{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IRZZMG2B4U6RZURQJ5LHP2XTFT","short_pith_number":"pith:IRZZMG2B","schema_version":"1.0","canonical_sha256":"4473961b41e53d1cd2304f5677eaf32ccc3992d09991c6ae81c5704a1b4d4696","source":{"kind":"arxiv","id":"2508.07747","version":1},"attestation_state":"computed","paper":{"title":"Grouped Speculative Decoding for Autoregressive Image Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Eunhyeok Park, Hyunho Kook, Juncheol Shin, Junhyuk So","submitted_at":"2025-08-11T08:27:57Z","abstract_excerpt":"Recently, autoregressive (AR) image models have demonstrated remarkable generative capabilities, positioning themselves as a compelling alternative to diffusion models. However, their sequential nature leads to long inference times, limiting their practical scalability. In this work, we introduce Grouped Speculative Decoding (GSD), a novel, training-free acceleration method for AR image models. While recent studies have explored Speculative Decoding (SD) as a means to speed up AR image generation, existing approaches either provide only modest acceleration or require additional training. Our i"},"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":"2508.07747","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-11T08:27:57Z","cross_cats_sorted":[],"title_canon_sha256":"cdc10979f72d9b38d9c8c30d01c7c069ba1a6da3b508cc15a132a7428cd78a0d","abstract_canon_sha256":"630db3788f3df681d251ca800a0348792ca9721f29c3cfca37a426397a43f0ca"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:51:51.467337Z","signature_b64":"sUcvSW30nWiUA/JwubbQrqj/oge8bjlpXwiJXAm0OGp2vpPRR1Abro0dNAAAm/tb2CIOR/Err2GIDX5QgNOBBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4473961b41e53d1cd2304f5677eaf32ccc3992d09991c6ae81c5704a1b4d4696","last_reissued_at":"2026-07-05T11:51:51.466912Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:51:51.466912Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Grouped Speculative Decoding for Autoregressive Image Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Eunhyeok Park, Hyunho Kook, Juncheol Shin, Junhyuk So","submitted_at":"2025-08-11T08:27:57Z","abstract_excerpt":"Recently, autoregressive (AR) image models have demonstrated remarkable generative capabilities, positioning themselves as a compelling alternative to diffusion models. However, their sequential nature leads to long inference times, limiting their practical scalability. In this work, we introduce Grouped Speculative Decoding (GSD), a novel, training-free acceleration method for AR image models. While recent studies have explored Speculative Decoding (SD) as a means to speed up AR image generation, existing approaches either provide only modest acceleration or require additional training. Our i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.07747","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/2508.07747/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":"2508.07747","created_at":"2026-07-05T11:51:51.466967+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.07747v1","created_at":"2026-07-05T11:51:51.466967+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07747","created_at":"2026-07-05T11:51:51.466967+00:00"},{"alias_kind":"pith_short_12","alias_value":"IRZZMG2B4U6R","created_at":"2026-07-05T11:51:51.466967+00:00"},{"alias_kind":"pith_short_16","alias_value":"IRZZMG2B4U6RZURQ","created_at":"2026-07-05T11:51:51.466967+00:00"},{"alias_kind":"pith_short_8","alias_value":"IRZZMG2B","created_at":"2026-07-05T11:51:51.466967+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.24211","citing_title":"Speculative Coupled Decoding for Training-Free Lossless Acceleration of Autoregressive Visual Generation","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07230","citing_title":"CASCADE: Context-Aware Relaxation for Speculative Image Decoding","ref_index":38,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IRZZMG2B4U6RZURQJ5LHP2XTFT","json":"https://pith.science/pith/IRZZMG2B4U6RZURQJ5LHP2XTFT.json","graph_json":"https://pith.science/api/pith-number/IRZZMG2B4U6RZURQJ5LHP2XTFT/graph.json","events_json":"https://pith.science/api/pith-number/IRZZMG2B4U6RZURQJ5LHP2XTFT/events.json","paper":"https://pith.science/paper/IRZZMG2B"},"agent_actions":{"view_html":"https://pith.science/pith/IRZZMG2B4U6RZURQJ5LHP2XTFT","download_json":"https://pith.science/pith/IRZZMG2B4U6RZURQJ5LHP2XTFT.json","view_paper":"https://pith.science/paper/IRZZMG2B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.07747&json=true","fetch_graph":"https://pith.science/api/pith-number/IRZZMG2B4U6RZURQJ5LHP2XTFT/graph.json","fetch_events":"https://pith.science/api/pith-number/IRZZMG2B4U6RZURQJ5LHP2XTFT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IRZZMG2B4U6RZURQJ5LHP2XTFT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IRZZMG2B4U6RZURQJ5LHP2XTFT/action/storage_attestation","attest_author":"https://pith.science/pith/IRZZMG2B4U6RZURQJ5LHP2XTFT/action/author_attestation","sign_citation":"https://pith.science/pith/IRZZMG2B4U6RZURQJ5LHP2XTFT/action/citation_signature","submit_replication":"https://pith.science/pith/IRZZMG2B4U6RZURQJ5LHP2XTFT/action/replication_record"}},"created_at":"2026-07-05T11:51:51.466967+00:00","updated_at":"2026-07-05T11:51:51.466967+00:00"}