You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

LAP-VQA — GPT-3-Medium-style decoder

Best checkpoint from Diff-VQA online multi-objective (reg-steps) training.

Weights only — this file contains state_dict + architecture config (vis_dim, d_model, num_layers, ...). It does not contain optimizer or LR-scheduler state, so it cannot be used to exactly resume an interrupted training run (fine for eval / deployment / fine-tuning from weights).

⚠️ Gated: medical-imaging model weights, access granted on request.

Downloads last month: -; Downloads are not tracked for this model. How to track