GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
phanviethoang1512
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
phanviethoang1512
Model Tree
Qwen/Qwen3-8B
Input Modalities
Output Modalities
Supported Functionality
Mean task score of the five Qwen3-8B agents from the paper across eight UserRL environments, three of them held out from training (IntentionGym, TelepathyGym, SearchGym), under nine evaluation user models that were not used in training.
| Agent | Training user | Objective | Mean score |
|---|---|---|---|
| Base | none | SFT | 18.49 |
| UserRL | GPT-5.5 | GRPO | 26.10 |
| InfoPO | MIMESIS-9B | InfoPO | 26.65 |
| UserRL+ | MIMESIS-9B | GRPO | 29.54 |
| CSD (this model) | MIMESIS-9B | GRPO + CSD | 31.09 |
With the training user held fixed, CSD improves on GRPO (UserRL+) under all nine evaluation users. Each score is the mean over the nine evaluation users; per-user and per-environment scores are in Appendix D of the paper.
RL starts from the Base checkpoint, Qwen3-8B after a supervised warm start, and trains on five UserRL environments: TravelGym, TurtleGym, FunctionGym, TauGym and PersuadeGym. The CSD loss weight is α = 0.01 and the gate sharpness is β = 5.0. The full configuration is in Appendix D of the paper, and the training scripts are in the agent directory of the code release.
The model loads like Qwen3-8B. To evaluate it in the eight environments, install the agent code, serve a user model as described there, and run from the agent directory:
bash
bash scripts/run_eval.sh phanviethoang1512/Qwen3-8B-CSD-Agent csdbibtex
@article{phan2026mimesis, title = {{MIMESIS}: Learning User Simulators as Training Environments for Interactive Agents}, author = {Phan, Hoang and Huynh, Dat and Zhmoginov, Andrey and Zeng, Qi and Mu, Wancen and Cao, Yue and Bi, Shengjie and He, Yun and Oh, Changdae and Lei, Deren}, journal = {arXiv preprint arXiv:2610.09484}, year = {2026}}@article{phan2026mimesis, title = {{MIMESIS}: Learning User Simulators as Training Environments for Interactive Agents}, author = {Phan, Hoang and Huynh, Dat and Zhmoginov, Andrey and Zeng, Qi and Mu, Wancen and Cao, Yue and Bi, Shengjie and He, Yun and Oh, Changdae and Lei, Deren}, journal = {arXiv preprint arXiv:2610.09484}, year = {2026}}