shahedm2001/qwen3-vl-2b-cataract-grpo

Upload model comparison benchmark figure figure_task_granularity_comparison.png

Upload model comparison benchmark figure figure_model_comparison_benchmark.png

Upload model comparison benchmark figure figure_task_granularity_comparison.png

Upload model comparison benchmark figure figure_model_comparison_benchmark.png

Add SFT Stages 1 & 2 concatenated training dynamics plot

Add complete SFT Stages 1 & 2 logs archive

Update figure_continuous_distributions.png regenerated purely from Hugging Face telemetry dataset

Update figure_continuous_rewards.png regenerated purely from Hugging Face telemetry dataset

Update figure_binary_diagnostics.png regenerated purely from Hugging Face telemetry dataset

Update figure_binary_tasks.png regenerated purely from Hugging Face telemetry dataset

Add sample mapping reproduction script and tutorial

Add complete slice-to-sample training telemetry CSV

Add final complete TensorBoard archive (Stages 1 & 2 complete)

Update complete final TensorBoard data (Stages 1 & 2 complete)

Add final complete logs archive (Stages 1 & 2 complete)

Update complete final training logs (Stages 1 & 2 complete)

Add dedicated data group evaluation plots and Two-Stage GRPO details

Upload figure_group3_continuous_temporal.png

Upload figure_group2_phase_recognition.png

Upload figure_group1_surgical_mcqs.png

Update model card with Two-Stage GRPO training architecture & evaluation

Upload publication plot figure_reward_types_diagnostics.png

Upload publication plot figure_continuous_distributions.png

Upload publication plot figure_continuous_rewards.png

Upload publication plot figure_binary_tasks.png

Upload publication plot figure_step_diagnostics.png

Upload publication plot figure_step_core_dynamics.png

Upload publication plot figure_stage4_phase_specialization.png

Training configs (2026-09-25)

Training logs (2026-09-25)

Fine-tuned model (2026-09-25)

Sync TensorBoard data archive at step 60

Sync TensorBoard data into checkpoint-60_stage4_phase_20260925_144344

Add cmd.txt to checkpoint-60_stage4_phase_20260925_144344

Add config.txt to checkpoint-60_stage4_phase_20260925_144344

Sync training logs archive at step 60

Sync training logs into checkpoint-60_stage4_phase_20260925_144344

Add GRPO adapter checkpoint-60_stage4_phase_20260925_144344 (step 60)

Sync TensorBoard data archive at step 40

Sync TensorBoard data into checkpoint-40_stage4_phase_20260925_131602

Add cmd.txt to checkpoint-40_stage4_phase_20260925_131602

Add config.txt to checkpoint-40_stage4_phase_20260925_131602

Sync training logs archive at step 40

Sync training logs into checkpoint-40_stage4_phase_20260925_131602

Add GRPO adapter checkpoint-40_stage4_phase_20260925_131602 (step 40)

Sync TensorBoard data archive at step 38

Sync TensorBoard data into checkpoint-20_stage4_phase_20260925_130751

Add cmd.txt to checkpoint-20_stage4_phase_20260925_130751

Add config.txt to checkpoint-20_stage4_phase_20260925_130751

Sync training logs archive at step 38