{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:MQNZRO47JQNGCCKDFBTIE3FCYU","short_pith_number":"pith:MQNZRO47","schema_version":"1.0","canonical_sha256":"641b98bb9f4c1a6109432866826ca2c52aea61653933ebb7face3ec6bfd7bfe0","source":{"kind":"arxiv","id":"2410.10733","version":8},"attestation_state":"computed","paper":{"title":"Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Enze Xie, Han Cai, Haotian Tang, Junsong Chen, Junyu Chen, Muyang Li, Shang Yang, Song Han, Yao Lu","submitted_at":"2024-10-14T17:15:07Z","abstract_excerpt":"We present Deep Compression Autoencoder (DC-AE), a new family of autoencoder models for accelerating high-resolution diffusion models. Existing autoencoder models have demonstrated impressive results at a moderate spatial compression ratio (e.g., 8x), but fail to maintain satisfactory reconstruction accuracy for high spatial compression ratios (e.g., 64x). We address this challenge by introducing two key techniques: (1) Residual Autoencoding, where we design our models to learn residuals based on the space-to-channel transformed features to alleviate the optimization difficulty of high spatial"},"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":"2410.10733","kind":"arxiv","version":8},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-10-14T17:15:07Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c92634aca93537e7fa9164d9062660b3f3472f4167643d24575e79fb7b363840","abstract_canon_sha256":"0c9b0ca5147a895f1a1d06c9393b89dd53651a900a75b590012fabdb14febd83"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:04:40.499335Z","signature_b64":"ltRLz/dz1ZaX4AUbZWtZo1xmPsSBMVJWacu47Y1vZakoNHIIvKKeCjiUUrVJzmHRBH7NSLjE62M0A98ugqlhCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"641b98bb9f4c1a6109432866826ca2c52aea61653933ebb7face3ec6bfd7bfe0","last_reissued_at":"2026-07-05T11:04:40.498867Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:04:40.498867Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Enze Xie, Han Cai, Haotian Tang, Junsong Chen, Junyu Chen, Muyang Li, Shang Yang, Song Han, Yao Lu","submitted_at":"2024-10-14T17:15:07Z","abstract_excerpt":"We present Deep Compression Autoencoder (DC-AE), a new family of autoencoder models for accelerating high-resolution diffusion models. Existing autoencoder models have demonstrated impressive results at a moderate spatial compression ratio (e.g., 8x), but fail to maintain satisfactory reconstruction accuracy for high spatial compression ratios (e.g., 64x). We address this challenge by introducing two key techniques: (1) Residual Autoencoding, where we design our models to learn residuals based on the space-to-channel transformed features to alleviate the optimization difficulty of high spatial"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.10733","kind":"arxiv","version":8},"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/2410.10733/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":"2410.10733","created_at":"2026-07-05T11:04:40.498924+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.10733v8","created_at":"2026-07-05T11:04:40.498924+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.10733","created_at":"2026-07-05T11:04:40.498924+00:00"},{"alias_kind":"pith_short_12","alias_value":"MQNZRO47JQNG","created_at":"2026-07-05T11:04:40.498924+00:00"},{"alias_kind":"pith_short_16","alias_value":"MQNZRO47JQNGCCKD","created_at":"2026-07-05T11:04:40.498924+00:00"},{"alias_kind":"pith_short_8","alias_value":"MQNZRO47","created_at":"2026-07-05T11:04:40.498924+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":31,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07565","citing_title":"Collaborative Synthetic Data Generation for Knowledge Transfer in Federated Learning","ref_index":77,"is_internal_anchor":true},{"citing_arxiv_id":"2606.23041","citing_title":"SPAR: Semantic-Pixel Self-Alignment and Adaptive Routing for Unified Multimodal Models","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23041","citing_title":"SPAR: Semantic-Pixel Self-Alignment and Adaptive Routing for Unified Multimodal Models","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03578","citing_title":"Diffusing in the Right Space: A Systematic Study of Latent Diffusability","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01620","citing_title":"Real-Time Generation of Streamable Talking Portrait Video with Reference-Guided Deep Compression VAEs","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14531","citing_title":"Language Generation as Optimal Control: Closed-Loop Diffusion in Latent Control Space","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20708","citing_title":"Rethinking Cross-Layer Information Routing in Diffusion Transformers","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28758","citing_title":"X-Mind: Efficient Visual Chain-of-Thought via Predictive World Model for End-to-End Driving","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23902","citing_title":"PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusion","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19385","citing_title":"LatentBox: Storing AI-Generated Images at Scale via a Latent-First Design","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20808","citing_title":"Spatial Gram Alignment for Ultra-High-Resolution Image Synthesis","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20708","citing_title":"Rethinking Cross-Layer Information Routing in Diffusion Transformers","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2512.14692","citing_title":"Native and Compact Structured Latents for 3D Generation","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15684","citing_title":"ElasticDiT: Efficient Diffusion Transformers via Elastic Architecture and Sparse Attention for High-Resolution Image Generation on Mobile Devices","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19385","citing_title":"LatentBox: Storing AI-Generated Images at Scale via a Latent-First Design","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14531","citing_title":"Language Generation as Optimal Control: Closed-Loop Diffusion in Latent Control Space","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2510.26782","citing_title":"Cloning Deterministic Worlds: The Critical Role of Latent Geometry in Long-Horizon World Models","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2511.19365","citing_title":"DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2503.19325","citing_title":"Long-Context Autoregressive Video Modeling with Next-Frame Prediction","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2602.02493","citing_title":"PixelGen: Improving Pixel Diffusion with Perceptual Supervision","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2503.22020","citing_title":"CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15178","citing_title":"SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion Transformer","ref_index":83,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14531","citing_title":"Language Generation as Optimal Control: Closed-Loop Diffusion in Latent Control Space","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2603.26357","citing_title":"MPDiT: Multi-Patch Global-to-Local Transformer Architecture For Efficient Flow Matching and Diffusion Model","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03064","citing_title":"Gram-MMD: A Texture-Aware Metric for Image Realism Assessment","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MQNZRO47JQNGCCKDFBTIE3FCYU","json":"https://pith.science/pith/MQNZRO47JQNGCCKDFBTIE3FCYU.json","graph_json":"https://pith.science/api/pith-number/MQNZRO47JQNGCCKDFBTIE3FCYU/graph.json","events_json":"https://pith.science/api/pith-number/MQNZRO47JQNGCCKDFBTIE3FCYU/events.json","paper":"https://pith.science/paper/MQNZRO47"},"agent_actions":{"view_html":"https://pith.science/pith/MQNZRO47JQNGCCKDFBTIE3FCYU","download_json":"https://pith.science/pith/MQNZRO47JQNGCCKDFBTIE3FCYU.json","view_paper":"https://pith.science/paper/MQNZRO47","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.10733&json=true","fetch_graph":"https://pith.science/api/pith-number/MQNZRO47JQNGCCKDFBTIE3FCYU/graph.json","fetch_events":"https://pith.science/api/pith-number/MQNZRO47JQNGCCKDFBTIE3FCYU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MQNZRO47JQNGCCKDFBTIE3FCYU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MQNZRO47JQNGCCKDFBTIE3FCYU/action/storage_attestation","attest_author":"https://pith.science/pith/MQNZRO47JQNGCCKDFBTIE3FCYU/action/author_attestation","sign_citation":"https://pith.science/pith/MQNZRO47JQNGCCKDFBTIE3FCYU/action/citation_signature","submit_replication":"https://pith.science/pith/MQNZRO47JQNGCCKDFBTIE3FCYU/action/replication_record"}},"created_at":"2026-07-05T11:04:40.498924+00:00","updated_at":"2026-07-05T11:04:40.498924+00:00"}