-
Notifications
You must be signed in to change notification settings - Fork 2
Expand file tree
/
Copy pathindex.html
More file actions
959 lines (901 loc) · 59 KB
/
index.html
File metadata and controls
959 lines (901 loc) · 59 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<!-- Primary Meta Tags -->
<!-- TODO: Replace with your paper title and author names -->
<meta name="title" content="World-VLA-Loop: Closed-Loop Learning of Video World Model and VLA Policy">
<!-- TODO: Write a compelling 150-160 character description of your research -->
<meta name="description" content="World-VLA-Loop is a framework that establishes a mutually beneficial co-evolving cycle between a robotic world model and a VLA policy to enable efficient reinforcement learning with minimal physical interaction.">
<!-- TODO: Add 5-10 relevant keywords for your research area -->
<meta name="keywords" content="Video Generation, World Model, Robotics, VLA, Reinforcement Learning">
<!-- TODO: List all authors -->
<meta name="author" content="Xiaokang Liu, Zechen Bai, Hai Ci, Kevin Yuchen Ma, Mike Zheng Shou">
<meta name="robots" content="index, follow">
<meta name="language" content="English">
<!-- Open Graph / Facebook -->
<meta property="og:type" content="article">
<!-- TODO: Replace with your institution or lab name -->
<meta property="og:site_name" content="Show Lab, National University of Singapore">
<!-- TODO: Same as paper title above -->
<meta property="og:title" content="World-VLA-Loop: Closed-Loop Learning of Video World Model and VLA Policy">
<!-- TODO: Same as description above -->
<meta property="og:description" content="World-VLA-Loop is a framework that establishes a mutually beneficial co-evolving cycle between a robotic world model and a VLA policy to enable efficient reinforcement learning with minimal physical interaction.">
<!-- TODO: Replace with your actual website URL -->
<meta property="og:url" content="https://hiskiv.github.io/World-VLA-Loop/">
<!-- TODO: Create a 1200x630px preview image and update path -->
<meta property="og:image" content="https://imgur.com/a/k6raYAC">
<meta property="og:image:width" content="1200">
<meta property="og:image:height" content="630">
<meta property="og:image:alt" content="World-VLA-Loop: Closed-Loop Learning of Video World Model and VLA Policy - Research Preview">
<meta property="article:published_time" content="2024-01-01T00:00:00.000Z">
<meta property="article:author" content="Xiaokang Liu">
<meta property="article:section" content="Research">
<meta property="article:tag" content="KEYWORD1">
<meta property="article:tag" content="KEYWORD2">
<!-- Twitter (skip) -->
<meta name="twitter:card" content="summary_large_image">
<!-- TODO: Replace with your lab/institution Twitter handle -->
<meta name="twitter:site" content="@YOUR_TWITTER_HANDLE">
<!-- TODO: Replace with first author's Twitter handle -->
<meta name="twitter:creator" content="@AUTHOR_TWITTER_HANDLE">
<!-- TODO: Same as paper title above -->
<meta name="twitter:title" content="World-VLA-Loop: Closed-Loop Learning of Video World Model and VLA Policy">
<!-- TODO: Same as description above -->
<meta name="twitter:description" content="BRIEF_DESCRIPTION_OF_YOUR_RESEARCH_CONTRIBUTION_AND_FINDINGS">
<!-- TODO: Same as social preview image above -->
<meta name="twitter:image" content="https://YOUR_DOMAIN.com/static/images/social_preview.png">
<meta name="twitter:image:alt" content="World-VLA-Loop: Closed-Loop Learning of Video World Model and VLA Policy - Research Preview">
<!-- Academic/Research Specific (skip) -->
<meta name="citation_title" content="World-VLA-Loop: Closed-Loop Learning of Video World Model and VLA Policy">
<meta name="citation_author" content="Liu, Xiaokang">
<meta name="citation_author" content="Bai, Zechen">
<meta name="citation_publication_date" content="2024">
<meta name="citation_conference_title" content="CONFERENCE_NAME">
<meta name="citation_pdf_url" content="https://YOUR_DOMAIN.com/static/pdfs/paper.pdf">
<!-- Additional SEO -->
<meta name="theme-color" content="#2563eb">
<meta name="msapplication-TileColor" content="#2563eb">
<meta name="apple-mobile-web-app-capable" content="yes">
<meta name="apple-mobile-web-app-status-bar-style" content="default">
<!-- Preconnect for performance -->
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link rel="preconnect" href="https://ajax.googleapis.com">
<link rel="preconnect" href="https://documentcloud.adobe.com">
<link rel="preconnect" href="https://cdn.jsdelivr.net">
<!-- TODO: Replace with your paper title and authors -->
<title>World-VLA-Loop: Closed-Loop Learning of Video World Model and VLA Policy - Xiaokang Liu et al. | Academic Research</title>
<!-- Favicon and App Icons -->
<link rel="icon" type="image/jpeg" href="figures/avatar.jpeg">
<link rel="apple-touch-icon" href="figures/avatar.jpeg">
<!-- Critical CSS - Load synchronously -->
<link rel="stylesheet" href="static/css/bulma.min.css">
<link rel="stylesheet" href="static/css/index.css">
<!-- Non-critical CSS - Load asynchronously -->
<link rel="preload" href="static/css/bulma-carousel.min.css" as="style" onload="this.onload=null;this.rel='stylesheet'">
<link rel="preload" href="static/css/bulma-slider.min.css" as="style" onload="this.onload=null;this.rel='stylesheet'">
<link rel="preload" href="static/css/fontawesome.all.min.css" as="style" onload="this.onload=null;this.rel='stylesheet'">
<link rel="preload" href="https://cdn.jsdelivr.net/gh/jpswalsh/academicons@1/css/academicons.min.css" as="style" onload="this.onload=null;this.rel='stylesheet'">
<!-- Fallback for browsers that don't support preload -->
<noscript>
<link rel="stylesheet" href="static/css/bulma-carousel.min.css">
<link rel="stylesheet" href="static/css/bulma-slider.min.css">
<link rel="stylesheet" href="static/css/fontawesome.all.min.css">
<link rel="stylesheet" href="https://cdn.jsdelivr.net/gh/jpswalsh/academicons@1/css/academicons.min.css">
</noscript>
<!-- Fonts - Optimized loading -->
<link href="https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700;800&display=swap" rel="stylesheet">
<!-- Defer non-critical JavaScript -->
<script defer src="https://ajax.googleapis.com/ajax/libs/jquery/3.5.1/jquery.min.js"></script>
<script defer src="https://documentcloud.adobe.com/view-sdk/main.js"></script>
<script defer src="static/js/fontawesome.all.min.js"></script>
<script defer src="static/js/bulma-carousel.min.js"></script>
<script defer src="static/js/bulma-slider.min.js"></script>
<script defer src="static/js/index.js"></script>
<!-- Structured Data for Academic Papers -->
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "ScholarlyArticle",
"headline": "World-VLA-Loop: Closed-Loop Learning of Video World Model and VLA Policy",
"description": "BRIEF_DESCRIPTION_OF_YOUR_RESEARCH_CONTRIBUTION_AND_FINDINGS",
"author": [
{
"@type": "Person",
"name": "Xiaokang Liu",
"affiliation": {
"@type": "Organization",
"name": "INSTITUTION_NAME"
}
},
{
"@type": "Person",
"name": "SECOND_AUTHOR_NAME",
"affiliation": {
"@type": "Organization",
"name": "INSTITUTION_NAME"
}
}
],
"datePublished": "2024-01-01",
"publisher": {
"@type": "Organization",
"name": "CONFERENCE_OR_JOURNAL_NAME"
},
"url": "https://YOUR_DOMAIN.com/YOUR_PROJECT_PAGE",
"image": "https://YOUR_DOMAIN.com/static/images/social_preview.png",
"keywords": ["KEYWORD1", "KEYWORD2", "KEYWORD3", "machine learning", "computer vision"],
"abstract": "FULL_ABSTRACT_TEXT_HERE",
"citation": "BIBTEX_CITATION_HERE",
"isAccessibleForFree": true,
"license": "https://creativecommons.org/licenses/by/4.0/",
"mainEntity": {
"@type": "WebPage",
"@id": "https://YOUR_DOMAIN.com/YOUR_PROJECT_PAGE"
},
"about": [
{
"@type": "Thing",
"name": "RESEARCH_AREA_1"
},
{
"@type": "Thing",
"name": "RESEARCH_AREA_2"
}
]
}
</script>
<!-- Website/Organization Structured Data -->
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Organization",
"name": "INSTITUTION_OR_LAB_NAME",
"url": "https://YOUR_INSTITUTION_WEBSITE.com",
"logo": "https://YOUR_DOMAIN.com/static/images/favicon.ico",
"sameAs": [
"https://twitter.com/YOUR_TWITTER_HANDLE",
"https://github.com/YOUR_GITHUB_USERNAME"
]
}
</script>
</head>
<body>
<!-- Scroll to Top Button -->
<button class="scroll-to-top" onclick="scrollToTop()" title="Scroll to top" aria-label="Scroll to top">
<i class="fas fa-chevron-up"></i>
</button>
<main id="main-content">
<section class="hero">
<div class="hero-body">
<div class="container is-max-desktop">
<div class="columns is-centered">
<div class="column has-text-centered">
<!-- TODO: Replace with your paper title -->
<h1 class="title is-1 publication-title">World-VLA-Loop : Closed-Loop Learning of Video World Model and VLA Policy</h1>
<div class="is-size-5 publication-authors">
<!-- TODO: Replace with your paper authors and their personal links -->
<span class="author-block">
<a href="https://scholar.google.com/citations?user=dAEHm8AAAAAJ&hl=en" target="_blank">Xiaokang Liu</a><sup>*</sup>,</span>
<span class="author-block">
<a href="https://scholar.google.com/citations?user=aIdQ8GwAAAAJ&hl=en" target="_blank">Zechen Bai</a><sup>*</sup>,</span>
<span class="author-block">
<a href="https://scholar.google.com/citations?user=GMrjppAAAAAJ&hl=en" target="_blank">Hai Ci</a>,</span>
<span class="author-block">
<a href="https://scholar.google.com/citations?user=_6xM_IcAAAAJ&hl=en" target="_blank">Kevin Yuchen Ma</a>,</span>
<span class="author-block">
<a href="https://scholar.google.com/citations?user=h1-3lSoAAAAJ&hl=en" target="_blank">Mike Zheng Shou</a>
</span>
</div>
<div class="is-size-5 publication-authors">
<!-- TODO: Replace with your institution and conference/journal info -->
<span class="author-block">Show Lab<br>National University of Singapore</span>
<!-- TODO: Remove this line if no equal contribution -->
<span class="eql-cntrb"><small><br><sup>*</sup>Indicates Equal Contribution</small></span>
</div>
<div class="column has-text-centered">
<div class="publication-links">
<!-- TODO: Update with your arXiv paper ID -->
<span class="link-block">
<a href="https://arxiv.org/pdf/2602.06508.pdf" target="_blank"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fas fa-file-pdf"></i>
</span>
<span>Paper</span>
</a>
</span>
<!-- TODO: Add your supplementary material PDF or remove this section -->
<!-- <span class="link-block">
<a href="static/pdfs/supplementary_material.pdf" target="_blank"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fas fa-file-pdf"></i>
</span>
<span>Supplementary</span>
</a>
</span> -->
<!-- TODO: Replace with your GitHub repository URL -->
<span class="link-block">
<a href="https://github.com/showlab/World-VLA-Loop" target="_blank"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fab fa-github"></i>
</span>
<span>Github</span>
</a>
</span>
<!-- TODO: Update with your arXiv paper ID -->
<span class="link-block">
<a href="https://www.arxiv.org/abs/2602.06508" target="_blank"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="ai ai-arxiv"></i>
</span>
<span>arXiv</span>
</a>
</span>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<!-- Teaser figure-->
<section class="hero teaser">
<div class="container is-max-desktop">
<div class="hero-body">
<!-- Teaser original a video -->
<!-- <video poster="" id="tree" autoplay controls muted loop height="100%" preload="metadata"> -->
<!-- TODO: Add your video file path here -->
<!-- <source src="static/videos/banner_video.mp4" type="video/mp4"> -->
<!-- </video> -->
<!-- Teaser figure -->
<div style="text-align: center;">
<img src="figures/teaser.png" alt="World-VLA-Loop Teaser" style="width: 100%; max-width: 100%; height: auto;" />
</div>
<!-- TODO: Replace with your figure description -->
<h2 class="subtitle has-text-centered">
(a) Paradigms for world-model-based VLA reinforcement learning. Existing methodologies typically rely on reconstructing the environment within 3D world or training video world models that simulate the environment. To address the imprecise action-following inherent in existing video-based simulators, we propose World-VLA-Loop, a closed-loop paradigm that jointly optimizes the world model and the VLA policy to iteratively enhance the performance and grounding of both. (b) We show that the real-world policy success rate is improved by 36.7% after two iterations of joint optimization with VLA model and world model.
</h2>
</div>
</div>
</section>
<!-- End teaser figure -->
<!-- Paper abstract -->
<section class="section hero is-light">
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column is-four-fifths">
<h2 class="title is-3">Abstract</h2>
<div class="content has-text-justified">
<!-- TODO: Replace with your paper abstract -->
<p>
Recent progress in robotic world models has leveraged video diffusion transformers to predict future observations conditioned on historical states and actions. While these models can simulate realistic visual outcomes, they often exhibit poor action-following precision, hindering their utility for downstream robotic learning. In this work, we introduce World-VLA-Loop, a closed-loop framework for the joint refinement of world models and Vision-Language-Action (VLA) policies. We propose a state-aware video world model that functions as a high-fidelity interactive simulator by jointly predicting future observations and reward signals. To enhance reliability, we introduce the Sans dataset, which incorporates near-Success Trajectory to improve action-outcome alignment within the world model. This framework enables a closed-loop for reinforcement learning (RL) post-training of VLA policies entirely within a virtual environment. Crucially, our approach facilitates a co-evolving cycle: Failure Trajectory rollouts generated by the VLA policy are iteratively fed back to refine the world model’s precision, which in turn enhances subsequent RL optimization. Evaluations across simulation and real-world tasks demonstrate that our framework significantly boosts VLA performance with minimal physical interaction, establishing a mutually beneficial relationship between world modeling and policy learning for general-purpose robotics.
</p>
</div>
</div>
</div>
</div>
</section>
<!-- End paper abstract -->
<section class="section hero is-small">
<div class="container is-max-desktop">
<div class="columns is-centered">
<div class="column is-full">
<div class="content">
<h2 class="title is-3 has-text-centered">Our Method</h2>
<div class="level-set has-text-justified">
<p>
World-VLA-Loop framework consists of four phases.
<ol>
<li>Curate success and near-success dataset (SANS) mainly via manual teleoperation. Few demonstrations are needed.
</li>
<li>Fine-tune the action-conditioned world model on SANS dataset with joint reward and video supervision.</li>
<li>Execute VLA policy rollouts within the world model and perform RL (GRPO) optimization.</li>
<li>Deploy the refined policy in real-world. And in real-world deployment, new rollouts could be used to collect new Failure Trajectory and success data for further SANS dataset augmentation, which can be used to iteratively improve the world model and policy.</li>
</ol>
This cycle enables the joint optimization of the world model and the VLA policy, iteratively enhancing both performance.
</p>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="hero is-small">
<div class="hero-body">
<div class="container">
<div class="columns is-centered">
<div class="column is-two-thirds">
<div class="item">
<!-- Your image here -->
<img src="figures/framework.png" alt="A table comparing probing to previous approaches"/>
<h2 class="subtitle has-text-centered">
<b>Full pipeline of our proposed framework.</b>
</h2>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="section hero is-small">
<div class="container is-max-desktop">
<div class="columns is-centered">
<div class="column is-full">
<div class="content">
<h2 class="title is-3 has-text-centered">Results</h2>
<div class="level-set has-text-justified">
<p>
Success rates are computed across 500 rollouts for the LIBERO suites and 30 physical rollouts for our real-world experiments.
</p>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="hero is-small">
<div class="hero-body">
<div class="container">
<div class="columns is-centered">
<div class="column is-two-thirds">
<div class="item">
<!-- Your image here -->
<img src="figures/RL_performance.png" alt="A table comparing probing to previous approaches"/>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="section hero is-small">
<div class="hero-body">
<div class="container">
<h2 class="title is-3 has-text-centered">World Model Generation Results</h2>
<div class="columns is-centered"><p>Through iterative updating, our world model can generate videos that are highly aligned with conditioning action and adhere to physical dynamics. <b>Note:</b> there is a minor temporal mismatch between the ground-truth and generated videos. This discrepancy arises because we truncate the ground-truth action trajectories to the nearest multiple of our fixed action chunk size rather than using their full original length. To ensure consistent visual comparison, we synchronize the frame rates (FPS) of both videos to achieve an identical total duration.</p></div>
<div class="columns is-centered">
<div class="column is-four-fifths">
<div id="results-carousel" class="carousel results-carousel">
<!-- Page 1: Real-World -->
<div class="item">
<h3 class="title is-4 has-text-centered" style="margin-bottom: 20px;">Real-World</h3>
<div class="columns is-multiline">
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page1/1.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Success Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page1/1-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page1/2.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Failure Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page1/2-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page1/3.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Success Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page1/3-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page1/4.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Failure Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page1/4-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
</div>
</div>
<!-- Page 2: LIBERO-Object -->
<div class="item">
<h3 class="title is-4 has-text-centered" style="margin-bottom: 20px;">LIBERO-Object</h3>
<div class="columns is-multiline">
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page2/1.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Success Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page2/1-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page2/2.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Failure Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page2/2-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page2/3.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Success Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page2/3-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page2/4.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Failure Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page2/4-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
</div>
</div>
<!-- Page 3: LIBERO-Goal -->
<div class="item">
<h3 class="title is-4 has-text-centered" style="margin-bottom: 20px;">LIBERO-Goal</h3>
<div class="columns is-multiline">
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page3/1.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Success Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page3/1-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page3/2.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Failure Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page3/2-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page3/3.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Success Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page3/3-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page3/4.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Failure Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page3/4-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
</div>
</div>
<!-- Page 4: LIBERO-Spatial -->
<div class="item">
<h3 class="title is-4 has-text-centered" style="margin-bottom: 20px;">LIBERO-Spatial</h3>
<div class="columns is-multiline">
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page4/1.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Success Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page4/1-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page4/2.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Failure Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page4/2-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page4/3.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Success Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page4/3-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page4/4.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Failure Trajectory</p>
</div>
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/worldgen_videos/page4/4-GT-p.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 12px; color: #666;">Ground Truth</p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="section hero is-small">
<div class="hero-body">
<div class="container">
<h2 class="title is-3 has-text-centered">World Model Generation from Unseen Trajectory</h2>
<div class="columns is-centered"><p>Our world model still faithfully follow the action condition inputs even with random action sequence unseen during training.</p></div>
<div class="columns is-centered">
<div class="column is-four-fifths">
<div>
<div class="item">
<div class="columns is-multiline">
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop autoplay preload="metadata" style="width: 100%; display: block;">
<source src="figures/ood_cases/1.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop autoplay preload="metadata" style="width: 100%; display: block;">
<source src="figures/ood_cases/2.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop autoplay preload="metadata" style="width: 100%; display: block;">
<source src="figures/ood_cases/3.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop autoplay preload="metadata" style="width: 100%; display: block;">
<source src="figures/ood_cases/4.mp4" type="video/mp4">
</video>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="section hero is-small">
<div class="hero-body">
<div class="container">
<h2 class="title is-3 has-text-centered">RL Fine-tuned VLA Performance</h2>
<div class="columns is-centered"><p>After RL fine-tuning within our world model, VLA actions produce more accurate grasping.</p></div>
<div class="columns is-centered">
<div class="column is-four-fifths">
<div id="results-carousel" class="carousel results-carousel">
<!-- Page 1: Real-World -->
<div class="item">
<h3 class="title is-4 has-text-centered" style="margin-bottom: 20px;">Real-World (Original Speed)</h3>
<div class="columns is-multiline">
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page1/1.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page1/3.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page1/2.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page1/4.mp4" type="video/mp4">
</video>
</div>
</div>
</div>
</div>
<!-- Page 2: LIBERO-Object -->
<div class="item">
<h3 class="title is-4 has-text-centered" style="margin-bottom: 20px;">LIBERO-Object</h3>
<div class="columns is-multiline">
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page2/1.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page2/2.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page2/3.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page2/4.mp4" type="video/mp4">
</video>
</div>
</div>
</div>
</div>
<!-- Page 3: LIBERO-Goal -->
<div class="item">
<h3 class="title is-4 has-text-centered" style="margin-bottom: 20px;">LIBERO-Goal</h3>
<div class="columns is-multiline">
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page3/1.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page3/2.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page3/3.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page3/4.mp4" type="video/mp4">
</video>
</div>
</div>
</div>
</div>
<!-- Page 4: LIBERO-Spatial -->
<div class="item">
<h3 class="title is-4 has-text-centered" style="margin-bottom: 20px;">LIBERO-Spatial</h3>
<div class="columns is-multiline">
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page4/1.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page4/2.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page4/3.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop preload="metadata" style="width: 100%; display: block;">
<source src="figures/VLA_videos/page4/4.mp4" type="video/mp4">
</video>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="section hero is-small">
<div class="hero-body">
<div class="container">
<h2 class="title is-3 has-text-centered">Robustness to OOD Cases</h2>
<div class="columns is-centered"><p>Our model demonstrates robustness to out-of-distribution cases, indicating that the model retains original robustness after our RL fine-tuning.</p></div>
<div class="columns is-centered">
<div class="column is-four-fifths">
<div class="item">
<div class="columns is-multiline">
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop autoplay preload="metadata" style="width: 100%; display: block;">
<source src="figures/unseen_cases/1.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Unseen Lighting Condition</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop autoplay preload="metadata" style="width: 100%; display: block;">
<source src="figures/unseen_cases/2.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Unseen Distractor</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop autoplay preload="metadata" style="width: 100%; display: block;">
<source src="figures/unseen_cases/3.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Unseen Distractors</p>
</div>
</div>
<div class="column is-one-quarter">
<div style="margin-bottom: 10px;">
<video controls muted loop autoplay preload="metadata" style="width: 100%; display: block;">
<source src="figures/unseen_cases/4.mp4" type="video/mp4">
</video>
<p class="has-text-centered" style="margin-top: 5px; margin-bottom: 0; font-weight: 600; font-size: 14px;">Unseen Distractors with Disturbance</p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<!--BibTex citation -->
<section class="section" id="BibTeX">
<div class="container is-max-desktop content">
<div class="bibtex-header">
<h2 class="title">BibTeX</h2>
<button class="copy-bibtex-btn" onclick="copyBibTeX()" title="Copy BibTeX to clipboard">
<i class="fas fa-copy"></i>
<span class="copy-text">Copy</span>
</button>
</div>
<pre id="bibtex-code"><code>@article{liu2026worldvlaloop,
title={World-VLA-Loop: Closed-Loop Learning of Video World Model and VLA Policy},
author={Liu, Xiaokang and Bai, Zechen and Ci, Hai and Ma, Kevin Yuchen and Shou, Mike Zheng},
journal={arXiv preprint arXiv:2602.06508},
year={2026}
}</code></pre>
</div>
</section>
<!--End BibTex citation -->
<footer class="footer">
<div class="container">
<div class="columns is-centered">
<div class="column is-8">
<div class="content">
<p>
This page was built using the <a href="https://github.com/eliahuhorwitz/Academic-project-page-template" target="_blank">Academic Project Page Template</a> which was adopted from the <a href="https://nerfies.github.io" target="_blank">Nerfies</a> project page.
You are free to borrow the source code of this website, we just ask that you link back to this page in the footer. <br> This website is licensed under a <a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/" target="_blank">Creative
Commons Attribution-ShareAlike 4.0 International License</a>.
</p>
</div>
</div>
</div>
</div>
</footer>
<!-- Statcounter tracking code -->
<!-- You can add a tracker to track page visits by creating an account at statcounter.com -->
<!-- End of Statcounter Code -->
</body>
</html>