{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/long-tail-learning-with-attributes","title":"From Generalized zero-shot learning to long-tail with class descriptors","arxiv_id":"2004.02235","date":"2020-04-05","proceeding":null,"authors":["Dvir Samuel","Yuval Atzmon","Gal Chechik"],"abstract":"Real-world data is predominantly unbalanced and long-tailed, but deep models struggle to recognize rare classes in the presence of frequent classes. Often, classes can be accompanied by side information like textual descriptions, but it is not fully clear how to use them for learning with unbalanced long-tail data. Such descriptions have been mostly used in (Generalized) Zero-shot learning (ZSL), suggesting that ZSL with class descriptions may also be useful for long-tail distributions. We describe DRAGON, a late-fusion architecture for long-tail learning with class descriptors. It learns to (1) correct the bias towards head classes on a sample-by-sample basis; and (2) fuse information from class-descriptions to improve the tail-class accuracy. We also introduce new benchmarks CUB-LT, SUN-LT, AWA-LT for long-tail learning with class-descriptions, building on existing learning-with-attributes datasets and a version of Imagenet-LT with class descriptors. DRAGON outperforms state-of-the-art models on the new benchmark. It is also a new SoTA on existing benchmarks for GFSL with class descriptors (GFSL-d) and standard (vision-only) long-tailed learning ImageNet-LT, CIFAR-10, 100, and Places365.","url_abs":"https://arxiv.org/abs/2004.02235v4","url_pdf":"https://arxiv.org/pdf/2004.02235v4.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":"long-tail-learning-with-attributes","repo_url":"https://github.com/dvirsamuel/DRAGON","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"few-shot-learning","task_name":"Few-Shot Learning"},{"task_slug":"generalized-few-shot-learning","task_name":"Generalized Few-Shot Learning"},{"task_slug":"generalized-zero-shot-learning","task_name":"Generalized Zero-Shot Learning"},{"task_slug":"long-tail-learning","task_name":"Long-tail Learning"},{"task_slug":"long-tail-learning-with-class-descriptors","task_name":"Long-tail learning with class descriptors"},{"task_slug":"zero-shot-learning","task_name":"Zero-Shot Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/generalized-few-shot-learning-on-awa2","task":"Generalized Few-Shot Learning","dataset":"AwA2","model":"DRAGON","rank_in_archive_order":4,"of":6,"metrics":{"Per-Class Accuracy (1-shot)":"67.1","Per-Class Accuracy (10-shots)":"81.9","Per-Class Accuracy (2-shots)":"69.1","Per-Class Accuracy (20-shots)":"83.3","Per-Class Accuracy (5-shots)":"76.7"},"uses_additional_data":false},{"leaderboard":"/sota/generalized-few-shot-learning-on-sun","task":"Generalized Few-Shot Learning","dataset":"SUN","model":"DRAGON","rank_in_archive_order":1,"of":5,"metrics":{"Per-Class Accuracy (1-shot)":"41.0","Per-Class Accuracy (10-shots)":"48.2","Per-Class Accuracy (2-shots)":"43.8","Per-Class Accuracy (5-shots)":"46.7"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-cifar-10-lt-r-10","task":"Long-tail Learning","dataset":"CIFAR-10-LT (ρ=10)","model":"smDRAGON","rank_in_archive_order":38,"of":50,"metrics":{"Error Rate":"11.84"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-cifar-10-lt-r-100","task":"Long-tail Learning","dataset":"CIFAR-10-LT (ρ=100)","model":"smDRAGON","rank_in_archive_order":22,"of":28,"metrics":{"Error Rate":"20.37"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-cifar-100-lt-r-10","task":"Long-tail Learning","dataset":"CIFAR-100-LT (ρ=10)","model":"smDRAGON","rank_in_archive_order":28,"of":31,"metrics":{"Error Rate":"41.23"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-cifar-100-lt-r-100","task":"Long-tail Learning","dataset":"CIFAR-100-LT (ρ=100)","model":"smDRAGON","rank_in_archive_order":56,"of":66,"metrics":{"Error Rate":"56.50"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-imagenet-lt","task":"Long-tail Learning","dataset":"ImageNet-LT","model":"smDRAGON","rank_in_archive_order":60,"of":69,"metrics":{"Top-1 Accuracy":"42.0"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-places-lt","task":"Long-tail Learning","dataset":"Places-LT","model":"smDRAGON","rank_in_archive_order":24,"of":29,"metrics":{"Top-1 Accuracy":"38.1"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-with-class-descriptors-on-2","task":"Long-tail learning with class descriptors","dataset":"AWA-LT","model":"DRAGON + Bal'Loss","rank_in_archive_order":1,"of":5,"metrics":{"Long-Tailed Accuracy":"92.2","Per-Class Accuracy":"76.2"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-with-class-descriptors-on-2","task":"Long-tail learning with class descriptors","dataset":"AWA-LT","model":"DRAGON","rank_in_archive_order":2,"of":5,"metrics":{"Long-Tailed Accuracy":"94.1","Per-Class Accuracy":"74.1"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-with-class-descriptors-on","task":"Long-tail learning with class descriptors","dataset":"CUB-LT","model":"DRAGON + Bal'Loss","rank_in_archive_order":1,"of":5,"metrics":{"Long-Tailed Accuracy":"66.5","Per-Class Accuracy":"60.1"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-with-class-descriptors-on","task":"Long-tail learning with class descriptors","dataset":"CUB-LT","model":"DRAGON","rank_in_archive_order":2,"of":5,"metrics":{"Long-Tailed Accuracy":"67.7","Per-Class Accuracy":"57.8"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-with-class-descriptors-on-3","task":"Long-tail learning with class descriptors","dataset":"ImageNet-LT-d","model":"DRAGON + Bal'Loss","rank_in_archive_order":1,"of":5,"metrics":{"Per-Class Accuracy":"53.5"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-with-class-descriptors-on-3","task":"Long-tail learning with class descriptors","dataset":"ImageNet-LT-d","model":"DRAGON","rank_in_archive_order":2,"of":5,"metrics":{"Per-Class Accuracy":"51.2"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-with-class-descriptors-on-1","task":"Long-tail learning with class descriptors","dataset":"SUN-LT","model":"DRAGON + Bal'Loss","rank_in_archive_order":1,"of":5,"metrics":{"Long-Tailed Accuracy":"38.5","Per-Class Accuracy":"36.1"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-with-class-descriptors-on-1","task":"Long-tail learning with class descriptors","dataset":"SUN-LT","model":"DRAGON","rank_in_archive_order":2,"of":5,"metrics":{"Long-Tailed Accuracy":"40.4","Per-Class Accuracy":"34.8"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers","held":null,"reason":"arxiv_id_not_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"by_repo":[{"repo":"dvirsamuel/DRAGON","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"30f5d01a21d0bc405daa4e9b3750507fbb6e3d5827ce0ff8bf8dc781c1647cda","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}