{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WZIS3STU3NQD6UVGVNJUUSRVJE","short_pith_number":"pith:WZIS3STU","schema_version":"1.0","canonical_sha256":"b6512dca74db603f52a6ab534a4a3549061c9d48643ae457356f2363335dd817","source":{"kind":"arxiv","id":"2507.02860","version":1},"attestation_state":"computed","paper":{"title":"Less is Enough: Training-Free Video Diffusion Acceleration via Runtime-Adaptive Caching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dingkang Liang, Feiyang Tan, Hengshuang Zhao, Hongkai Lin, Kaijin Chen, Tianrui Feng, Xiang Bai, Xin Zhou, Xiwu Chen, Yikang Ding","submitted_at":"2025-07-03T17:59:54Z","abstract_excerpt":"Video generation models have demonstrated remarkable performance, yet their broader adoption remains constrained by slow inference speeds and substantial computational costs, primarily due to the iterative nature of the denoising process. Addressing this bottleneck is essential for democratizing advanced video synthesis technologies and enabling their integration into real-world applications. This work proposes EasyCache, a training-free acceleration framework for video diffusion models. EasyCache introduces a lightweight, runtime-adaptive caching mechanism that dynamically reuses previously c"},"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":"2507.02860","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-03T17:59:54Z","cross_cats_sorted":[],"title_canon_sha256":"d15b8fe6c18edfa297ef17a71df160628d03135730f538916d17f5bb8c9dc43a","abstract_canon_sha256":"346da847d91ac2b1fe13cf9e9f4af7953c896453d0ac9fb946aa3c3118dd5168"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:31:35.550278Z","signature_b64":"YsUuaWFT80CASLkp1HYVu09H1ahCdRkRb+HjkAWH5iZnONVv611MEQ/efWo/TzEzRAYmC8SyYN++k8lBexB4CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b6512dca74db603f52a6ab534a4a3549061c9d48643ae457356f2363335dd817","last_reissued_at":"2026-07-05T11:31:35.549822Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:31:35.549822Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Less is Enough: Training-Free Video Diffusion Acceleration via Runtime-Adaptive Caching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dingkang Liang, Feiyang Tan, Hengshuang Zhao, Hongkai Lin, Kaijin Chen, Tianrui Feng, Xiang Bai, Xin Zhou, Xiwu Chen, Yikang Ding","submitted_at":"2025-07-03T17:59:54Z","abstract_excerpt":"Video generation models have demonstrated remarkable performance, yet their broader adoption remains constrained by slow inference speeds and substantial computational costs, primarily due to the iterative nature of the denoising process. Addressing this bottleneck is essential for democratizing advanced video synthesis technologies and enabling their integration into real-world applications. This work proposes EasyCache, a training-free acceleration framework for video diffusion models. EasyCache introduces a lightweight, runtime-adaptive caching mechanism that dynamically reuses previously c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.02860","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/2507.02860/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":"2507.02860","created_at":"2026-07-05T11:31:35.549877+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.02860v1","created_at":"2026-07-05T11:31:35.549877+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.02860","created_at":"2026-07-05T11:31:35.549877+00:00"},{"alias_kind":"pith_short_12","alias_value":"WZIS3STU3NQD","created_at":"2026-07-05T11:31:35.549877+00:00"},{"alias_kind":"pith_short_16","alias_value":"WZIS3STU3NQD6UVG","created_at":"2026-07-05T11:31:35.549877+00:00"},{"alias_kind":"pith_short_8","alias_value":"WZIS3STU","created_at":"2026-07-05T11:31:35.549877+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26778","citing_title":"LearniBridge: Learnable Calibration of Feature Caching for Diffusion Models Acceleration","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.26795","citing_title":"NaviCache: Test-Time Self-Calibration Caching for Video Generation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23743","citing_title":"Sol Video Inference Engine: Agent-Native Full-Stack Acceleration Framework for Efficient Video Generation","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06309","citing_title":"RhymeFlow: Training-Free Acceleration for Video Generation with Asynchronous Denoising Flow Scheduling","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06060","citing_title":"ReCache: Learning Budget-Aware Caching Schedules for Diffusion Models via REINFORCE","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30849","citing_title":"SyncCache: Exploiting Asymmetric Dynamics for Fast Audio-Driven Portrait Animation","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25801","citing_title":"PixelWizard: Towards Efficient High-Fidelity Video Generation at Ultra-Large Spatial Resolution","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23381","citing_title":"VDE: Training-Free Accelerating Rectified Flow Model via Velocity Decomposition and Estimation","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16789","citing_title":"Accelerating Rectified Flow Models via Trajectory-Aware Caching","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04018","citing_title":"1.x-Distill: Breaking the Diversity, Quality, and Efficiency Barrier in Distribution Matching Distillation","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06892","citing_title":"Not All Tokens Need 40 Steps: Heterogeneous Step Allocation in Diffusion Transformers for Efficient Video Generation","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08546","citing_title":"When Numbers Speak: Aligning Textual Numerals and Visual Instances in Text-to-Video Diffusion Models","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02757","citing_title":"Seeing Realism from Simulation: Efficient Video Transfer for Vision-Language-Action Data Augmentation","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WZIS3STU3NQD6UVGVNJUUSRVJE","json":"https://pith.science/pith/WZIS3STU3NQD6UVGVNJUUSRVJE.json","graph_json":"https://pith.science/api/pith-number/WZIS3STU3NQD6UVGVNJUUSRVJE/graph.json","events_json":"https://pith.science/api/pith-number/WZIS3STU3NQD6UVGVNJUUSRVJE/events.json","paper":"https://pith.science/paper/WZIS3STU"},"agent_actions":{"view_html":"https://pith.science/pith/WZIS3STU3NQD6UVGVNJUUSRVJE","download_json":"https://pith.science/pith/WZIS3STU3NQD6UVGVNJUUSRVJE.json","view_paper":"https://pith.science/paper/WZIS3STU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.02860&json=true","fetch_graph":"https://pith.science/api/pith-number/WZIS3STU3NQD6UVGVNJUUSRVJE/graph.json","fetch_events":"https://pith.science/api/pith-number/WZIS3STU3NQD6UVGVNJUUSRVJE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WZIS3STU3NQD6UVGVNJUUSRVJE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WZIS3STU3NQD6UVGVNJUUSRVJE/action/storage_attestation","attest_author":"https://pith.science/pith/WZIS3STU3NQD6UVGVNJUUSRVJE/action/author_attestation","sign_citation":"https://pith.science/pith/WZIS3STU3NQD6UVGVNJUUSRVJE/action/citation_signature","submit_replication":"https://pith.science/pith/WZIS3STU3NQD6UVGVNJUUSRVJE/action/replication_record"}},"created_at":"2026-07-05T11:31:35.549877+00:00","updated_at":"2026-07-05T11:31:35.549877+00:00"}