Visuotactile Paper Artifacts
Small trained artifacts and feature caches for this paper's visuotactile training pipeline.
Source du modèle
- Type
- Checkpoint, LoRA
- Taille du fichier
- 578 KB–420 MB
- Version
- vRevision 0cde066
- Mis à jour
- 2 oct. 2026
- Éditeur
- @wy891
- Téléchargements
- 0
- Source
- Hugging Face
Description de la source
Visuotactile Paper Artifacts
Small trained artifacts and feature caches for this paper's visuotactile training pipeline.
Source
These artifacts come from the training pipeline in https://github.com/howard-lynn-ye/vr-teleop-isaacsim. Paper draft: https://github.com/howard-lynn-ye/vr-teleop-isaacsim/blob/main/paper/main.tex.
Pinned source commit for these uploaded artifacts: 8084ec32d37f07d57bd71e424d6144aae9e4f8b9.
Contents
phase1/seed_42/best.pt,phase1/seed_42/last.pt: Phase 1 alignment encoder checkpoints.phase1/features_cache/: canonical Phase 1 DINOv2 feature caches and split manifest.phase3_3/seed_42/best.pt,phase3_3/seed_42/last.pt,phase3_3/seed_42/train.log: Phase 3.3 sim alignment artifacts.
Reproducibility
- Canonical Phase 1 data split:
DATA_SEED=42. - Phase 1 training script:
https://github.com/howard-lynn-ye/vr-teleop-isaacsim/blob/8084ec32d37f07d57bd71e424d6144aae9e4f8b9/path_d/code/train/phase1_5_5seed.py. - Phase 3.3 training script:
https://github.com/howard-lynn-ye/vr-teleop-isaacsim/blob/8084ec32d37f07d57bd71e424d6144aae9e4f8b9/path_d/code/train/phase3_3_alignment.py. - Phase 1 feature cache manifests record
cache_version=2,n_rows=4583,top_n=30, and the sharedsample_ids_sha256.
Critical Caveats
- Phase 1 checkpoint:
best_ep=67,val_loss=2.5115,top5=0.6876. This post-restore retrain matches the original within epsilon. - Phase 3.3 uses the held-out-scale split required by gamma_gates compliance: train on s15+s20 and validate on s25.
- These are paper artifacts, not a turnkey production model package. Preserve the split and script commit when comparing numbers.
Related Repositories
- SSVTP source corpus: https://huggingface.co/datasets/mlfu7/Touch-Vision-Language-Dataset
- Imputed force sidecars: https://huggingface.co/datasets/wy891/ssvtp-claude-imputed-force
- Unified v2t generator: https://huggingface.co/wy891/v2t-unified-ssvtp-diffusion
Sources
1 source
Sources
1 sourceVérifié 2 oct.
Artefacts du modèle
61 artefacts
Artefacts du modèle
61 artefactsgenerator_rescue/round91_pix2pix_v2_regularized/main_checkpoint/pix2pix_step_1000.pt
pt · 53,4 MB · SHA-256 71a8c3d4f160…ede5 · Hugging Face
Téléchargergenerator_rescue/round93_deformation_amp_rescue/main_checkpoint/pix2pix_step_1000.pt
pt · 53,4 MB · SHA-256 59d8888aa666…58da · Hugging Face
Téléchargergenerator_rescue/round95_r93_finetune_rescue/main_checkpoint/pix2pix_r95c_ft_deformmean_step100.pt
pt · 53,4 MB · SHA-256 9ead671430b2…7112 · Hugging Face
Téléchargergenerator_rescue/round96_early_finetune_grid/main_checkpoint/pix2pix_r96c_mildcontact_lr2e5_step150.pt
pt · 53,4 MB · SHA-256 fbb397e514f0…8f05 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N1000_seed42/best.pt
pt · 1,70 MB · SHA-256 0d0694d344a4…4f90 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N1000_seed42/last.pt
pt · 1,70 MB · SHA-256 814794236515…d107 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N1000_seed43/best.pt
pt · 1,70 MB · SHA-256 ad9e4dda4152…b7f9 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N1000_seed43/last.pt
pt · 1,70 MB · SHA-256 27f9a6e6ecd9…f960 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N1000_seed44/best.pt
pt · 1,70 MB · SHA-256 243096e915d1…7442 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N1000_seed44/last.pt
pt · 1,70 MB · SHA-256 17198bf2bf69…0a11 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N1000_seed45/best.pt
pt · 1,70 MB · SHA-256 c300c5729521…8d1b · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N1000_seed45/last.pt
pt · 1,70 MB · SHA-256 e8a70abac324…81fc · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N1000_seed46/best.pt
pt · 1,70 MB · SHA-256 479e8a1e6c89…fc24 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N1000_seed46/last.pt
pt · 1,70 MB · SHA-256 5468577b7fff…37f7 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N2000_seed42/best.pt
pt · 1,70 MB · SHA-256 9db415982b2f…0af1 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N2000_seed42/last.pt
pt · 1,70 MB · SHA-256 db9a502b64fc…d2bc · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N2000_seed43/best.pt
pt · 1,70 MB · SHA-256 278a381f0f6a…2e74 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N2000_seed43/last.pt
pt · 1,70 MB · SHA-256 dffe6d6d5d79…9171 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N2000_seed44/best.pt
pt · 1,70 MB · SHA-256 afb742960e5d…211d · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N2000_seed44/last.pt
pt · 1,70 MB · SHA-256 dbafba600395…aaa4 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N2000_seed45/best.pt
pt · 1,70 MB · SHA-256 b5ea8eced430…cb9f · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N2000_seed45/last.pt
pt · 1,70 MB · SHA-256 eee986f73ce0…767c · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N2000_seed46/best.pt
pt · 1,70 MB · SHA-256 89b482241044…6137 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N2000_seed46/last.pt
pt · 1,70 MB · SHA-256 5a20c454d773…2c38 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N4124_seed44/best.pt
pt · 1,70 MB · SHA-256 9177f63fe35c…06ae · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N4124_seed44/last.pt
pt · 1,70 MB · SHA-256 1c2ca74cc93f…b7f0 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N4124_seed45/best.pt
pt · 1,70 MB · SHA-256 a6578c2f51b1…c8cc · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N4124_seed45/last.pt
pt · 1,70 MB · SHA-256 576325bf77dd…881b · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N4124_seed46/best.pt
pt · 1,70 MB · SHA-256 22d2a705ec7a…35f9 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N4124_seed46/last.pt
pt · 1,70 MB · SHA-256 544442d7fc77…da43 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N500_seed44/best.pt
pt · 1,70 MB · SHA-256 ff73c9618873…370f · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N500_seed44/last.pt
pt · 1,70 MB · SHA-256 9b9589108622…771e · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N500_seed45/best.pt
pt · 1,70 MB · SHA-256 f3a02c72ed2e…5af2 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N500_seed45/last.pt
pt · 1,70 MB · SHA-256 a5b4758a94b3…9314 · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N500_seed46/best.pt
pt · 1,70 MB · SHA-256 238027492ae7…5beb · Hugging Face
Téléchargerpaper/nips2026/round126b_full_grid_artifacts/N500_seed46/last.pt
pt · 1,70 MB · SHA-256 055fb62117b1…ba28 · Hugging Face
Téléchargerround121_text_film_generator/checkpoints/best_text_pix2pix.pt
pt · 56,8 MB · SHA-256 f96d3b7b88f3…a6e7 · Hugging Face
Téléchargerround121_text_film_generator/checkpoints/final_text_pix2pix.pt
pt · 56,8 MB · SHA-256 1676d8b30296…5116 · Hugging Face
Téléchargerround124_cliptext_bgsub/checkpoints/cliptext_pix2pix_step_12000.pt
pt · 68,0 MB · SHA-256 18b85172f97c…fb38 · Hugging Face
Téléchargerround126_cliptext_strong/checkpoints/cliptext_pix2pix_step_12000.pt
pt · 68,0 MB · SHA-256 91a0a21e21eb…9eb1 · Hugging Face
Téléchargerround126_cliptext_strong/checkpoints/cliptext_pix2pix_step_3000.pt
pt · 68,0 MB · SHA-256 5a32d2d2833b…5d85 · Hugging Face
Téléchargerround127_cliptext_textdominant/checkpoints/cliptext_pix2pix_step_11000.pt
pt · 68,0 MB · SHA-256 a51f5a461d99…875c · Hugging Face
Téléchargerround127_cliptext_textdominant/checkpoints/cliptext_pix2pix_step_12000.pt
pt · 68,0 MB · SHA-256 ff4ac7558da6…c096 · Hugging Face
Téléchargerround127_cliptext_textdominant/checkpoints/cliptext_pix2pix_step_3000.pt
pt · 68,0 MB · SHA-256 44ee5d9f95e0…1c9f · Hugging Face
Téléchargerround127_cliptext_textdominant/checkpoints/cliptext_pix2pix_step_8000.pt
pt · 68,0 MB · SHA-256 33d16da54401…9184 · Hugging Face
Téléchargerround130_vtl_aligned_generator/checkpoints/r130b_cliptext_step1000.pt
pt · 68,0 MB · SHA-256 7c6c5daf54ba…f887 · Hugging Face
Téléchargerround134_135_trainable_qwen_prefix/r134_prefix_ce_r131e_realgen_e12_lr2e3/trainable_qwen_tactile_prefix.pt
pt · 66,1 MB · SHA-256 b1f630c3c65e…ab25 · Hugging Face
Téléchargerround134_135_trainable_qwen_prefix/r134_prefix_margin_r131a_realgen_e20_lr1e3/trainable_qwen_tactile_prefix.pt
pt · 66,1 MB · SHA-256 f82265bd9771…c260 · Hugging Face
Téléchargerround134_135_trainable_qwen_prefix/r134_prefix_margin_r131e_real_e20_lr1e3/trainable_qwen_tactile_prefix.pt
pt · 66,1 MB · SHA-256 327fada8d3a0…5403 · Hugging Face
Téléchargerround134_135_trainable_qwen_prefix/r134_prefix_margin_r131e_realgen_e20_lr1e3/trainable_qwen_tactile_prefix.pt
pt · 66,1 MB · SHA-256 f079d7425787…e12c · Hugging Face
Téléchargerround134_135_trainable_qwen_prefix/r135_lora_margin_r131c_realgen_e3_lr2e4/qwen_lora_adapter/adapter_model.safetensors
safetensors · 35,5 MB · SHA-256 650797ee5925…9d2f · Hugging Face
Téléchargerround134_135_trainable_qwen_prefix/r135_lora_margin_r131c_realgen_e3_lr2e4/trainable_qwen_tactile_prefix.pt
pt · 66,1 MB · SHA-256 273d36eb7fee…83a8 · Hugging Face
Téléchargerround134_135_trainable_qwen_prefix/r135_lora_margin_r131e_realgen_e3_lr2e4/qwen_lora_adapter/adapter_model.safetensors
safetensors · 35,5 MB · SHA-256 fbaba6aba89c…bfd3 · Hugging Face
Téléchargerround134_135_trainable_qwen_prefix/r135_lora_margin_r131e_realgen_e3_lr2e4/trainable_qwen_tactile_prefix.pt
pt · 66,1 MB · SHA-256 5423e03637fb…d472 · Hugging Face
Téléchargerround134_135_trainable_qwen_prefix/r135b_qformer_ce_r131e_realgen_e10_lr3e4/trainable_qwen_tactile_prefix.pt
pt · 228 MB · SHA-256 34973a38bcee…f659 · Hugging Face
Téléchargerround134_135_trainable_qwen_prefix/r135b_qformer_margin_r131e_realgen_e16_lr1e4/trainable_qwen_tactile_prefix.pt
pt · 228 MB · SHA-256 65e5219398d4…ea8f · Hugging Face
Téléchargerround134_135_trainable_qwen_prefix/r135b_qformer_margin_r131e_realgen_e20_lr8e4/trainable_qwen_tactile_prefix.pt
pt · 420 MB · SHA-256 674882b6fbae…0b87 · Hugging Face
TéléchargerExtraits de sources
3 extraits
Extraits de sources
3 extraits--- license: cc-by-4.0 tags: - tactile - vision-tactile - visuotactile - physical-property - ssvtp - dinov2 --- # Visuotactile Paper Artifacts Small trained artifacts and feature caches for this paper's visuotactile training pipeline. ## Source These artifacts come from the training pipeline in `https://github.com/howard-lynn-ye/vr-teleop-isaacsim`. Paper draft: `https://github.com/howard-lynn-ye/vr-teleop-isaacsim/blob/main/paper/main.tex`. Pinned source commit for these uploaded artifacts: `8084ec32d37f07d57bd71e424d6144aae9e4f8b9`. ## Contents - `phase1/seed_42/best.pt`, `phase1/seed_42/last.pt`: Phase 1 alignment encoder checkpoints. - `phase1/features_cache/`: canonical Phase 1 DINOv2 feature caches and split manifest. - `phase3_3/seed_42/best.pt`, `phase3_3/seed_42/last.pt`, `phase3_3/seed_42/train.log`: Phase 3.3 sim alignment artifacts. ## Reproducibility - Canonical Phase 1 data split: `DATA_SEED=42`. - Phase 1 training script: `https://github.com/howard-lynn-ye/vr-teleop-isaacsim/blob/8084ec32d37f07d57bd71e424d6144aae9e4f8b9/path_d/code/train/phase1_5_5seed.py`. - Phase 3.3 training script: `https://github.com/howard-lynn-ye/vr-teleop-isaacsim/blob/8084ec32d37f07d57bd71e424d6144aae9e4f8b9/path_d/code/train/phase3_3_alignment.py`. - Phase 1 feature cache manifests record `cache_version=2`, `n_rows=4583`, `top_n=30`, and the shared `sample_ids_sha256`. ## Critical Caveats - Phase 1 checkpoint: `best_ep=67`, `val_loss=2.5115`, `top5=0.6876`. This post-restore retrain matches the original within epsilon. - Phase 3.3 uses the held-out-scale split required by gamma_gates compliance: train on s15+s20 and validate on s25. - These are paper artifacts, not a turnkey production model package. Preserve the split and script commit when comparing numbers. ## Related Repositories - SSVTP source corpus: https://huggingface.co/datasets/mlfu7/Touch-Vision-Language-Dataset - Imputed force sidecars: https://huggingface.co/datasets/wy891/ssvtp-claude-imputed-force - Unified v2t generator: https://huggingface.co/wy891/v2t-unified-ssvtp-diffusion
Source context: 0 downloads · 0 likes · Repo wy891/visuotactile-paper-artifacts
Source context: 0 downloads · 0 likes · Repo wy891/visuotactile-paper-artifacts
