Skip to content
ModelListedExternal
Download

PRSIMVL LoRA V1

This folder stores the released PRSIMVL LoRA adapter layout used by inference and evaluation.

Sourcehuggingface.co/kepeng/PRSIMVL-LoRA-V1/tree/217866a7fb18e5cca1795eaade5664f357c5e932
KindOtherVersionAdapter modelBgrade

Model source

Kind
Other
File size
45.4–98.1 MB
Version
vAdapter model
Updated
Oct 10, 2026
Source
Hugging Face

Source description

Released PRSIMVL Weights

This folder stores the released PRSIMVL LoRA adapter layout used by inference and evaluation.

Expected Checkpoints

SizeBase ModelLocal LoRA Checkpoint
2BQwen/Qwen3-VL-2B-InstructBANALCED_150K_META_VIT_PROXY/output-Qwen3-VL-2B-Instruct/v8-20260421-133546/checkpoint-95000
4BQwen/Qwen3-VL-4B-InstructBANALCED_150K_META_VIT_PROXY/output-Qwen3-VL-4B-Instruct/v12-20260425-113029/checkpoint-85000
8BQwen/Qwen3-VL-8B-InstructBANALCED_150K_META_VIT_PROXY/output-Qwen3-VL-8B-Instruct/v2-20260423-205317/checkpoint-95000
@misc{xu2026allegory,
  title         = {Allegory of the Cave: Measurement-Grounded Vision-Language Learning},
  author        = {Xu, Kepeng and Xu, Li and He, Gang and Yu, Wenxin},
  year          = {2026},
  eprint        = {2605.11727},
  archivePrefix = {arXiv},
  url           = {https://arxiv.org/abs/2605.11727}
}

Sources

1 source
Hugging FacePrimaryVersion 217866a7fb18e5cca1795eaade5664f357c5e932
huggingface.co/kepeng/PRSIMVL-LoRA-V1/tree/217866a7fb18e5cca1795eaade5664f357c5e932

Verified Oct 10

Model artifacts

3 artifacts

output-Qwen3-VL-2B-Instruct/v8-20260421-133546/checkpoint-95000/adapter_model.safetensors

safetensors · 45.4 MB · SHA-256 1c29fc75f547…955d · Hugging Face

Download

output-Qwen3-VL-4B-Instruct/v12-20260425-113029/checkpoint-85000/adapter_model.safetensors

safetensors · 75.2 MB · SHA-256 59a402a46e0b…9dcf · Hugging Face

Download

output-Qwen3-VL-8B-Instruct/v2-20260423-205317/checkpoint-95000/adapter_model.safetensors

safetensors · 98.1 MB · SHA-256 ea38db41f51b…fb36 · Hugging Face

Download

Source excerpts

2 excerpts
ReadmeAug 1huggingface.co/kepeng/PRSIMVL-LoRA-V1

--- license: cc-by-nc-4.0 task_categories: - visual-question-answering - image-to-text language: - en tags: - vision-language - multimodal - visual-question-answering - measurement-grounding - raw-image - camera-raw - meas-xyz - low-light - hdr - benchmark - prsimvl pretty_name: kepeng/PRSIMVL-LoRA-V1 --- # Released PRSIMVL Weights This folder stores the released PRSIMVL LoRA adapter layout used by inference and evaluation. ## Expected Checkpoints | Size | Base Model | Local LoRA Checkpoint | |---|---|---| | 2B | `Qwen/Qwen3-VL-2B-Instruct` | `BANALCED_150K_META_VIT_PROXY/output-Qwen3-VL-2B-Instruct/v8-20260421-133546/checkpoint-95000` | | 4B | `Qwen/Qwen3-VL-4B-Instruct` | `BANALCED_150K_META_VIT_PROXY/output-Qwen3-VL-4B-Instruct/v12-20260425-113029/checkpoint-85000` | | 8B | `Qwen/Qwen3-VL-8B-Instruct` | `BANALCED_150K_META_VIT_PROXY/output-Qwen3-VL-8B-Instruct/v2-20260423-205317/checkpoint-95000` | ``` @misc{xu2026allegory, title = {Allegory of the Cave: Measurement-Grounded Vision-Language Learning}, author = {Xu, Kepeng and Xu, Li and He, Gang and Yu, Wenxin}, year = {2026}, eprint = {2605.11727}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2605.11727} } ```

NoteAug 1huggingface.co/kepeng/PRSIMVL-LoRA-V1

Source context: 0 downloads · 11 likes · Pipeline visual-question-answering · Repo kepeng/PRSIMVL-LoRA-V1