| --- |
| license: apache-2.0 |
| library_name: pytorch |
| tags: |
| - multimodal-emotion-recognition |
| - knowledge-distillation |
| - light-mer |
| - swd-h |
| - qwen3 |
| --- |
| |
| # Light-MER Checkpoints |
|
|
| This repository hosts model checkpoints for **Light-MER**, the official implementation of: |
|
|
| **Do We Really Need Multimodal Emotion Language Models Larger Than 1B Parameters?** |
|
|
| GitHub: https://github.com/kevinkke233-maker/Light-MER |
|
|
| ## Checkpoints |
|
|
| | File | Description | Status | |
| |---|---|---| |
| | `light-mer-teacher-qwen3-8b.pth` | Qwen3-8B teacher checkpoint for Stage 1 SWD-H distillation | Released | |
| | `stage1-swdh-qwen3-0.6b/checkpoint_000060_loss_1.291.pth` | Recommended Qwen3-0.6B student checkpoint after Stage 1 SWD-H distillation | Released | |
| | `stage1-swdh-qwen3-0.6b/checkpoint_*.pth` | Stage 1 SWD-H checkpoint trajectory from epoch 5 to epoch 60 | Released | |
| | `stage1-swdh-qwen3-0.6b/config.yaml` | Stage 1 SWD-H training config used for the released checkpoints | Released | |
| | `light-mer-stage2-mgrpo-qwen3-0.6b.pth` | Qwen3-0.6B student after Stage 2 M-GRPO refinement | COMING SOON | |
|
|
| ## Usage |
|
|
| Download the teacher checkpoint and place it under the GitHub repo checkpoint directory: |
|
|
| ```text |
| checkpoints/light-mer-teacher-qwen3-8b.pth |
| ``` |
|
|
| Then run Stage 1 SWD-H distillation with: |
|
|
| ```bash |
| CONDA_ENV_NAME=swdh-stage1 \ |
| TEACHER_CKPT=checkpoints/light-mer-teacher-qwen3-8b.pth \ |
| bash scripts/train_stage1_swdh.sh |
| ``` |
|
|
| For Stage 1 inference or evaluation, download one of the released student checkpoints, for example: |
|
|
| ```text |
| checkpoints/light-mer-stage1-swdh-qwen3-0.6b.pth |
| ``` |
|
|
| The recommended Stage 1 checkpoint is: |
|
|
| ```text |
| stage1-swdh-qwen3-0.6b/checkpoint_000060_loss_1.291.pth |
| ``` |
|
|
| ## License |
|
|
| This checkpoint repository is released under the Apache License 2.0. Please also follow the licenses and usage terms of the external datasets and pretrained models used with Light-MER. |
|
|
