{"schemaVersion":1,"target":{"type":"reading","id":"5955562f957b1a1be86e3c9519a417bcd35c7ec0df24e601cf922d88ab13bccd","revision":"9f90a49f-3882-4d8c-8794-b54afd408f1e","revisionKind":"publication-id","title":"arXiv cs.AI: PACE: Pareto-Adaptive Compression for Efficient Native MLLMs","url":"https://escapemeridian.com/readings/arxiv-cs-ai-pace-pareto-adaptive-compression-for-efficient-5955562f"},"source":{"type":"source","id":"https://arxiv.org/abs/2610.24005","revision":"59120c026123833bbfcdb26b2497f481fa32811f77fbe32e8ef44344789224e4","revisionKind":"sha256-source-text","title":"PACE: Pareto-Adaptive Compression for Efficient Native MLLMs","url":"https://arxiv.org/abs/2610.24005"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter."}