific Global-AdaLN conditioning, which cuts parameters by $20\\%$ through modular adapta"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.04996","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/2509.04996/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":"2509.04996","created_at":"2026-07-05T12:05:34.177010+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.04996v1","created_at":"2026-07-05T12:05:34.177010+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.04996","created_at":"2026-07-05T12:05:34.177010+00:00"},{"alias_kind":"pith_short_12","alias_value":"LJBFNZFCVT5Y","created_at":"2026-07-05T12:05:34.177010+00:00"},{"alias_kind":"pith_short_16","alias_value":"LJBFNZFCVT5YQTXB","created_at":"2026-07-05T12:05:34.177010+00:00"},{"alias_kind":"pith_short_8","alias_value":"LJBFNZFC","created_at":"2026-07-05T12:05:34.177010+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":20,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26588","citing_title":"Inference-Time Robot Behavior Steering through Physically-Aware Reconfiguration of Task-Structure","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22471","citing_title":"Scalable Multi-Task Data Generation via Reinforcement Learning for Language-Conditioned Bimanual Dexterous Manipulation","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21470","citing_title":"ASCII Art Turns LLMs into VLA Controllers","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19998","citing_title":"Tri-Info: Generalizable, Interpretable Failure Prediction for VLA Models via Information Theory","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17639","citing_title":"ERQA-Plus: A Diagnostic Benchmark for Reasoning in Embodied AI","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17924","citing_title":"PearlVLA: Progressive Embodied Action-Plan Refinement in Latent Space","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06245","citing_title":"MPCoT: Reward-Guided Multi-Path Latent Reasoning for Test-Time Scalable Vision-Language-Action","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06461","citing_title":"Flow-based Policy Adaptation without Policy Updates","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24890","citing_title":"QuoVLA: Quotient Space for Vision-Language-Action Models","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22471","citing_title":"Scalable Multi-Task Data Generation via Reinforcement Learning for Language-Conditioned Bimanual Dexterous Manipulation","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2602.18532","citing_title":"VLANeXt: Recipes for Building Strong VLA Models","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2603.14371","citing_title":"OxyGen: Unified KV Cache Management for VLA Inference under Multi-Task Parallelism","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21414","citing_title":"PointACT: Vision-Language-Action Models with Multi-Scale Point-Action Interaction","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2511.18960","citing_title":"AVA-VLA: Improving Vision-Language-Action models with Active Visual Attention","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2602.11236","citing_title":"ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learning","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2602.20309","citing_title":"QuantVLA: Scale-Calibrated Post-Training Quantization for Vision-Language-Action Models","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2512.15692","citing_title":"mimic-video: Video-Action Models for Generalizable Robot Control Beyond VLAs","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2603.25661","citing_title":"Fast-dVLA: Accelerating Discrete Diffusion VLA to Real-Time Performance","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2603.25903","citing_title":"Emergent Neural Automaton Policies: Learning Symbolic Structure from Visuomotor Trajectories","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24622","citing_title":"CF-VLA: Efficient Coarse-to-Fine Action Generation for Vision-Language-Action Policies","ref_index":35,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LJBFNZFCVT5YQTXBFDOARYXPBG","json":"https://pith.science/pith/LJBFNZFCVT5YQTXBFDOARYXPBG.json","graph_json":"https://pith.science/api/pith-number/LJBFNZFCVT5YQTXBFDOARYXPBG/graph.json","events_json":"https://pith.science/api/pith-number/LJBFNZFCVT5YQTXBFDOARYXPBG/events.json","paper":"https://pith.science/paper/LJBFNZFC"},"agent_actions":{"view_html":"https://pith.science/pith/LJBFNZFCVT5YQTXBFDOARYXPBG","download_json":"https://pith.science/pith/LJBFNZFCVT5YQTXBFDOARYXPBG.json","view_paper":"https://pith.science/paper/LJBFNZFC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.04996&json=true","fetch_graph":"https://pith.science/api/pith-number/LJBFNZFCVT5YQTXBFDOARYXPBG/graph.json","fetch_events":"https://pith.science/api/pith-number/LJBFNZFCVT5YQTXBFDOARYXPBG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LJBFNZFCVT5YQTXBFDOARYXPBG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LJBFNZFCVT5YQTXBFDOARYXPBG/action/storage_attestation","attest_author":"https://pith.science/pith/LJBFNZFCVT5YQTXBFDOARYXPBG/action/author_attestation","sign_citation":"https://pith.science/pith/LJBFNZFCVT5YQTXBFDOARYXPBG/action/citation_signature","submit_replication":"https://pith.science/pith/LJBFNZFCVT5YQTXBFDOARYXPBG/action/replication_record"}},"created_at":"2026-07-05T12:05:34.177010+00:00","updated_at":"2026-07-05T12:05:34.177010+00:00"}