Hugging Face's logo

Spaces:
bertfil
/
VzdSHEab4G
Running

App Files Community
VzdSHEab4G
File size: 3,024 Bytes
42b6cb6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
{
  "schema_version": 1,
  "title": "Repro - Pretrained Vision-Language-Action Models are Surprisingly Resistant to Forgetting in Continual Learning",
  "emoji": "\ud83d\udd2c",
  "space_id": "bertfil/VzdSHEab4G",
  "paper": {
    "arxiv_id": "2603.03818",
    "openreview_id": "VzdSHEab4G",
    "title": "Pretrained Vision-Language-Action Models are Surprisingly Resistant to Forgetting in Continual Learning",
    "url": "https://arxiv.org/abs/2603.03818"
  },
  "tags": [
    "icml2026-repro",
    "paper-VzdSHEab4G"
  ],
  "updated_at": "2026-07-20T03:52:38+00:00",
  "root": {
    "slug": "index",
    "title": "Repro - Pretrained Vision-Language-Action Models are Surprisingly Resistant to Forgetting in Continual Learning",
    "file": "pages/index.md",
    "children": [
      {
        "slug": "00-scored-evidence-summary",
        "title": "00 - Scored evidence summary",
        "file": "pages/00-scored-evidence-summary/page.md",
        "children": []
      },
      {
        "slug": "claim-1-at-a-20-replay-buffer-size-pretrained-vlas-pi0-and",
        "title": "Claim 1 - At a 20% replay buffer size, pretrained VLAs Pi0 and GR00T a\u2026",
        "file": "pages/claim-1-at-a-20-replay-buffer-size-pretrained-vlas-pi0-and/page.md",
        "children": []
      },
      {
        "slug": "claim-2-at-a-2-replay-buffer-size-vlas-maintain-nbt-of-0-1",
        "title": "Claim 2 - At a 2% replay buffer size, VLAs maintain NBT of 0.1-0.2 whi\u2026",
        "file": "pages/claim-2-at-a-2-replay-buffer-size-vlas-maintain-nbt-of-0-1/page.md",
        "children": []
      },
      {
        "slug": "claim-3-ablating-pi0-s-pretraining-source-shows-pi0-pretra",
        "title": "Claim 3 - Ablating Pi0's pretraining source shows Pi0 pretrained on vi\u2026",
        "file": "pages/claim-3-ablating-pi0-s-pretraining-source-shows-pi0-pretra/page.md",
        "children": []
      },
      {
        "slug": "claim-4-recovery-efficiency-analysis-shows-pi0-requires-on",
        "title": "Claim 4 - Recovery efficiency analysis shows Pi0 requires only 6-10% o\u2026",
        "file": "pages/claim-4-recovery-efficiency-analysis-shows-pi0-requires-on/page.md",
        "children": []
      },
      {
        "slug": "claim-5-experiments-use-the-libero-benchmark-suite-libero",
        "title": "Claim 5 - Experiments use the LIBERO benchmark suite (LIBERO-Spatial, \u2026",
        "file": "pages/claim-5-experiments-use-the-libero-benchmark-suite-libero/page.md",
        "children": []
      },
      {
        "slug": "reproduction-protocol-and-provenance",
        "title": "Reproduction protocol and provenance",
        "file": "pages/reproduction-protocol-and-provenance/page.md",
        "children": []
      },
      {
        "slug": "conclusion",
        "title": "Conclusion",
        "file": "pages/conclusion/page.md",
        "children": []
      }
    ]
  },
  "agent_view_tokens": 6198,
  "revision": "1784519558222318200"
}