{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2V6OW2Z6KYYWLJIBLIQQIOPXSS","short_pith_number":"pith:2V6OW2Z6","schema_version":"1.0","canonical_sha256":"d57ceb6b3e563165a5015a210439f794a3a158d592184bbde68b80b6dcdc3e4d","source":{"kind":"arxiv","id":"2506.08257","version":1},"attestation_state":"computed","paper":{"title":"Highly Compressed Tokenizer Can Generate Without Training","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"K. He, L. Lao Beyer, S. Karaman, T. Li, X. Chen","submitted_at":"2025-06-09T21:45:03Z","abstract_excerpt":"Commonly used image tokenizers produce a 2D grid of spatially arranged tokens. In contrast, so-called 1D image tokenizers represent images as highly compressed one-dimensional sequences of as few as 32 discrete tokens. We find that the high degree of compression achieved by a 1D tokenizer with vector quantization enables image editing and generative capabilities through heuristic manipulation of tokens, demonstrating that even very crude manipulations -- such as copying and replacing tokens between latent representations of images -- enable fine-grained image editing by transferring appearance"},"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":"2506.08257","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-09T21:45:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"72aaef375ec04a73cbef446175afea53fe668f34d4b1263128df9bf04805e124","abstract_canon_sha256":"e395e5e4ea58074b43b9d1a0c8758a6e3c99f10725467c3d5874dbc7851d804f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:47.079945Z","signature_b64":"oPN04yiZhkPRlaFrz7vFv+TsoBz1EoKfBQQuUalYuVrEwg9GDL5r8x++S42eNSG3rm4sPmrnLvtODTwCc/yEDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d57ceb6b3e563165a5015a210439f794a3a158d592184bbde68b80b6dcdc3e4d","last_reissued_at":"2026-07-05T11:18:47.079465Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:47.079465Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Highly Compressed Tokenizer Can Generate Without Training","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"K. He, L. Lao Beyer, S. Karaman, T. Li, X. Chen","submitted_at":"2025-06-09T21:45:03Z","abstract_excerpt":"Commonly used image tokenizers produce a 2D grid of spatially arranged tokens. In contrast, so-called 1D image tokenizers represent images as highly compressed one-dimensional sequences of as few as 32 discrete tokens. We find that the high degree of compression achieved by a 1D tokenizer with vector quantization enables image editing and generative capabilities through heuristic manipulation of tokens, demonstrating that even very crude manipulations -- such as copying and replacing tokens between latent representations of images -- enable fine-grained image editing by transferring appearance"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.08257","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/2506.08257/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":"2506.08257","created_at":"2026-07-05T11:18:47.079524+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.08257v1","created_at":"2026-07-05T11:18:47.079524+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.08257","created_at":"2026-07-05T11:18:47.079524+00:00"},{"alias_kind":"pith_short_12","alias_value":"2V6OW2Z6KYYW","created_at":"2026-07-05T11:18:47.079524+00:00"},{"alias_kind":"pith_short_16","alias_value":"2V6OW2Z6KYYWLJIB","created_at":"2026-07-05T11:18:47.079524+00:00"},{"alias_kind":"pith_short_8","alias_value":"2V6OW2Z6","created_at":"2026-07-05T11:18:47.079524+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17590","citing_title":"TivTok: Broadcasting Time-Invariant Tokens for Scalable Video Tokenization","ref_index":95,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05552","citing_title":"Balancing Image Compression and Generation with Bootstrapped Tokenization","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03578","citing_title":"Diffusing in the Right Space: A Systematic Study of Latent Diffusability","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08810","citing_title":"Continuous Language Diffusion as a Decoder-Interface Problem","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2V6OW2Z6KYYWLJIBLIQQIOPXSS","json":"https://pith.science/pith/2V6OW2Z6KYYWLJIBLIQQIOPXSS.json","graph_json":"https://pith.science/api/pith-number/2V6OW2Z6KYYWLJIBLIQQIOPXSS/graph.json","events_json":"https://pith.science/api/pith-number/2V6OW2Z6KYYWLJIBLIQQIOPXSS/events.json","paper":"https://pith.science/paper/2V6OW2Z6"},"agent_actions":{"view_html":"https://pith.science/pith/2V6OW2Z6KYYWLJIBLIQQIOPXSS","download_json":"https://pith.science/pith/2V6OW2Z6KYYWLJIBLIQQIOPXSS.json","view_paper":"https://pith.science/paper/2V6OW2Z6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.08257&json=true","fetch_graph":"https://pith.science/api/pith-number/2V6OW2Z6KYYWLJIBLIQQIOPXSS/graph.json","fetch_events":"https://pith.science/api/pith-number/2V6OW2Z6KYYWLJIBLIQQIOPXSS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2V6OW2Z6KYYWLJIBLIQQIOPXSS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2V6OW2Z6KYYWLJIBLIQQIOPXSS/action/storage_attestation","attest_author":"https://pith.science/pith/2V6OW2Z6KYYWLJIBLIQQIOPXSS/action/author_attestation","sign_citation":"https://pith.science/pith/2V6OW2Z6KYYWLJIBLIQQIOPXSS/action/citation_signature","submit_replication":"https://pith.science/pith/2V6OW2Z6KYYWLJIBLIQQIOPXSS/action/replication_record"}},"created_at":"2026-07-05T11:18:47.079524+00:00","updated_at":"2026-07-05T11:18:47.079524+00:00"}