{"url":"/sota/generalized-referring-expression-segmentation","task":{"name":"Generalized Referring Expression Segmentation","url":"/task/generalized-referring-expression-segmentation","note":null},"dataset":{"name":"gRefCOCO","url":"/dataset/grefcoco"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"Generalized Referring Expression Segmentation (GRES), introduced by [Liu et al in CVPR 2023](https://henghuiding.github.io/GRES/), allows expressions indicating any number of target objects. GRES takes an image and a referring expression as input, and requires mask prediction of the target object(s).","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["gIoU","cIoU"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"gIoU":null,"cIoU":null}},"counts":{"rows":13,"rows_with_code":11,"rows_with_paper_page":13,"rows_dated":13,"rows_using_additional_data":4},"rows":[{"rank_in_archive_order":1,"model":"DeRIS-L","metrics":{"cIoU":"72.00","gIoU":"77.67"},"uses_additional_data":false,"paper_date":"2025-07-02","paper":"/paper/deris-decoupling-perception-and-cognition-for","paper_url":"https://arxiv.org/abs/2507.01738v1","paper_title":"DeRIS: Decoupling Perception and Cognition for Enhanced Referring Image Segmentation through Loopback Synergy","code":"https://github.com/Dmmm1997/DeRIS","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"GSVA-Llama2-13B","metrics":{"cIoU":"66.38","gIoU":"70.04"},"uses_additional_data":true,"paper_date":"2023-12-15","paper":"/paper/gsva-generalized-segmentation-via-multimodal","paper_url":"https://arxiv.org/abs/2312.10103v3","paper_title":"GSVA: Generalized Segmentation via Multimodal Large Language Models","code":"https://github.com/leaplabthu/gsva","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":8,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"MABP","metrics":{"cIoU":"65.69","gIoU":"68.79"},"uses_additional_data":false,"paper_date":"2024-05-24","paper":"/paper/bring-adaptive-binding-prototypes-to","paper_url":"https://arxiv.org/abs/2405.15169v2","paper_title":"Bring Adaptive Binding Prototypes to Generalized Referring Expression Segmentation","code":"https://github.com/buptlwz/mabp","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"HDC","metrics":{"cIoU":"65.42","gIoU":"68.28"},"uses_additional_data":false,"paper_date":"2024-05-24","paper":"/paper/hdc-hierarchical-semantic-decoding-with-1","paper_url":"https://arxiv.org/abs/2405.15658v2","paper_title":"CoHD: A Counting-Aware Hierarchical Decoding Framework for Generalized Referring Expression Segmentation","code":"https://github.com/robertluo1/cohd","n_code_links":2,"syntology":null},{"rank_in_archive_order":5,"model":"GSVA-Vicuna-13B-v1.1","metrics":{"cIoU":"64.05","gIoU":"68.01"},"uses_additional_data":true,"paper_date":"2023-12-15","paper":"/paper/gsva-generalized-segmentation-via-multimodal","paper_url":"https://arxiv.org/abs/2312.10103v3","paper_title":"GSVA: Generalized Segmentation via Multimodal Large Language Models","code":"https://github.com/leaplabthu/gsva","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":8,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"GROUNDHOG","metrics":{"gIoU":"66.70"},"uses_additional_data":true,"paper_date":"2024-02-26","paper":"/paper/groundhog-grounding-large-language-models-to","paper_url":"https://arxiv.org/abs/2402.16846v2","paper_title":"GROUNDHOG: Grounding Large Language Models to Holistic Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":7,"model":"GSVA-Vicuna-7B-v1.1","metrics":{"cIoU":"63.29","gIoU":"66.47"},"uses_additional_data":true,"paper_date":"2023-12-15","paper":"/paper/gsva-generalized-segmentation-via-multimodal","paper_url":"https://arxiv.org/abs/2312.10103v3","paper_title":"GSVA: Generalized Segmentation via Multimodal Large Language Models","code":"https://github.com/leaplabthu/gsva","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":8,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"ReLA","metrics":{"cIoU":"62.42","gIoU":"63.60"},"uses_additional_data":false,"paper_date":"2023-06-01","paper":"/paper/gres-generalized-referring-expression-1","paper_url":"https://arxiv.org/abs/2306.00968v1","paper_title":"GRES: Generalized Referring Expression Segmentation","code":"https://github.com/henghuiding/ReLA","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":3,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"LAVT","metrics":{"cIoU":"57.64","gIoU":"58.40"},"uses_additional_data":false,"paper_date":"2021-12-04","paper":"/paper/lavt-language-aware-vision-transformer-for","paper_url":"https://arxiv.org/abs/2112.02244v2","paper_title":"LAVT: Language-Aware Vision Transformer for Referring Image Segmentation","code":"https://github.com/yz93/lavt-ris","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"CRIS","metrics":{"cIoU":"55.34","gIoU":"56.27"},"uses_additional_data":false,"paper_date":"2021-11-30","paper":"/paper/cris-clip-driven-referring-image-segmentation","paper_url":"https://arxiv.org/abs/2111.15174v2","paper_title":"CRIS: CLIP-Driven Referring Image Segmentation","code":"https://github.com/DerrickWang005/CRIS.pytorch","n_code_links":1,"syntology":{"n_ran":7,"n_unverified":5,"n_samples":12,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"LTS","metrics":{"cIoU":"52.30","gIoU":"52.70"},"uses_additional_data":false,"paper_date":"2021-03-30","paper":"/paper/locate-then-segment-a-strong-pipeline-for","paper_url":"https://arxiv.org/abs/2103.16284v1","paper_title":"Locate then Segment: A Strong Pipeline for Referring Image Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":12,"model":"VLT","metrics":{"cIoU":"52.51","gIoU":"52.00"},"uses_additional_data":false,"paper_date":"2021-08-12","paper":"/paper/vision-language-transformer-and-query","paper_url":"https://arxiv.org/abs/2108.05565v1","paper_title":"Vision-Language Transformer and Query Generation for Referring Segmentation","code":"https://github.com/henghuiding/Vision-Language-Transformer","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"MattNet","metrics":{"cIoU":"47.51","gIoU":"48.24"},"uses_additional_data":false,"paper_date":"2018-01-24","paper":"/paper/mattnet-modular-attention-network-for","paper_url":"http://arxiv.org/abs/1801.08186v3","paper_title":"MAttNet: Modular Attention Network for Referring Expression Comprehension","code":"https://github.com/lichengunc/MAttNet","n_code_links":1,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,264 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":5,"rows_with_any_sample_ran":5,"distinct_papers_with_graph_line":3,"distinct_papers_with_any_sample_ran":3,"samples_over_distinct_papers":{"n_ran":11,"n_unverified":16,"n_samples":27,"n_pointer_only_licence":0,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":15,"n_unverified":32,"n_samples":47,"n_pointer_only_licence":0,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}