{"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/a-spitting-image-modular-superpixel","title":"A Spitting Image: Modular Superpixel Tokenization in Vision Transformers","arxiv_id":"2408.07680","date":"2024-08-14","proceeding":null,"authors":["Marius Aasan","Odd Kolbjørnsen","Anne Schistad Solberg","Adín Ramirez Rivera"],"abstract":"Vision Transformer (ViT) architectures traditionally employ a grid-based approach to tokenization independent of the semantic content of an image. We propose a modular superpixel tokenization strategy which decouples tokenization and feature extraction; a shift from contemporary approaches where these are treated as an undifferentiated whole. Using on-line content-aware tokenization and scale- and shape-invariant positional embeddings, we perform experiments and ablations that contrast our approach with patch-based tokenization and randomized partitions as baselines. We show that our method significantly improves the faithfulness of attributions, gives pixel-level granularity on zero-shot unsupervised dense prediction tasks, while maintaining predictive performance in classification tasks. Our approach provides a modular tokenization framework commensurable with standard architectures, extending the space of ViTs to a larger class of semantically-rich models.","url_abs":"https://arxiv.org/abs/2408.07680v2","url_pdf":"https://arxiv.org/pdf/2408.07680v2.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":"a-spitting-image-modular-superpixel","repo_url":"https://github.com/dsb-ifi/spit","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2408.07680","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2408.07680"}},"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":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/dsb-ifi/spit","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":16,"unverified":2},"by_repo_kind":{"official":{"samples":18,"ran":16,"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":0,"samples":[{"code_sha256_prefix":"50f7b854dc076644","entry":"argmax_cosine_pytorch","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/utils/cossimargmax.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/utils/cossimargmax.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":"50f7b854dc076644"}},{"code_sha256_prefix":"0aa7d593304219d8","entry":"asinh_contrast","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/tokenizer/proc.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/tokenizer/proc.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0aa7d593304219d8"}},{"code_sha256_prefix":"e8f0407093a971a2","entry":"cc_cpu","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/utils/concom.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/utils/concom.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":"e8f0407093a971a2"}},{"code_sha256_prefix":"52da940c3007a4cf","entry":"chunked_voronoi","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/tokenizer/voronoi.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/tokenizer/voronoi.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":"52da940c3007a4cf"}},{"code_sha256_prefix":"83dbe085029e03ec","entry":"fast_uidx_1d","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/utils/indexing.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/utils/indexing.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":"83dbe085029e03ec"}},{"code_sha256_prefix":"1e90c46bf4eea599","entry":"fast_uidx_long2d","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/utils/indexing.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/utils/indexing.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":"1e90c46bf4eea599"}},{"code_sha256_prefix":"9ac935a92bba0335","entry":"getmem","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/tokenizer/tokenizer.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/tokenizer/tokenizer.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":"9ac935a92bba0335"}},{"code_sha256_prefix":"97cb99b5105ed904","entry":"init_segedges","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/tokenizer/tokenizer.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/tokenizer/tokenizer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"97cb99b5105ed904"}},{"code_sha256_prefix":"068991f0659ecb8d","entry":"kuma_contrast1","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/tokenizer/proc.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/tokenizer/proc.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"068991f0659ecb8d"}},{"code_sha256_prefix":"64dd70cd2c44acf4","entry":"kuma_contrast1_","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/tokenizer/proc.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/tokenizer/proc.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"64dd70cd2c44acf4"}},{"code_sha256_prefix":"76a753ccff598f7b","entry":"packed_scatter_argmax","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/utils/cossimargmax.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/utils/cossimargmax.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":"76a753ccff598f7b"}},{"code_sha256_prefix":"9165c5ad579d7e88","entry":"scatter_add_1d","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/utils/scatter.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/utils/scatter.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":"9165c5ad579d7e88"}},{"code_sha256_prefix":"8cf0344d23aba65d","entry":"scatter_mean_2d","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/utils/scatter.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/utils/scatter.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":"8cf0344d23aba65d"}},{"code_sha256_prefix":"f8658b85d5a90abb","entry":"scatter_range_2d","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/utils/scatter.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/utils/scatter.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":"f8658b85d5a90abb"}},{"code_sha256_prefix":"6b43e964eb98fb3e","entry":"unravel_index","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/utils/indexing.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/utils/indexing.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":"6b43e964eb98fb3e"}},{"code_sha256_prefix":"f36f819b58509a1e","entry":"voronoi","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/tokenizer/voronoi.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/tokenizer/voronoi.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":"f36f819b58509a1e"}},{"code_sha256_prefix":"228314a8add82334","entry":"fast_pseudo_voronoi","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/tokenizer/voronoi.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/tokenizer/voronoi.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":"228314a8add82334"}},{"code_sha256_prefix":"771897acedfab466","entry":"img_coords","repo":"dsb-ifi/spit","repo_kind":"official","path":"spit/tokenizer/tokenizer.py","file_url":"https://github.com/dsb-ifi/spit/blob/HEAD/spit/tokenizer/tokenizer.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":"771897acedfab466"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}