Browse › Computer Vision › Text-to-Image Generation › COCO (Common Objects in Context)

Text-to-Image Generation archive 2025-07-28

COCO (Common Objects in Context) Benchmark (Text-to-Image Generation)

69 rows 59 with code listed 8 metrics Dataset page

The development of the brain's blood supply in an embryo involves a complex process with several stages. Initially, there is a network of connections between the carotid and basilar artery systems that usually disappear during development. However, in some cases, these connections remain, leading to variations in the adult cerebral circulation. Around the 24th day of embryonic life (at the 3 mm embryonic stage), the internal carotid arteries (ICA) emerge. They are formed from a combination of the 3rd branchial arch arteries and the distal segments of the paired dorsal aortae. The ICA initially provides all the blood needed by the developing brain. As the brain grows, particularly the occipital region, brainstem, and cerebellum, the blood supply from the ICA becomes insufficient, prompting the development of the posterior circulation. At this stage (4-5 mm embryonic stage), the posterior circulation, specifically the hindbrain, is supplied by two longitudinal neural arteries. These arteries receive blood from several temporary connections (anastomoses) with the ICA, including: Trigeminal artery (TA), Otic artery (OA), Hypoglossal artery (HA), Proatlantal artery (ProA). These carotid-vertebrobasilar anastomoses act as temporary bridges, providing blood flow to the developing posterior circulation until the vertebral arteries (VA) and the basilar artery (BA) are fully formed.6

The BA forms between the 5-8 mm stage by the fusion of the longitudinal neural arteries. As the posterior communicating artery develops and connects with the distal BA, the TA, OA, and HA typically regress. Unlike the TA, OA, and HA, the ProA persists until the VA are fully developed. A segment of the ProA becomes incorporated into the V3 segment of the VA and the distal portions of the occipital artery. The VA develop between the 7-12 mm stage from transverse connections between cervical intersegmental arteries, starting with the ProA and progressing downwards to the 6th intersegmental artery. This 6th intersegmental artery eventually forms the origin of the adult VA from the subclavian artery. 7-8

Persistence of the HA into adulthood, the focus of this discussion, is a rare occurrence. It is the second most common persistent carotid-vertebrobasilar anastomosis, following the persistent trigeminal artery. The failure of the HA to regress during embryological development leads to PPHA. This condition can impact the normal blood flow dynamics in the brain and, in certain cases, can be associated with cerebrovascular issues like aneurysms or ischemic events.

The archive carries no text for this table; the description above is the archive's text for the task Text-to-Image Generation. archive 2025-07-28

Over time archive 2025-07-28

The chart needs JavaScript; the table below carries every value.

Direction inferred from the metric name, not from the archive: FID (lower is better), Inception score (higher is better), FID-1 (lower is better), FID-2 (lower is better), FID-4 (lower is better), FID-8 (lower is better), Zero shot FID (lower is better). Not inferred (points only, no best-so-far line): SOA-C. Points are placed at the row's paper date; 69 of 69 rows carry one.

Results archive 2025-07-28

Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.

Paper Code Ran Syntology Report
1 RAT-Diffusion 5.00 ✓ Paper Code 2024 linked, not harvested report
2 Re-Imagen (Finetuned) 5.25 – Paper – 2022 no code linked report
3 U-ViT-S/2-Deep 5.48 – Paper Code 2022 3 of 3 ran · 0 unverified report
4 GLIGEN (fine-tuned, Detection + Caption data) 5.61 – Paper Code 2023 2 of 2 ran · 0 unverified report
5 GLIGEN (fine-tuned, Detection data only) 5.82 – Paper Code 2023 2 of 2 ran · 0 unverified report
6 U-ViT-S/2 5.95 – Paper Code 2022 3 of 3 ran · 0 unverified report
7 ConPreDiff 6.216.21 – Paper – 2024 no code linked report
8 TLDM 6.29 – Paper Code 2022 3 of 3 ran · 0 unverified report
9 GLIGEN (fine-tuned, Grounding data) 6.38 – Paper Code 2023 2 of 2 ran · 0 unverified report
10 RAPHAEL (zero-shot) 6.61 – Paper Code 2023 linked, not harvested report
11 ERNIE-ViLG 2.0 (zero-shot) 6.75 – Paper Code 2022 linked, not harvested report
12 Re-Imagen 6.88 – Paper – 2022 no code linked report
13 eDiff-I (zero-shot) 6.95 – Paper Code 2022 4 of 13 ran · 9 unverified report
14 Swinv2-Imagen 7.2131.46 ✓ Paper – 2022 no code linked report
15 Imagen (zero-shot) 7.27 ✓ Paper Code 2022 5 of 11 ran · 6 unverified report
16 GigaGAN (Zero-shot, 64x64) 7.28 – Paper Code 2023 8 of 14 ran · 6 unverified report
17 StyleGAN-T (Zero-shot, 64x64) 7.3 – Paper Code 2023 linked, not harvested report
18 Make-a-Scene (unfiltered) 7.55 ✓ Paper Code 2022 4 of 6 ran · 2 unverified report
19 Kandinsky 8.03 – Paper Code 2023 12 of 18 ran · 6 unverified report
20 Lafite 8.1232.3461.09 – Paper Code 2021 4 of 18 ran · 14 unverified report
21 SiD-LSG (Data-free distillation, zero-shot FID) 8.158.15 – Paper Code 2024 18 of 24 ran · 6 unverified report
22 simple diffusion (U-ViT) 8.3 – Paper Code 2023 linked, not harvested report
23 GigaGAN (Zero-shot, 256x256) 9.09 – Paper Code 2023 8 of 14 ran · 6 unverified report
24 XMC-GAN (256 x 256) 9.330.5 – Paper Code 2021 3 of 3 ran · 0 unverified report
25 XMC-GAN 9.33 ✓ Paper Code 2021 4 of 4 ran · 0 unverified report
26 DALL-E 2 10.39 ✓ Paper Code 2022 29 of 38 ran · 9 unverified report
27 Corgi-Semi 10.6 – Paper Code 2022 7 of 17 ran · 10 unverified report
28 Corgi 10.88 – Paper Code 2022 7 of 17 ran · 10 unverified report
29 TR0N (StyleGAN-XL, LAION2BCLIP, BLIP-2, zero-shot) 10.9 – Paper Code 2023 9 of 19 ran · 10 unverified report
30 Make-a-Scene (unfiltered) 11.84 ✓ Paper Code 2022 4 of 6 ran · 2 unverified report
31 GLIDE (zero-shot) 12.24 ✓ Paper Code 2021 9 of 15 ran · 6 unverified report
32 KNN-Diffusion 12.5 – Paper – 2022 no code linked report
33 GALIP (CC12m) 12.54 – Paper Code 2023 7 of 11 ran · 4 unverified report
34 Latent Diffusion (LDM-KL-8-G) 12.63 ✓ Paper Code 2021 19 of 28 ran · 9 unverified report
35 Stable Diffusion 12.63 – Paper – 2022 no code linked report
36 NÜWA (256 x 256) 12.9 27.2 – Paper Code 2021 3 of 3 ran · 0 unverified report
37 VQ-Diffusion-F 13.86 ✓ Paper Code 2021 linked, not harvested report
38 StyleGAN-T (Zero-shot, 256x256) 13.9 – Paper Code 2023 linked, not harvested report
39 RAT-GAN 14.6 – Paper Code 2022 linked, not harvested report
40 ERNIE-ViLG 14.7 – Paper Code 2021 linked, not harvested report
41 RA-CM3 (2.7B) 15.7 – Paper – 2022 no code linked report
42 CogView2(6B, Finetuned) 17.7 – Paper Code 2022 0 of 8 ran · 8 unverified report
43 VQ-Diffusion-B 19.75 ✓ Paper Code 2021 linked, not harvested report
44 DM-GAN+CL 20.7933.34 – Paper Code 2021 linked, not harvested report
45 FuseDream (few-shot, k=5) 21.1634.26 – Paper Code 2021 linked, not harvested report
46 FuseDream (k=5, 256) 21.1634.26 – Paper Code 2021 linked, not harvested report
47 FuseDream (k=10, 256) 21.8934.67 – Paper Code 2021 linked, not harvested report
48 AttnGAN+CL 23.9325.70 – Paper Code 2021 linked, not harvested report
49 CogView2(6B, Finetuned) 24 – Paper Code 2022 0 of 8 ran · 8 unverified report
50 OP-GAN 24.7027.8835.85 – Paper Code 2019 4 of 23 ran · 19 unverified report
51 DM-GAN (256 x 256) 26.032.2 – Paper Code 2021 3 of 3 ran · 0 unverified report
52 Lafite (zero-shot) 26.9426.0222.9718.7015.7214.79 – Paper Code 2021 4 of 18 ran · 14 unverified report
53 CogView 27.118.219.413.919.423.6 ✓ Paper Code 2021 4 of 6 ran · 2 unverified report
54 CogView (256 x 256) 27.118.2 – Paper Code 2021 3 of 3 ran · 0 unverified report
55 DALL-E (256 x 256) 27.517.9 – Paper Code 2021 3 of 3 ran · 0 unverified report
56 DALL-E (12B) 28 – Paper – 2022 no code linked report
57 AttnGAN + VICTR 29.2628.18 – Paper Code 2020 linked, not harvested report
58 Vanilla CM3 29.5 – Paper – 2022 no code linked report
59 DM-GAN + VICTR 32.3732.37 – Paper Code 2020 linked, not harvested report
60 DM-GAN 32.6430.4933.44 – Paper Code 2019 4 of 5 ran · 1 unverified report
61 AttnGAN + OP 33.3524.7625.46 – Paper Code 2019 linked, not harvested report
62 AttnGAN (256 x 256) 35.223.3 – Paper Code 2021 3 of 3 ran · 0 unverified report
63 L-Verse-CC 37.231.625.721.421.1 – Paper Code 2021 linked, not harvested report
64 L-Verse 45.841.935.530.229.83 – Paper Code 2021 linked, not harvested report
65 StackGAN + OP 55.3012.12 – Paper Code 2019 linked, not harvested report
66 StackGAN-v1 74.058.45 – Paper Code 2017 10 of 31 ran · 21 unverified report
67 DF-GAN (256 x 256) 18.7 – Paper Code 2021 3 of 3 ran · 0 unverified report
68 StackGAN + VICTR 10.38 – Paper Code 2020 linked, not harvested report
69 ChatPainter 9.74 – Paper – 2018 no code linked report

All 69 rows shown. 69 link to a paper page on this site; 12 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28

Syntology Ran reads "N of M ran · U unverified": of the M code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and the other U = M−N are unverified (harvested, no recorded run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 37 rows have a graph line, from 23 distinct papers; 35 rows (22 papers) have at least one sample that ran. Counting each paper once: Syntology ran 172 of 320 samples; 148 unverified. Separately, 40 of those 320 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Read from the graph 2026-09-24. Per-sample status is on the paper page.

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