{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ULGARW6GJFQ4JHXMJFMMMEOZOC","short_pith_number":"pith:ULGARW6G","schema_version":"1.0","canonical_sha256":"a2cc08dbc64961c49eec4958c611d97080a3f91413a10cc159c9a93dd21c1a98","source":{"kind":"arxiv","id":"2608.01665","version":1},"attestation_state":"computed","paper":{"title":"Allocation Before Ranking: Decoupled Token Compression for OmniLLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"cs.AI","authors_text":"Chengming Zhang, Miao Yin, Omprakash Gnawali, Rabimba Karanjai, Weidong Shi, Yilin Yang, Yuanbin Man, Zhenghui Guo","submitted_at":"2026-08-03T03:56:33Z","abstract_excerpt":"Token compression in OmniLLMs is typically posed as a single saliency-ranking problem: score each multimodal token, keep the top-K. We argue this abstraction is mis-specified. The same attention score simultaneously decides two things: how much retained capacity each modality receives, and which tokens within a modality are kept. A shared top-K rule therefore inherits this audio-favoring allocation prior, spending retained capacity on audio before video tokens have a chance to compete. We propose Macer, a training-free compressor that first assigns explicit audio and video budgets, then perfor"},"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":"2608.01665","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-03T03:56:33Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"ab2054c13027a420e241603a5810b2d7a45b948a6139f47316e18757bb0740ea","abstract_canon_sha256":"16053d24854cfd74d0c30f631723eff873dd8958ed8012ecaa5e26649bbbe00a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T02:06:22.406971Z","signature_b64":"EeNWOX4H3b6yi/0kfxMz8idFM7Y4r3JSkFCkjBLc+uPBEH5DZN+Spz6951bY7JIxgPmYNjkVuNp6YWQz+OMuAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a2cc08dbc64961c49eec4958c611d97080a3f91413a10cc159c9a93dd21c1a98","last_reissued_at":"2026-08-04T02:06:22.405453Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T02:06:22.405453Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Allocation Before Ranking: Decoupled Token Compression for OmniLLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"cs.AI","authors_text":"Chengming Zhang, Miao Yin, Omprakash Gnawali, Rabimba Karanjai, Weidong Shi, Yilin Yang, Yuanbin Man, Zhenghui Guo","submitted_at":"2026-08-03T03:56:33Z","abstract_excerpt":"Token compression in OmniLLMs is typically posed as a single saliency-ranking problem: score each multimodal token, keep the top-K. We argue this abstraction is mis-specified. The same attention score simultaneously decides two things: how much retained capacity each modality receives, and which tokens within a modality are kept. A shared top-K rule therefore inherits this audio-favoring allocation prior, spending retained capacity on audio before video tokens have a chance to compete. We propose Macer, a training-free compressor that first assigns explicit audio and video budgets, then perfor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.01665","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/2608.01665/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":"2608.01665","created_at":"2026-08-04T02:06:22.406848+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.01665v1","created_at":"2026-08-04T02:06:22.406848+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.01665","created_at":"2026-08-04T02:06:22.406848+00:00"},{"alias_kind":"pith_short_12","alias_value":"ULGARW6GJFQ4","created_at":"2026-08-04T02:06:22.406848+00:00"},{"alias_kind":"pith_short_16","alias_value":"ULGARW6GJFQ4JHXM","created_at":"2026-08-04T02:06:22.406848+00:00"},{"alias_kind":"pith_short_8","alias_value":"ULGARW6G","created_at":"2026-08-04T02:06:22.406848+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/ULGARW6GJFQ4JHXMJFMMMEOZOC","json":"https://pith.science/pith/ULGARW6GJFQ4JHXMJFMMMEOZOC.json","graph_json":"https://pith.science/api/pith-number/ULGARW6GJFQ4JHXMJFMMMEOZOC/graph.json","events_json":"https://pith.science/api/pith-number/ULGARW6GJFQ4JHXMJFMMMEOZOC/events.json","paper":"https://pith.science/paper/ULGARW6G"},"agent_actions":{"view_html":"https://pith.science/pith/ULGARW6GJFQ4JHXMJFMMMEOZOC","download_json":"https://pith.science/pith/ULGARW6GJFQ4JHXMJFMMMEOZOC.json","view_paper":"https://pith.science/paper/ULGARW6G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.01665&json=true","fetch_graph":"https://pith.science/api/pith-number/ULGARW6GJFQ4JHXMJFMMMEOZOC/graph.json","fetch_events":"https://pith.science/api/pith-number/ULGARW6GJFQ4JHXMJFMMMEOZOC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ULGARW6GJFQ4JHXMJFMMMEOZOC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ULGARW6GJFQ4JHXMJFMMMEOZOC/action/storage_attestation","attest_author":"https://pith.science/pith/ULGARW6GJFQ4JHXMJFMMMEOZOC/action/author_attestation","sign_citation":"https://pith.science/pith/ULGARW6GJFQ4JHXMJFMMMEOZOC/action/citation_signature","submit_replication":"https://pith.science/pith/ULGARW6GJFQ4JHXMJFMMMEOZOC/action/replication_record"}},"created_at":"2026-08-04T02:06:22.406848+00:00","updated_at":"2026-08-04T02:06:22.406848+00:00"}