{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GWHYNDZPVKCNO3DC2G57JM3RCF","short_pith_number":"pith:GWHYNDZP","schema_version":"1.0","canonical_sha256":"358f868f2faa84d76c62d1bbf4b371116b69dcb9f8414d31f3600f99aa919337","source":{"kind":"arxiv","id":"2608.02791","version":1},"attestation_state":"computed","paper":{"title":"Better, Stronger, Faster, and Broader: Structured All-Mask Prediction for MLLM-Based Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiazhen Liu, Long Chen, Mingkuan Feng","submitted_at":"2026-08-03T18:42:11Z","abstract_excerpt":"MLLM-based segmentation faces a core segmentation trilemma: high segmentation performance, preserved dialogue ability, and fast inference. Embedding-prediction methods may disrupt language modeling through pixel-level objectives, whereas next-token generation is inefficient for dense masks. We propose All-Mask Prediction, decoupling autoregressive dialogue from non-autoregressive mask prediction. Its binary instantiation, STAMP (Simultaneous Textual All-Mask Prediction), emits an in-vocabulary <SEG> trigger, fuses image-aligned mask tokens with corresponding patch features, and uses hybrid att"},"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":"2608.02791","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-08-03T18:42:11Z","cross_cats_sorted":[],"title_canon_sha256":"a8dd99c4292c6b0041de49ff67222239bd5be433ab57e3033b48b354d409cded","abstract_canon_sha256":"a8a0201d0f3de018c4cf60e688445b6ad93f4baa5d82f8b76fe5d3808fdaf30e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-05T00:40:04.269707Z","signature_b64":"wx/Ytc53yeF3SlH2U2/RGvPswlTdpReZxT7Ts+anilgmLCL4pBlWBqMtDZ0Idm4p0RSMEtUCSi47Q6GnO+WGDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"358f868f2faa84d76c62d1bbf4b371116b69dcb9f8414d31f3600f99aa919337","last_reissued_at":"2026-08-05T00:40:04.266424Z","signature_status":"signed_v1","first_computed_at":"2026-08-05T00:40:04.266424Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Better, Stronger, Faster, and Broader: Structured All-Mask Prediction for MLLM-Based Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiazhen Liu, Long Chen, Mingkuan Feng","submitted_at":"2026-08-03T18:42:11Z","abstract_excerpt":"MLLM-based segmentation faces a core segmentation trilemma: high segmentation performance, preserved dialogue ability, and fast inference. Embedding-prediction methods may disrupt language modeling through pixel-level objectives, whereas next-token generation is inefficient for dense masks. We propose All-Mask Prediction, decoupling autoregressive dialogue from non-autoregressive mask prediction. Its binary instantiation, STAMP (Simultaneous Textual All-Mask Prediction), emits an in-vocabulary <SEG> trigger, fuses image-aligned mask tokens with corresponding patch features, and uses hybrid att"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.02791","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/2608.02791/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":"2608.02791","created_at":"2026-08-05T00:40:04.267781+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.02791v1","created_at":"2026-08-05T00:40:04.267781+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.02791","created_at":"2026-08-05T00:40:04.267781+00:00"},{"alias_kind":"pith_short_12","alias_value":"GWHYNDZPVKCN","created_at":"2026-08-05T00:40:04.267781+00:00"},{"alias_kind":"pith_short_16","alias_value":"GWHYNDZPVKCNO3DC","created_at":"2026-08-05T00:40:04.267781+00:00"},{"alias_kind":"pith_short_8","alias_value":"GWHYNDZP","created_at":"2026-08-05T00:40:04.267781+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/GWHYNDZPVKCNO3DC2G57JM3RCF","json":"https://pith.science/pith/GWHYNDZPVKCNO3DC2G57JM3RCF.json","graph_json":"https://pith.science/api/pith-number/GWHYNDZPVKCNO3DC2G57JM3RCF/graph.json","events_json":"https://pith.science/api/pith-number/GWHYNDZPVKCNO3DC2G57JM3RCF/events.json","paper":"https://pith.science/paper/GWHYNDZP"},"agent_actions":{"view_html":"https://pith.science/pith/GWHYNDZPVKCNO3DC2G57JM3RCF","download_json":"https://pith.science/pith/GWHYNDZPVKCNO3DC2G57JM3RCF.json","view_paper":"https://pith.science/paper/GWHYNDZP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.02791&json=true","fetch_graph":"https://pith.science/api/pith-number/GWHYNDZPVKCNO3DC2G57JM3RCF/graph.json","fetch_events":"https://pith.science/api/pith-number/GWHYNDZPVKCNO3DC2G57JM3RCF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GWHYNDZPVKCNO3DC2G57JM3RCF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GWHYNDZPVKCNO3DC2G57JM3RCF/action/storage_attestation","attest_author":"https://pith.science/pith/GWHYNDZPVKCNO3DC2G57JM3RCF/action/author_attestation","sign_citation":"https://pith.science/pith/GWHYNDZPVKCNO3DC2G57JM3RCF/action/citation_signature","submit_replication":"https://pith.science/pith/GWHYNDZPVKCNO3DC2G57JM3RCF/action/replication_record"}},"created_at":"2026-08-05T00:40:04.267781+00:00","updated_at":"2026-08-05T00:40:04.267781+00:00"}