davanstrien / qwen35-4b-iconclass-grpo-v5full

huggingface.co
Total runs: 4
24-hour runs: 0
7-day runs: 1
30-day runs: 1
Model's Last Updated: June 05 2026
image-text-to-text

Introduction of qwen35-4b-iconclass-grpo-v5full

Model Details of qwen35-4b-iconclass-grpo-v5full

qwen35-4b-iconclass-grpo-v5full

GRPO reward-ablation checkpoint (Qwen3.5-4B-VL, from davanstrien/qwen35-4b-iconclass-vlm ). Part of an experiment testing whether a richer reward bundle beats plain hierarchical-F1 ( gt_match ) for iconclass classification.

  • Reward config : recall + validity + count + diversity
  • Result (completeness-corrected H-F1, 40-image test) : 61.8%
  • Verdict : no improvement over plain gt_match (all variants 61–64%, within n=40 noise). Reward tuning is not the lever — the model is capability-bound. The approach that worked is anchored fusion (see qwen35-4b-iconclass-sft-brillfull ).

Base: davanstrien/qwen35-4b-iconclass-vlm . Trained with Unsloth + TRL.

Runs of davanstrien qwen35-4b-iconclass-grpo-v5full on huggingface.co

4
Total runs
0
24-hour runs
2
3-day runs
1
7-day runs
1
30-day runs

More Information About qwen35-4b-iconclass-grpo-v5full huggingface.co Model

More qwen35-4b-iconclass-grpo-v5full license Visit here:

https://choosealicense.com/licenses/apache-2.0

qwen35-4b-iconclass-grpo-v5full huggingface.co

qwen35-4b-iconclass-grpo-v5full huggingface.co is an AI model on huggingface.co that provides qwen35-4b-iconclass-grpo-v5full's model effect (), which can be used instantly with this davanstrien qwen35-4b-iconclass-grpo-v5full model. huggingface.co supports a free trial of the qwen35-4b-iconclass-grpo-v5full model, and also provides paid use of the qwen35-4b-iconclass-grpo-v5full. Support call qwen35-4b-iconclass-grpo-v5full model through api, including Node.js, Python, http.

qwen35-4b-iconclass-grpo-v5full huggingface.co Url

https://huggingface.co/davanstrien/qwen35-4b-iconclass-grpo-v5full

davanstrien qwen35-4b-iconclass-grpo-v5full online free

qwen35-4b-iconclass-grpo-v5full huggingface.co is an online trial and call api platform, which integrates qwen35-4b-iconclass-grpo-v5full's modeling effects, including api services, and provides a free online trial of qwen35-4b-iconclass-grpo-v5full, you can try qwen35-4b-iconclass-grpo-v5full online for free by clicking the link below.

davanstrien qwen35-4b-iconclass-grpo-v5full online free url in huggingface.co:

https://huggingface.co/davanstrien/qwen35-4b-iconclass-grpo-v5full

qwen35-4b-iconclass-grpo-v5full install

qwen35-4b-iconclass-grpo-v5full is an open source model from GitHub that offers a free installation service, and any user can find qwen35-4b-iconclass-grpo-v5full on GitHub to install. At the same time, huggingface.co provides the effect of qwen35-4b-iconclass-grpo-v5full install, users can directly use qwen35-4b-iconclass-grpo-v5full installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

qwen35-4b-iconclass-grpo-v5full install url in huggingface.co:

https://huggingface.co/davanstrien/qwen35-4b-iconclass-grpo-v5full

Url of qwen35-4b-iconclass-grpo-v5full

qwen35-4b-iconclass-grpo-v5full huggingface.co Url

Provider of qwen35-4b-iconclass-grpo-v5full huggingface.co

davanstrien
ORGANIZATIONS

Other API from davanstrien

huggingface.co

Total runs: 14
Run Growth: 4
Growth Rate: 28.57%
Updated:September 24 2024