{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:QCST7RHSGYJX6JI2IPKCYALP3H","short_pith_number":"pith:QCST7RHS","schema_version":"1.0","canonical_sha256":"80a53fc4f236137f251a43d42c016fd9d22b84e294515fb363de65f8d4cbbe5c","source":{"kind":"arxiv","id":"2605.17093","version":1},"attestation_state":"computed","paper":{"title":"HEED: Density-Weighted Residual Alignment for Hybrid Vision-Language Model Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Niraj K. Jha, Yihao Liang","submitted_at":"2026-05-16T17:33:24Z","abstract_excerpt":"Distilling vision-language models into faster hybrid architectures, such as 3:1 Mamba-2/attention mixes, is now standard practice for making inference efficient. Aggregate benchmarks suggest that this works but they hide selective failures. When we distill Qwen3-VL-8B-Instruct into a 3:1 Mamba-2/attention hybrid, student model stays within 2 points of the teacher across visual reasoning benchmarks like MMStar, MMBench, and MMMU-Pro, while dropping 13 points on optical-character-recognition and document tasks. The student can still understand the scene but loses the fine-grained text needed to "},"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":"2605.17093","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-16T17:33:24Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"7e9dbc2eaa848cdcc74f2b8fffe3ff888505f8bbd7e6c17ad612278e7b2ce3bc","abstract_canon_sha256":"2c447ed347cd2a2711367cac9225ff32011bc371f65cc138d3b1d41bcbb13b42"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:03:39.786975Z","signature_b64":"bLvuVhp5G9t5e7V6815NXKJ2C3TgEEb0SBHIPWns0tQ9KLZx5ttCjvVwPqWQnwrc/RsE8ZdRMmqWAHAJdyKyAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"80a53fc4f236137f251a43d42c016fd9d22b84e294515fb363de65f8d4cbbe5c","last_reissued_at":"2026-05-20T00:03:39.786310Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:03:39.786310Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HEED: Density-Weighted Residual Alignment for Hybrid Vision-Language Model Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Niraj K. Jha, Yihao Liang","submitted_at":"2026-05-16T17:33:24Z","abstract_excerpt":"Distilling vision-language models into faster hybrid architectures, such as 3:1 Mamba-2/attention mixes, is now standard practice for making inference efficient. Aggregate benchmarks suggest that this works but they hide selective failures. When we distill Qwen3-VL-8B-Instruct into a 3:1 Mamba-2/attention hybrid, student model stays within 2 points of the teacher across visual reasoning benchmarks like MMStar, MMBench, and MMMU-Pro, while dropping 13 points on optical-character-recognition and document tasks. The student can still understand the scene but loses the fine-grained text needed to "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.17093","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/2605.17093/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T22:33:23.800514Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T22:21:57.734545Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"35746a90979cd8063f4bdef402bfb1234eed8ebc174748f9578a96c486aa0dde"},"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":"2605.17093","created_at":"2026-05-20T00:03:39.786423+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.17093v1","created_at":"2026-05-20T00:03:39.786423+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17093","created_at":"2026-05-20T00:03:39.786423+00:00"},{"alias_kind":"pith_short_12","alias_value":"QCST7RHSGYJX","created_at":"2026-05-20T00:03:39.786423+00:00"},{"alias_kind":"pith_short_16","alias_value":"QCST7RHSGYJX6JI2","created_at":"2026-05-20T00:03:39.786423+00:00"},{"alias_kind":"pith_short_8","alias_value":"QCST7RHS","created_at":"2026-05-20T00:03:39.786423+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/QCST7RHSGYJX6JI2IPKCYALP3H","json":"https://pith.science/pith/QCST7RHSGYJX6JI2IPKCYALP3H.json","graph_json":"https://pith.science/api/pith-number/QCST7RHSGYJX6JI2IPKCYALP3H/graph.json","events_json":"https://pith.science/api/pith-number/QCST7RHSGYJX6JI2IPKCYALP3H/events.json","paper":"https://pith.science/paper/QCST7RHS"},"agent_actions":{"view_html":"https://pith.science/pith/QCST7RHSGYJX6JI2IPKCYALP3H","download_json":"https://pith.science/pith/QCST7RHSGYJX6JI2IPKCYALP3H.json","view_paper":"https://pith.science/paper/QCST7RHS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.17093&json=true","fetch_graph":"https://pith.science/api/pith-number/QCST7RHSGYJX6JI2IPKCYALP3H/graph.json","fetch_events":"https://pith.science/api/pith-number/QCST7RHSGYJX6JI2IPKCYALP3H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QCST7RHSGYJX6JI2IPKCYALP3H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QCST7RHSGYJX6JI2IPKCYALP3H/action/storage_attestation","attest_author":"https://pith.science/pith/QCST7RHSGYJX6JI2IPKCYALP3H/action/author_attestation","sign_citation":"https://pith.science/pith/QCST7RHSGYJX6JI2IPKCYALP3H/action/citation_signature","submit_replication":"https://pith.science/pith/QCST7RHSGYJX6JI2IPKCYALP3H/action/replication_record"}},"created_at":"2026-05-20T00:03:39.786423+00:00","updated_at":"2026-05-20T00:03:39.786423+00:00"}