{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/autoregressive-image-generation-without","title":"Autoregressive Image Generation without Vector Quantization","arxiv_id":"2406.11838","date":"2024-06-17","proceeding":null,"authors":["Tianhong Li","Yonglong Tian","He Li","Mingyang Deng","Kaiming He"],"abstract":"Conventional wisdom holds that autoregressive models for image generation are typically accompanied by vector-quantized tokens. We observe that while a discrete-valued space can facilitate representing a categorical distribution, it is not a necessity for autoregressive modeling. In this work, we propose to model the per-token probability distribution using a diffusion procedure, which allows us to apply autoregressive models in a continuous-valued space. Rather than using categorical cross-entropy loss, we define a Diffusion Loss function to model the per-token probability. This approach eliminates the need for discrete-valued tokenizers. We evaluate its effectiveness across a wide range of cases, including standard autoregressive models and generalized masked autoregressive (MAR) variants. By removing vector quantization, our image generator achieves strong results while enjoying the speed advantage of sequence modeling. We hope this work will motivate the use of autoregressive generation in other continuous-valued domains and applications. Code is available at: https://github.com/LTH14/mar.","url_abs":"https://arxiv.org/abs/2406.11838v3","url_pdf":"https://arxiv.org/pdf/2406.11838v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"autoregressive-image-generation-without","repo_url":"https://github.com/lth14/mar","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"autoregressive-image-generation-without","repo_url":"https://github.com/shaochenze/ear","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"quantization","task_name":"Quantization"}],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"},{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-generation-on-imagenet-256x256","task":"Image Generation","dataset":"ImageNet 256x256","model":"MAR-H, Diff Loss","rank_in_archive_order":24,"of":94,"metrics":{"FID":"1.55"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-256x256","task":"Image Generation","dataset":"ImageNet 256x256","model":"MAR-L, Diff Loss","rank_in_archive_order":39,"of":94,"metrics":{"FID":"1.78"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-256x256","task":"Image Generation","dataset":"ImageNet 256x256","model":"MAR-B, Diff Loss","rank_in_archive_order":56,"of":94,"metrics":{"FID":"2.31"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"MAR-L, Diff Loss","rank_in_archive_order":17,"of":52,"metrics":{"FID":"1.73"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2406.11838","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2406.11838"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"deterministic:regex_extraction","url":"https://github.com/LTH14/mar","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lth14/mar","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/shaochenze/ear","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":5,"ran_fixture":2,"ran_honours":1,"unverified":5},"by_repo_kind":{"official":{"samples":12,"ran":7,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"89c137ff7c13fc84","entry":"FinalLayer","repo":"lth14/mar","repo_kind":"official","path":"models/diffloss.py","file_url":"https://github.com/lth14/mar/blob/HEAD/models/diffloss.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"89c137ff7c13fc84"}},{"code_sha256_prefix":"de57ba56d49f5ccb","entry":"GaussianDiffusion","repo":"lth14/mar","repo_kind":"official","path":"models/diffloss.py","file_url":"https://github.com/lth14/mar/blob/HEAD/models/diffloss.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"de57ba56d49f5ccb"}},{"code_sha256_prefix":"944b8eccd301a676","entry":"ResBlock","repo":"lth14/mar","repo_kind":"official","path":"models/diffloss.py","file_url":"https://github.com/lth14/mar/blob/HEAD/models/diffloss.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"944b8eccd301a676"}},{"code_sha256_prefix":"11f5d9b171f4a6e3","entry":"SimpleMLPAdaLN","repo":"lth14/mar","repo_kind":"official","path":"models/diffloss.py","file_url":"https://github.com/lth14/mar/blob/HEAD/models/diffloss.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"11f5d9b171f4a6e3"}},{"code_sha256_prefix":"68c1d0608671c310","entry":"_WrappedModel","repo":"lth14/mar","repo_kind":"official","path":"models/diffloss.py","file_url":"https://github.com/lth14/mar/blob/HEAD/models/diffloss.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"68c1d0608671c310"}},{"code_sha256_prefix":"086e19314a5cc8b9","entry":"_extract_into_tensor","repo":"lth14/mar","repo_kind":"official","path":"models/diffloss.py","file_url":"https://github.com/lth14/mar/blob/HEAD/models/diffloss.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"086e19314a5cc8b9"}},{"code_sha256_prefix":"62fcb3912a967a50","entry":"modulate","repo":"lth14/mar","repo_kind":"official","path":"models/diffloss.py","file_url":"https://github.com/lth14/mar/blob/HEAD/models/diffloss.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"62fcb3912a967a50"}},{"code_sha256_prefix":"ea9dbc131adf582e","entry":"space_timesteps","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"ea9dbc131adf582e"}},{"code_sha256_prefix":"0b1b4cb568fcf57f","entry":"DiffLoss","repo":"lth14/mar","repo_kind":"official","path":"models/diffloss.py","file_url":"https://github.com/lth14/mar/blob/HEAD/models/diffloss.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0b1b4cb568fcf57f"}},{"code_sha256_prefix":"e58dd85eb8661db5","entry":"ModelVarType","repo":"lth14/mar","repo_kind":"official","path":"models/diffloss.py","file_url":"https://github.com/lth14/mar/blob/HEAD/models/diffloss.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e58dd85eb8661db5"}},{"code_sha256_prefix":"5f227e2400952f44","entry":"SpacedDiffusion","repo":"lth14/mar","repo_kind":"official","path":"models/diffloss.py","file_url":"https://github.com/lth14/mar/blob/HEAD/models/diffloss.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5f227e2400952f44"}},{"code_sha256_prefix":"2b4d845f9d4a841f","entry":"create_diffusion","repo":"lth14/mar","repo_kind":"official","path":"models/diffloss.py","file_url":"https://github.com/lth14/mar/blob/HEAD/models/diffloss.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2b4d845f9d4a841f"}},{"code_sha256_prefix":"d076098d89dd7262","entry":"mask_by_order","repo":"lth14/mar","repo_kind":"official","path":"models/mar.py","file_url":"https://github.com/lth14/mar/blob/HEAD/models/mar.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d076098d89dd7262"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}