Papers › ToViLaG: Your Visual-Language Generative Model is Also An Evildoer

ToViLaG: Your Visual-Language Generative Model is Also An Evildoer

13 Dec 2023arXiv:2312.11523archive 2025-07-28

Xinpeng Wang, Xiaoyuan Yi, Han Jiang, Shanlin Zhou, Zhihua Wei, Xing Xie

Warning: this paper includes model outputs showing offensive content. Recent large-scale Visual-Language Generative Models (VLGMs) have achieved unprecedented improvement in multimodal image/text generation. However, these models might also generate toxic content, e.g., offensive text and pornography images, raising significant ethical risks. Despite exhaustive studies on toxic degeneration of language models, this problem remains largely unexplored within the context of visual-language generation. This work delves into the propensity for toxicity generation and susceptibility to toxic data across various VLGMs. For this purpose, we built ToViLaG, a dataset comprising 32K co-toxic/mono-toxic text-image pairs and 1K innocuous but evocative text that tends to stimulate toxicity. Furthermore, we propose WInToRe, a novel toxicity metric tailored to visual-language generation, which theoretically reflects different aspects of toxicity considering both input and output. On such a basis, we benchmarked the toxicity of a diverse spectrum of VLGMs and discovered that some models do more evil than expected while some are more vulnerable to infection, underscoring the necessity of VLGMs detoxification. Therefore, we develop an innovative bottleneck-based detoxification method. Our method could reduce toxicity while maintaining comparable generation quality, providing a promising initial solution to this line of research.

PaperPDFCodeCode Syntology ran

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 18 code samples harvested from 2 repositories linked to this paper: 15 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 1 violated, 2 with no contract checked; 12 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 8 where our contract draft was wrong; 4 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 7 samples from 1 repository, 6 ran (1 with no instrument failure; 5 where Syntology's instrument failed); found in paper text by Syntology: 11 samples from 1 repository, 9 ran (2 with no instrument failure; 7 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. 1 of the 15 samples that ran has no recorded date.

victorup/ToViLaG officialmentioned on GitHubpytorchSyntology: 7 samples, 6 ran (5 instrument-failed) · 1 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

18 samples harvested: 15 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 1 violated, 2 with no contract checked; 12 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · violated contract
8ran · our draft was wrong
4ran; the contract check could not run
2ran
3unverified

Licence: 7 of the 18 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 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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”. “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.

victorup/ToViLaG official repository 7 samples, 6 ran (5 instrument-failed) · 1 unverified
autocontrast_func in 17 papers method/BLIP/transform/randaugment.py ran; the contract check could not run (RAISES) no licence file found · pointer only · 78ca1debd9c4ed87 · report fixture: signature img:ndarray/3/int, cutoff:int/0/int · returned ndarray · level recorded 2026-09-22
compute_acc in 61 papers
census: 3 implementations from 5 papers, 3 distinct outputs across 3 buckets, one shared input per bucket
method/BLIP/utils.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted no licence file found · pointer only · aa81d03f07735ab8 · report fixture: signature logits:Tensor/2/float, label:Tensor/1/int, reduction:str/0/str · returned float · level recorded 2026-09-22
compute_n_params in 30 papers method/BLIP/utils.py ran · our draft was wrong no licence file found · pointer only · ad12a494674d23fb · report fixture: signature model:callable/0/?, return_str:bool/0/bool · returned str · level recorded 2026-09-22
equalize_func in 17 papers method/BLIP/transform/randaugment.py ran; the contract check could not run (RAISES) no licence file found · pointer only · 4c53169f63a1fcef · report fixture: signature img:ndarray/3/int · returned ndarray · level recorded 2026-09-22
identity_func in 19 papers method/BLIP/transform/randaugment.py ran · violated contract fingerprinted no licence file found · pointer only · 1dd2ab205689d45b · report fixture: signature img:Tensor/4/float · returned Tensor · level recorded 2026-09-22
interpolate_pos_embed in 59 papers method/BLIP/models/vit.py ran · our draft was wrong no licence file found · pointer only · c6ec173f19f5c34d · report fixture: signature pos_embed_checkpoint:Tensor/3/float, visual_encoder:callable/0/? · returned Tensor · level recorded, date not recorded
detect in 8 papers metrics/toxicity/inference_imgclf.py unverified no licence file found · pointer only · daa6e468bb6841f2 · report
borisdayma/dalle-mini found in paper text by Syntology 11 samples, 9 ran (7 instrument-failed) · 2 unverified
blank_caption_function src/dalle_mini/data.py ran Apache-2.0 (permissive) · b46bc8a69ab97b12 · report fixture: signature example:dict/0/?, text_column:str/0/str, blank_caption_prob:float/0/float, rng:RandomState/0/? · returned dict · level recorded 2026-09-24
deepnet_init src/dalle_mini/model/modeling.py ran · our draft was wrong Apache-2.0 (permissive) · e254827ef156f929 · report fixture: signature init_std:float/0/float, gain:float/0/float · returned function · level recorded 2026-09-24
fix_html src/dalle_mini/model/text.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 848d81c889bb92eb · report fixture: signature t:str/0/str · returned str · level recorded 2026-09-24
norm in 120 papers
census: 23 implementations from 26 papers, 15 distinct outputs across 6 buckets, one shared input per bucket
src/dalle_mini/model/modeling.py ran · our draft was wrong Apache-2.0 (permissive) · 405ba2d3a0e6b5b3 · report fixture: signature type:str/0/str · returned RMSNorm · level recorded 2026-09-24
normalize_function src/dalle_mini/data.py ran Apache-2.0 (permissive) · a2d6ee58be3c306d · report fixture: signature example:dict/0/?, text_column:str/0/str, text_normalizer:callable/0/? · returned dict · level recorded 2026-09-24
replace_person_token src/dalle_mini/model/text.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 745c9aa603dc8439 · report fixture: signature t:str/0/str · returned str · level recorded 2026-09-24
replace_punctuation_with_commas src/dalle_mini/model/text.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 8af7dc10bc707987 · report fixture: signature t:str/0/str · returned str · level recorded 2026-09-24
shift_tokens_right in 49 papers
census: 9 implementations from 11 papers, 3 distinct outputs on one shared input
src/dalle_mini/data.py ran; the contract check could not run (RAISES) fingerprinted Apache-2.0 (permissive) · 45e195f64c082f0e · report fixture: signature input_ids:ndarray/2/int, decoder_start_token_id:int/0/int · returned ndarray · level recorded 2026-09-24
smelu src/dalle_mini/model/modeling.py ran · our draft was wrong Apache-2.0 (permissive) · 9914ba32442bd58d · report fixture: signature beta:float/0/float · returned custom_jvp · level recorded 2026-09-24
get_images_from_backend app/gradio/backend.py unverified Apache-2.0 (permissive) · 6189300424354113 · report
get_model_version app/gradio/backend.py unverified Apache-2.0 (permissive) · 7d648505413976ce · report

Tasks

Text Generation

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