{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:A3IJGP364V3NLALH52N3BIYU2E","short_pith_number":"pith:A3IJGP36","schema_version":"1.0","canonical_sha256":"06d0933f7ee576d58167ee9bb0a314d10acfce3c43a1a0617ddab39ba3d0ba9b","source":{"kind":"arxiv","id":"2606.08520","version":1},"attestation_state":"computed","paper":{"title":"Two Bridges, One Pathway: From VLMs to Generalizable VLAs with Embodied Trajectory-Coupled Data","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Chenchen Yang, Chenxin Li, Jingjing Gong, Lei Xiao, Linqi Yin, Pengfang Qian, Shenling Qiu, Shiduo Zhang, Xiang Wang, Xipeng Qiu, Xuanjing Huang, Yu-Gang Jiang, Zhaoyang Fu, Zhe Xu","submitted_at":"2026-06-07T08:57:51Z","abstract_excerpt":"Vision-language models (VLMs) are powerful general-purpose reasoners, yet converting them into robot control policies (VLAs) is surprisingly difficult. The root cause is a two-fold gap: VLMs are trained on internet-scale images with language-understanding objectives, while VLAs must perceive robot scenes and predict motor actions. Fine-tuning a VLM directly on robot action data forces the model to cross both gaps at once -- the learning curve is steep and the rich generalizations learned during pretraining tend to degrade rather than transfer. We argue that this gap can be bridged gradually wi"},"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":"2606.08520","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.RO","submitted_at":"2026-06-07T08:57:51Z","cross_cats_sorted":[],"title_canon_sha256":"d4c5a26b755afea0011a78032f014bd3f7c3035107692159830ec5054a45dd25","abstract_canon_sha256":"ace618e8e29052fda2eff2c0b2c64b9d91674cb26004beeff9409a6b716a4173"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:05:38.935248Z","signature_b64":"HCnOI3DGPo4sg0c3xKIORpc+WlvNwaIm+T7hNPH36k32zwp/IFEhnC2QzFAmqcDZ3FA0YX65RvEWlBHTTzt8Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"06d0933f7ee576d58167ee9bb0a314d10acfce3c43a1a0617ddab39ba3d0ba9b","last_reissued_at":"2026-06-09T01:05:38.934812Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:05:38.934812Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Two Bridges, One Pathway: From VLMs to Generalizable VLAs with Embodied Trajectory-Coupled Data","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Chenchen Yang, Chenxin Li, Jingjing Gong, Lei Xiao, Linqi Yin, Pengfang Qian, Shenling Qiu, Shiduo Zhang, Xiang Wang, Xipeng Qiu, Xuanjing Huang, Yu-Gang Jiang, Zhaoyang Fu, Zhe Xu","submitted_at":"2026-06-07T08:57:51Z","abstract_excerpt":"Vision-language models (VLMs) are powerful general-purpose reasoners, yet converting them into robot control policies (VLAs) is surprisingly difficult. The root cause is a two-fold gap: VLMs are trained on internet-scale images with language-understanding objectives, while VLAs must perceive robot scenes and predict motor actions. Fine-tuning a VLM directly on robot action data forces the model to cross both gaps at once -- the learning curve is steep and the rich generalizations learned during pretraining tend to degrade rather than transfer. We argue that this gap can be bridged gradually wi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.08520","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/2606.08520/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":"2606.08520","created_at":"2026-06-09T01:05:38.934880+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.08520v1","created_at":"2026-06-09T01:05:38.934880+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.08520","created_at":"2026-06-09T01:05:38.934880+00:00"},{"alias_kind":"pith_short_12","alias_value":"A3IJGP364V3N","created_at":"2026-06-09T01:05:38.934880+00:00"},{"alias_kind":"pith_short_16","alias_value":"A3IJGP364V3NLALH","created_at":"2026-06-09T01:05:38.934880+00:00"},{"alias_kind":"pith_short_8","alias_value":"A3IJGP36","created_at":"2026-06-09T01:05:38.934880+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/A3IJGP364V3NLALH52N3BIYU2E","json":"https://pith.science/pith/A3IJGP364V3NLALH52N3BIYU2E.json","graph_json":"https://pith.science/api/pith-number/A3IJGP364V3NLALH52N3BIYU2E/graph.json","events_json":"https://pith.science/api/pith-number/A3IJGP364V3NLALH52N3BIYU2E/events.json","paper":"https://pith.science/paper/A3IJGP36"},"agent_actions":{"view_html":"https://pith.science/pith/A3IJGP364V3NLALH52N3BIYU2E","download_json":"https://pith.science/pith/A3IJGP364V3NLALH52N3BIYU2E.json","view_paper":"https://pith.science/paper/A3IJGP36","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.08520&json=true","fetch_graph":"https://pith.science/api/pith-number/A3IJGP364V3NLALH52N3BIYU2E/graph.json","fetch_events":"https://pith.science/api/pith-number/A3IJGP364V3NLALH52N3BIYU2E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A3IJGP364V3NLALH52N3BIYU2E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A3IJGP364V3NLALH52N3BIYU2E/action/storage_attestation","attest_author":"https://pith.science/pith/A3IJGP364V3NLALH52N3BIYU2E/action/author_attestation","sign_citation":"https://pith.science/pith/A3IJGP364V3NLALH52N3BIYU2E/action/citation_signature","submit_replication":"https://pith.science/pith/A3IJGP364V3NLALH52N3BIYU2E/action/replication_record"}},"created_at":"2026-06-09T01:05:38.934880+00:00","updated_at":"2026-06-09T01:05:38.934880+00:00"}