Papers › Shatter and Gather: Learning Referring Image Segmentation with Text Supervision

Shatter and Gather: Learning Referring Image Segmentation with Text Supervision

29 Aug 2023ICCV 2023 1arXiv:2308.15512archive 2025-07-28

Dongwon Kim, Namyup Kim, Cuiling Lan, Suha Kwak

Referring image segmentation, the task of segmenting any arbitrary entities described in free-form texts, opens up a variety of vision applications. However, manual labeling of training data for this task is prohibitively costly, leading to lack of labeled data for training. We address this issue by a weakly supervised learning approach using text descriptions of training images as the only source of supervision. To this end, we first present a new model that discovers semantic entities in input image and then combines such entities relevant to text query to predict the mask of the referent. We also present a new loss function that allows the model to be trained without any further supervision. Our method was evaluated on four public benchmarks for referring image segmentation, where it clearly outperformed the existing method for the same task and recent open-vocabulary segmentation models on all the benchmarks.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 29 code samples harvested from 1 repository linked to this paper: 19 ran (of which 12 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 5 where Syntology's instrument failed) · 10 unverified. Where Syntology's instrument failed: 4 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 29 samples from 1 repository, 19 ran (14 with no instrument failure; 5 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

kdwonn/SaG officialmentioned on GitHubpytorchSyntology: 29 samples, 19 ran (5 instrument-failed) · 10 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

29 samples harvested: 19 ran (of which 12 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 5 where Syntology's instrument failed) · 10 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

4ran · our draft was wrong
1ran; the contract check could not run
14ran
10unverified

Licence: 29 of the 29 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from kdwonn/SaG. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 9 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: 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. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

kdwonn/SaG official repository 29 samples, 19 ran (5 instrument-failed) · 10 unverified
Attention in 344 papers
census: 13 implementations from 12 papers, 10 distinct outputs across 4 buckets, one shared input per bucket
model/cross_modal_attention.py ran no licence file found · pointer only · 12ea3b3517a96270 · report fixture: returned Attention · level recorded 2026-09-22
EncoderTextBERT model/cross_modal_attention.py ran no licence file found · pointer only · 028855901fa28457 · report fixture: signature x:Tensor/2/int, lengths:Tensor/1/int · returned tuple · level recorded 2026-09-24
FeedForward in 198 papers model/cross_modal_attention.py ran no licence file found · pointer only · b0e334ed471fe829 · report fixture: returned FeedForward · level recorded 2026-09-22
GEGLU in 20 papers
census: 12 implementations from 12 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
model/cross_modal_attention.py ran · metamorphic tier: invariant fingerprinted no licence file found · pointer only · 22d4e66449e24cb5 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
MLP in 375 papers model/cross_modal_attention.py ran no licence file found · pointer only · 694dbf618fcc8bc9 · report fixture: returned MLP · level recorded 2026-09-22
PatchEmbedding in 25 papers model/cross_modal_attention.py ran fingerprinted no licence file found · pointer only · f0d8b6422efbc32b · report fixture: returned PatchEmbedding · level recorded 2026-09-22
PositionEmbeddingLearned1D model/cross_modal_attention.py ran fingerprinted no licence file found · pointer only · 4540da9676418320 · report fixture: returned PositionEmbeddingLearned1D · level recorded 2026-09-22
PositionEmbeddingLearned2D model/cross_modal_attention.py ran fingerprinted no licence file found · pointer only · 569b6ec13ba1bb7c · report fixture: returned PositionEmbeddingLearned2D · level recorded 2026-09-22
PositionEmbeddingSine1D model/cross_modal_attention.py ran fingerprinted no licence file found · pointer only · 49e02a36dbacc263 · report fixture: returned PositionEmbeddingSine1D · level recorded 2026-09-22
PositionEmbeddingSine2D model/cross_modal_attention.py ran fingerprinted no licence file found · pointer only · 3eee2d3bcb861ce4 · report fixture: returned PositionEmbeddingSine2D · level recorded 2026-09-22
PostNorm model/cross_modal_attention.py ran no licence file found · pointer only · 903f3718322122c1 · report fixture: returned PostNorm · level recorded 2026-09-22
PreNorm in 34 papers model/cross_modal_attention.py ran no licence file found · pointer only · 446aa6353ce96ed3 · report fixture: returned PreNorm · level recorded 2026-09-22
Queries model/cross_modal_attention.py ran no licence file found · pointer only · 3f805427fbcf3779 · report fixture: returned Queries · level recorded 2026-09-22
ThinFeedForward model/cross_modal_attention.py ran no licence file found · pointer only · 8bcf0086b949bf6d · report fixture: returned ThinFeedForward · level recorded 2026-09-22
build_position_encoding in 42 papers model/cross_modal_attention.py ran · our draft was wrong no licence file found · pointer only · 1508f01cf4f224d9 · report fixture: signature dim:int/0/int, pos_type:str/0/str, axis:int/0/int · returned PositionEmbeddingSine2D · level recorded 2026-09-22
gumbel_softmax in 82 papers
census: 19 implementations from 20 papers, 12 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 2, kind float, dtype float32
model/cross_modal_attention.py ran · our draft was wrong fingerprinted no licence file found · pointer only · c8519676aac0e734 · report fixture: signature logits:Tensor/2/float, dim:int/0/int, tau:float/0/float · returned Tensor · level recorded 2026-09-22
hard_softmax in 5 papers model/cross_modal_attention.py ran · our draft was wrong fingerprinted no licence file found · pointer only · f66bb5953d92d51a · report fixture: signature logits:Tensor/2/float, dim:int/0/int · returned Tensor · level recorded 2026-09-22
l2norm in 84 papers
census: 12 implementations from 20 papers, 3 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
model/cross_modal_attention.py ran · our draft was wrong fingerprinted no licence file found · pointer only · e4fe846ff705def5 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
variable_len_pooling model/cross_modal_attention.py ran; the contract check could not run (RAISES) no licence file found · pointer only · cb5e2837a22f40a5 · report fixture: returned Tensor · level recorded 2026-09-22
Aggregator in 13 papers model/cross_modal_attention.py unverified no licence file found · pointer only · c40d3f0b01f5ae8d · report
CrossModalAttention model/cross_modal_attention.py unverified no licence file found · pointer only · 2dabab02658f440e · report
EncoderImage in 9 papers model/cross_modal_attention.py unverified no licence file found · pointer only · 5be60a5fd6a2851d · report
ImageTextEncoders model/cross_modal_attention.py unverified no licence file found · pointer only · e5aacceaaae4495b · report
SetPredictionModule model/cross_modal_attention.py unverified no licence file found · pointer only · 96fccb2494df3e07 · report
TransformerLayer in 34 papers model/cross_modal_attention.py unverified no licence file found · pointer only · 096e6b33de7b678c · report
VisionTransformer in 53 papers model/cross_modal_attention.py unverified no licence file found · pointer only · 19f4e674b83797e4 · report
get_img_backbone model/cross_modal_attention.py unverified no licence file found · pointer only · 5410c76c8cff05de · report
init_weights in 109 papers model/cross_modal_attention.py unverified no licence file found · pointer only · 41a107753d1ebf36 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for Shatter and Gather (graph method: Shatter and Gather). 33 bricks; individually on a fixture: 23 ran (of which 12 constructed an object rather than computing a result; 16 with no instrument failure: 0 honoured, 2 violated, 14 with no contract checked; 7 where Syntology's instrument failed) · 10 unverified. The assembly has not been run end-to-end.

  1. CrossModalAttention entry · level 0 · unverified · 2dabab02658f440e
  2. ImageTextEncoders model · level 0 · unverified · e5aacceaaae4495b
  3. MLP helper · level 1 · ran · 694dbf618fcc8bc9
  4. TransformerLayer helper · level 0 · unverified · 096e6b33de7b678c
  5. l2norm helper · level 1 · ran · our draft was wrong · e4fe846ff705def5
  6. EncoderImage helper · level 0 · unverified · 5be60a5fd6a2851d
  7. EncoderTextBERT helper · level 1 · ran · 028855901fa28457
  8. Attention helper · level 1 · ran · 12ea3b3517a96270
  9. FeedForward helper · level 1 · ran · b0e334ed471fe829
  10. PostNorm helper · level 1 · ran · 903f3718322122c1
  11. PreNorm helper · level 1 · ran · 446aa6353ce96ed3
  12. ThinFeedForward helper · level 1 · ran · 8bcf0086b949bf6d
  13. SetPredictionModule helper · level 0 · unverified · 96fccb2494df3e07
  14. get_img_backbone helper · level 0 · unverified · 5410c76c8cff05de
  15. variable_len_pooling helper · level 1 · ran; the contract check could not run (RAISES) · cb5e2837a22f40a5
  16. default helper · level 1 · ran · violated contract · 60fff7c3c400d7ff
  17. exists helper · level 1 · ran · violated contract · aa5486a3650902d8
  18. gumbel_softmax helper · level 1 · ran · our draft was wrong · c8519676aac0e734
  19. hard_softmax helper · level 1 · ran · our draft was wrong · f66bb5953d92d51a
  20. GEGLU helper · level 1 · ran · 22d4e66449e24cb5
  21. Aggregator helper · level 0 · unverified · c40d3f0b01f5ae8d
  22. VisionTransformer helper · level 0 · unverified · 19f4e674b83797e4
  23. Queries helper · level 1 · ran · 3f805427fbcf3779
  24. build_position_encoding helper · level 1 · ran · our draft was wrong · 1508f01cf4f224d9
  25. cache_fn helper · level 1 · ran · our draft was wrong · 8458f9ba0361ff43
  26. Block helper · level 0 · unverified · da0442761fe33eed
  27. PatchEmbedding helper · level 1 · ran · f0d8b6422efbc32b
  28. init_weights helper · level 0 · unverified · 41a107753d1ebf36
  29. resize_pos_embed helper · level 1 · ran · our draft was wrong · 039235a9d747b605
  30. PositionEmbeddingLearned1D helper · level 1 · ran · 4540da9676418320
  31. PositionEmbeddingLearned2D helper · level 1 · ran · 569b6ec13ba1bb7c
  32. PositionEmbeddingSine1D helper · level 1 · ran · 49e02a36dbacc263
  33. PositionEmbeddingSine2D helper · level 1 · ran · 3eee2d3bcb861ce4

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2308.15512") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Image SegmentationSegmentationSemantic SegmentationWeakly-supervised Learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections