{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OY3YYKGUXU6P7AWEHCLZ5GXEN5","short_pith_number":"pith:OY3YYKGU","schema_version":"1.0","canonical_sha256":"76378c28d4bd3cff82c438979e9ae46f46364f98b3b7d748d45dd51965304b20","source":{"kind":"arxiv","id":"2606.00390","version":1},"attestation_state":"computed","paper":{"title":"Zamba2-VL Technical Report","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Beren Millidge, Hassan Shapourian, Kasra Hejazi, Olabode M. Sule","submitted_at":"2026-05-29T22:12:40Z","abstract_excerpt":"We present Zamba2-VL, a suite of vision-language models built on Zamba2, a hybrid language-model architecture combining Mamba2 state-space layers with a small number of shared transformer blocks. Across a broad range of image understanding, reasoning, OCR, grounding, and counting benchmarks, Zamba2-VL is competitive with leading Transformer-based open-weight VLMs of comparable scale, including the Molmo2, Qwen3-VL, and InternVL3.5 families, and substantially outperforms prior SSM-based and hybrid VLMs such as VL-Mamba, Cobra, and mmMamba. Inheriting the near-linear prefill compute and small, n"},"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":"2606.00390","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-29T22:12:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f61e8b7dc3bef8d8dc7c0f38c27fa25a24d14fe91dd0ec55ca44794321f1d74a","abstract_canon_sha256":"f6380ed86e5152e008ee94e80d494eea35e2d66866ebd9c221622cae3ce6207d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:53.314199Z","signature_b64":"PTH9LI4o4HOXORCDWidDuBMGqIjemVS0/T+dEpzqtorzVtKCjUGvUgLwcmmjSC1j3Wv3TzJnuWBQo9c/1m5kDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"76378c28d4bd3cff82c438979e9ae46f46364f98b3b7d748d45dd51965304b20","last_reissued_at":"2026-06-02T01:03:53.313770Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:53.313770Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Zamba2-VL Technical Report","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Beren Millidge, Hassan Shapourian, Kasra Hejazi, Olabode M. Sule","submitted_at":"2026-05-29T22:12:40Z","abstract_excerpt":"We present Zamba2-VL, a suite of vision-language models built on Zamba2, a hybrid language-model architecture combining Mamba2 state-space layers with a small number of shared transformer blocks. Across a broad range of image understanding, reasoning, OCR, grounding, and counting benchmarks, Zamba2-VL is competitive with leading Transformer-based open-weight VLMs of comparable scale, including the Molmo2, Qwen3-VL, and InternVL3.5 families, and substantially outperforms prior SSM-based and hybrid VLMs such as VL-Mamba, Cobra, and mmMamba. Inheriting the near-linear prefill compute and small, n"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00390","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/2606.00390/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":"2606.00390","created_at":"2026-06-02T01:03:53.313834+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.00390v1","created_at":"2026-06-02T01:03:53.313834+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00390","created_at":"2026-06-02T01:03:53.313834+00:00"},{"alias_kind":"pith_short_12","alias_value":"OY3YYKGUXU6P","created_at":"2026-06-02T01:03:53.313834+00:00"},{"alias_kind":"pith_short_16","alias_value":"OY3YYKGUXU6P7AWE","created_at":"2026-06-02T01:03:53.313834+00:00"},{"alias_kind":"pith_short_8","alias_value":"OY3YYKGU","created_at":"2026-06-02T01:03:53.313834+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/OY3YYKGUXU6P7AWEHCLZ5GXEN5","json":"https://pith.science/pith/OY3YYKGUXU6P7AWEHCLZ5GXEN5.json","graph_json":"https://pith.science/api/pith-number/OY3YYKGUXU6P7AWEHCLZ5GXEN5/graph.json","events_json":"https://pith.science/api/pith-number/OY3YYKGUXU6P7AWEHCLZ5GXEN5/events.json","paper":"https://pith.science/paper/OY3YYKGU"},"agent_actions":{"view_html":"https://pith.science/pith/OY3YYKGUXU6P7AWEHCLZ5GXEN5","download_json":"https://pith.science/pith/OY3YYKGUXU6P7AWEHCLZ5GXEN5.json","view_paper":"https://pith.science/paper/OY3YYKGU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.00390&json=true","fetch_graph":"https://pith.science/api/pith-number/OY3YYKGUXU6P7AWEHCLZ5GXEN5/graph.json","fetch_events":"https://pith.science/api/pith-number/OY3YYKGUXU6P7AWEHCLZ5GXEN5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OY3YYKGUXU6P7AWEHCLZ5GXEN5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OY3YYKGUXU6P7AWEHCLZ5GXEN5/action/storage_attestation","attest_author":"https://pith.science/pith/OY3YYKGUXU6P7AWEHCLZ5GXEN5/action/author_attestation","sign_citation":"https://pith.science/pith/OY3YYKGUXU6P7AWEHCLZ5GXEN5/action/citation_signature","submit_replication":"https://pith.science/pith/OY3YYKGUXU6P7AWEHCLZ5GXEN5/action/replication_record"}},"created_at":"2026-06-02T01:03:53.313834+00:00","updated_at":"2026-06-02T01:03:53.313834+00:00"}