{"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/principal-components-enable-a-new-language-of","title":"\"Principal Components\" Enable A New Language of Images","arxiv_id":"2503.08685","date":"2025-03-11","proceeding":null,"authors":["Xin Wen","Bingchen Zhao","Ismail Elezi","Jiankang Deng","Xiaojuan Qi"],"abstract":"We introduce a novel visual tokenization framework that embeds a provable PCA-like structure into the latent token space. While existing visual tokenizers primarily optimize for reconstruction fidelity, they often neglect the structural properties of the latent space -- a critical factor for both interpretability and downstream tasks. Our method generates a 1D causal token sequence for images, where each successive token contributes non-overlapping information with mathematically guaranteed decreasing explained variance, analogous to principal component analysis. This structural constraint ensures the tokenizer extracts the most salient visual features first, with each subsequent token adding diminishing yet complementary information. Additionally, we identified and resolved a semantic-spectrum coupling effect that causes the unwanted entanglement of high-level semantic content and low-level spectral details in the tokens by leveraging a diffusion decoder. Experiments demonstrate that our approach achieves state-of-the-art reconstruction performance and enables better interpretability to align with the human vision system. Moreover, auto-regressive models trained on our token sequences achieve performance comparable to current state-of-the-art methods while requiring fewer tokens for training and inference.","url_abs":"https://arxiv.org/abs/2503.08685v1","url_pdf":"https://arxiv.org/pdf/2503.08685v1.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":"principal-components-enable-a-new-language-of","repo_url":"https://github.com/visual-gen/semanticist","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"}],"methods":[{"method_slug":"align","method_name":"ALIGN"},{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2503.08685","atlas_url":"https://app.syntology.ai/?focus=2503.08685","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2503.08685"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/visual-gen/semanticist","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":2,"ran":7,"ran_honours":3,"unverified":5},"by_repo_kind":{"official":{"samples":17,"ran":12,"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":"b05a447cc3b9e991","entry":"build_mlp","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage1/diffuse_slot.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage1/diffuse_slot.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b05a447cc3b9e991"}},{"code_sha256_prefix":"d0aca56def6a1f55","entry":"concat_all_gather","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/engine/trainer_utils.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/engine/trainer_utils.py","link_basis":"harvester_set","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":"d0aca56def6a1f55"}},{"code_sha256_prefix":"c10d644e68f17884","entry":"decode_n_tokens","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage2/generate.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage2/generate.py","link_basis":"harvester_set","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":"c10d644e68f17884"}},{"code_sha256_prefix":"f6e7c115a93e25a8","entry":"decode_one_token","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage2/generate.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage2/generate.py","link_basis":"harvester_set","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":"f6e7c115a93e25a8"}},{"code_sha256_prefix":"f6ff7671338c9c92","entry":"find_multiple","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage2/gpt.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage2/gpt.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f6ff7671338c9c92"}},{"code_sha256_prefix":"e13eb3c1606755e6","entry":"get_1d_sincos_pos_embed","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage1/pos_embed.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage1/pos_embed.py","link_basis":"harvester_set","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":"e13eb3c1606755e6"}},{"code_sha256_prefix":"c92c27c924b517e8","entry":"get_2d_sincos_pos_embed","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage1/diffusion_transfomer.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage1/diffusion_transfomer.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c92c27c924b517e8"}},{"code_sha256_prefix":"1be6845a690a8075","entry":"get_obj_from_str","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/engine/trainer_utils.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/engine/trainer_utils.py","link_basis":"harvester_set","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":"1be6845a690a8075"}},{"code_sha256_prefix":"1f5b50d685e20502","entry":"interpolate_features","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage1/diffuse_slot.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage1/diffuse_slot.py","link_basis":"harvester_set","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":"1f5b50d685e20502"}},{"code_sha256_prefix":"03310bba324ae4fb","entry":"modulate","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage1/diffusion_transfomer.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage1/diffusion_transfomer.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"03310bba324ae4fb"}},{"code_sha256_prefix":"62fcb3912a967a50","entry":"modulate","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage2/diffloss.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage2/diffloss.py","link_basis":"harvester_set","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":"d90c14200bec99a5","entry":"prefill","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage2/generate.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage2/generate.py","link_basis":"harvester_set","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":"d90c14200bec99a5"}},{"code_sha256_prefix":"4134f673786a2abe","entry":"drop_path","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage1/vision_transformer.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage1/vision_transformer.py","link_basis":"harvester_set","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":"4134f673786a2abe"}},{"code_sha256_prefix":"9417ae492629cf06","entry":"get_2d_sincos_pos_embed","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage1/pos_embed.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage1/pos_embed.py","link_basis":"harvester_set","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":"9417ae492629cf06"}},{"code_sha256_prefix":"665d8a4e8f673a4c","entry":"get_2d_sincos_pos_embed_from_grid","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage1/diffusion_transfomer.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage1/diffusion_transfomer.py","link_basis":"harvester_set","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":"665d8a4e8f673a4c"}},{"code_sha256_prefix":"4fd80de79832745d","entry":"get_2d_sincos_pos_embed_from_grid","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/stage1/pos_embed.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/stage1/pos_embed.py","link_basis":"harvester_set","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":"4fd80de79832745d"}},{"code_sha256_prefix":"e6197485fe096b58","entry":"instantiate_from_config","repo":"visual-gen/semanticist","repo_kind":"official","path":"semanticist/engine/trainer_utils.py","file_url":"https://github.com/visual-gen/semanticist/blob/HEAD/semanticist/engine/trainer_utils.py","link_basis":"harvester_set","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":"e6197485fe096b58"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}