Papers › DiscoSG: Towards Discourse-Level Text Scene Graph Parsing through Iterative Graph Refinement

DiscoSG: Towards Discourse-Level Text Scene Graph Parsing through Iterative Graph Refinement

18 Jun 2025arXiv:2506.15583archive 2025-07-28

Shaoqing Lin, Chong Teng, Fei Li, Donghong Ji, Lizhen Qu, Zhuang Li

Vision-Language Models (VLMs) now generate discourse-level, multi-sentence visual descriptions, challenging text scene graph parsers originally designed for single-sentence caption-to-graph mapping. Current approaches typically merge sentence-level parsing outputs for discourse input, often missing phenomena like cross-sentence coreference, resulting in fragmented graphs and degraded downstream VLM task performance. To address this, we introduce a new task, Discourse-level text Scene Graph parsing (DiscoSG), supported by our dataset DiscoSG-DS, which comprises 400 expert-annotated and 8,430 synthesised multi-sentence caption-graph pairs for images. Each caption averages 9 sentences, and each graph contains at least 3 times more triples than those in existing datasets. While fine-tuning large PLMs (i.e., GPT-4) on DiscoSG-DS improves SPICE by approximately 48% over the best sentence-merging baseline, high inference cost and restrictive licensing hinder its open-source use, and smaller fine-tuned PLMs struggle with complex graphs. We propose DiscoSG-Refiner, which drafts a base graph using one small PLM, then employs a second PLM to iteratively propose graph edits, reducing full-graph generation overhead. Using two Flan-T5-Base models, DiscoSG-Refiner still improves SPICE by approximately 30% over the best baseline while achieving 86 times faster inference than GPT-4. It also consistently improves downstream VLM tasks like discourse-level caption evaluation and hallucination detection. Code and data are available at: https://github.com/ShaoqLin/DiscoSG

PaperPDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2506.15583")

Code

Syntology Ran 1 of 19 code samples harvested from 1 repository linked to this paper; 18 have no recorded run. Of those that ran: 1 ran · violated contract.

By repository: official repository: 19 samples from 1 repository, 1 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

shaoqlin/discosg officialmentioned in paperpytorch report
zhuang-li/factual mentioned on GitHubpytorch 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

19 samples harvested; 1 ran; 0 honoured the contract we drafted; 18 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · violated contract
18unverified

Licence: 19 of the 19 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 ShaoqLin/DiscoSG. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

are_tuples_match ShaoqLin/DiscoSG/capture/capture.py official repository ran · violated contract no licence file found · pointer only · a834eb2f904a8668 · report
cal_agreement ShaoqLin/DiscoSG/caparena_metrics_all_score.py official repository unverified no licence file found · pointer only · d6e2c6fe36767d2c · report
cal_model_level_agreement ShaoqLin/DiscoSG/caparena_metrics_all_score.py official repository unverified no licence file found · pointer only · e6e6f6dc760babc9 · report
clean_text_caparena ShaoqLin/DiscoSG/convert_to_winner_style.py official repository unverified no licence file found · pointer only · 65be5c4b4da73e42 · report
collect_unique_captions ShaoqLin/DiscoSG/src/factual_scene_graph/dataset_utils.py official repository unverified no licence file found · pointer only · e5b0cb48bb329644 · report
collect_unique_captions_flickr8k ShaoqLin/DiscoSG/src/factual_scene_graph/dataset_utils.py official repository unverified no licence file found · pointer only · 0a2e445f0debd3d8 · report
compute_correlation ShaoqLin/DiscoSG/src/factual_scene_graph/eval_utils.py official repository unverified no licence file found · pointer only · 8b28307b99f57703 · report
determine_winner ShaoqLin/DiscoSG/convert_to_winner_style.py official repository unverified no licence file found · pointer only · 66b7066c73cbe754 · report
extract_triples ShaoqLin/DiscoSG/src/factual_scene_graph/triple_utils.py official repository unverified no licence file found · pointer only · ca3043f1941bb849 · report
find_matching_metrics ShaoqLin/DiscoSG/convert_to_winner_style.py official repository unverified no licence file found · pointer only · 891f7322773a89aa · report
get_synonyms ShaoqLin/DiscoSG/capture/capture.py official repository unverified no licence file found · pointer only · 6962fa94d60714a9 · report
load_parsed_captions ShaoqLin/DiscoSG/src/factual_scene_graph/parse_utils.py official repository unverified no licence file found · pointer only · fbcbdb56cb865fc0 · report
merge_sentence_results ShaoqLin/DiscoSG/capture/capture.py official repository unverified no licence file found · pointer only · 0cc82cec90a528c2 · report
normalize_triple ShaoqLin/DiscoSG/src/factual_scene_graph/triple_utils.py official repository unverified no licence file found · pointer only · 5275fcf2194b983c · report
parse_captions ShaoqLin/DiscoSG/src/factual_scene_graph/parse_utils.py official repository unverified no licence file found · pointer only · f0be599ef5e5abbf · report
parse_captions_fix ShaoqLin/DiscoSG/src/factual_scene_graph/parse_utils.py official repository unverified no licence file found · pointer only · 4c4b9f1c9d3001d8 · report
prepare_data_for_elo ShaoqLin/DiscoSG/caparena_metrics_all_score.py official repository unverified no licence file found · pointer only · a1d8dfe0ba8f9a0c · report
remove_space ShaoqLin/DiscoSG/src/factual_scene_graph/triple_utils.py official repository unverified no licence file found · pointer only · f1cc0f7668994de6 · report
save_results_to_json ShaoqLin/DiscoSG/src/factual_scene_graph/log_utils.py official repository unverified no licence file found · pointer only · a5b0a0b31855c4e0 · report

Tasks

Graph GenerationHallucinationSentence

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsBASEBPEDense ConnectionsDropoutGPT-4Label SmoothingLayer NormalizationSoftmaxTransformer

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