{"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/no-train-all-gain-self-supervised-gradients","title":"No Train, all Gain: Self-Supervised Gradients Improve Deep Frozen Representations","arxiv_id":"2407.10964","date":"2024-07-15","proceeding":null,"authors":["Walter Simoncini","Spyros Gidaris","Andrei Bursuc","Yuki M. Asano"],"abstract":"This paper introduces FUNGI, Features from UNsupervised GradIents, a method to enhance the features of transformer encoders by leveraging self-supervised gradients. Our method is simple: given any pretrained model, we first compute gradients from various self-supervised objectives for each input. These gradients are projected to a lower dimension and then concatenated with the model's output embedding. The resulting features are evaluated on k-nearest neighbor classification over 11 datasets from vision, 5 from natural language processing, and 2 from audio. Across backbones spanning various sizes and pretraining strategies, FUNGI features provide consistent performance improvements over the embeddings. We also show that using FUNGI features can benefit linear classification, clustering and image retrieval, and that they significantly improve the retrieval-based in-context scene understanding abilities of pretrained models, for example improving upon DINO by +17% for semantic segmentation - without any training.","url_abs":"https://arxiv.org/abs/2407.10964v2","url_pdf":"https://arxiv.org/pdf/2407.10964v2.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":"no-train-all-gain-self-supervised-gradients","repo_url":"https://github.com/waltersimoncini/fungivision","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"all","task_name":"All"},{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"scene-understanding","task_name":"Scene Understanding"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"dino","method_name":"DINO"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"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":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"vision-transformer","method_name":"Vision Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2407.10964","atlas_url":"https://app.syntology.ai/?focus=2407.10964","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2407.10964"}},"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":"deterministic:regex_extraction","url":"https://github.com/WalterSimoncini/fungivision","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/waltersimoncini/fungivision","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":2,"ran":5,"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":0,"samples":[{"code_sha256_prefix":"104dacd44ae83925","entry":"_layer_type","repo":"waltersimoncini/fungivision","repo_kind":"official","path":"fungivision/gradients/base_extractor.py","file_url":"https://github.com/waltersimoncini/fungivision/blob/HEAD/fungivision/gradients/base_extractor.py","link_basis":"first_harvest_node","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":"104dacd44ae83925"}},{"code_sha256_prefix":"1f2b26d7aa212e61","entry":"generate_projection_matrix","repo":"WalterSimoncini/fungivision","repo_kind":"official","path":"fungivision/utils/compression.py","file_url":"https://github.com/WalterSimoncini/fungivision/blob/HEAD/fungivision/utils/compression.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":"1f2b26d7aa212e61"}},{"code_sha256_prefix":"01e52b8e916acca2","entry":"is_supported","repo":"WalterSimoncini/fungivision","repo_kind":"official","path":"fungivision/utils/autograd_hacks.py","file_url":"https://github.com/WalterSimoncini/fungivision/blob/HEAD/fungivision/utils/autograd_hacks.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":"01e52b8e916acca2"}},{"code_sha256_prefix":"3db9ee14869a6845","entry":"model_feature_dim","repo":"WalterSimoncini/fungivision","repo_kind":"official","path":"fungivision/utils/misc.py","file_url":"https://github.com/WalterSimoncini/fungivision/blob/HEAD/fungivision/utils/misc.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":"3db9ee14869a6845"}},{"code_sha256_prefix":"30d027ce5196ab7c","entry":"preprocess_layer_paths","repo":"WalterSimoncini/fungivision","repo_kind":"official","path":"fungivision/utils/autograd_hacks.py","file_url":"https://github.com/WalterSimoncini/fungivision/blob/HEAD/fungivision/utils/autograd_hacks.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":"30d027ce5196ab7c"}},{"code_sha256_prefix":"7adb31904977e81b","entry":"rgetattr","repo":"waltersimoncini/fungivision","repo_kind":"official","path":"fungivision/gradients/base_extractor.py","file_url":"https://github.com/waltersimoncini/fungivision/blob/HEAD/fungivision/gradients/base_extractor.py","link_basis":"first_harvest_node","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":"7adb31904977e81b"}},{"code_sha256_prefix":"ba88ce6fa6ae61d3","entry":"suggested_scaling_factor","repo":"WalterSimoncini/fungivision","repo_kind":"official","path":"fungivision/utils/compression.py","file_url":"https://github.com/WalterSimoncini/fungivision/blob/HEAD/fungivision/utils/compression.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":"ba88ce6fa6ae61d3"}},{"code_sha256_prefix":"8c517a7e744ed152","entry":"BaseGradientExtractor","repo":"waltersimoncini/fungivision","repo_kind":"official","path":"fungivision/gradients/base_extractor.py","file_url":"https://github.com/waltersimoncini/fungivision/blob/HEAD/fungivision/gradients/base_extractor.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":"8c517a7e744ed152"}},{"code_sha256_prefix":"30dc14b4b04ac056","entry":"clear_backprops","repo":"waltersimoncini/fungivision","repo_kind":"official","path":"fungivision/gradients/base_extractor.py","file_url":"https://github.com/waltersimoncini/fungivision/blob/HEAD/fungivision/gradients/base_extractor.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":"30dc14b4b04ac056"}},{"code_sha256_prefix":"546242bc466400c2","entry":"compute_grad1","repo":"waltersimoncini/fungivision","repo_kind":"official","path":"fungivision/gradients/base_extractor.py","file_url":"https://github.com/waltersimoncini/fungivision/blob/HEAD/fungivision/gradients/base_extractor.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":"546242bc466400c2"}},{"code_sha256_prefix":"4632ca8fbf0dd221","entry":"rsetattr","repo":"WalterSimoncini/fungivision","repo_kind":"official","path":"fungivision/utils/misc.py","file_url":"https://github.com/WalterSimoncini/fungivision/blob/HEAD/fungivision/utils/misc.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":"4632ca8fbf0dd221"}},{"code_sha256_prefix":"56afcb3496c8fd91","entry":"symsqrt","repo":"WalterSimoncini/fungivision","repo_kind":"official","path":"fungivision/utils/autograd_hacks.py","file_url":"https://github.com/WalterSimoncini/fungivision/blob/HEAD/fungivision/utils/autograd_hacks.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":"56afcb3496c8fd91"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}