GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
DawnLiu
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
DawnLiu
Model Tree
Qwen/Qwen3-VL-8B-Instruct
Input Modalities
Output Modalities
Supported Functionality
RLVR rewards typically check only the final answer, not whether it came from the image or from language priors and dataset shortcuts. EVAN masks the visual evidence a question needs, lets the current policy answer without it, and penalizes plausible wrong answers under the clean image. EVAN-8B is trained with DAPO on ViRL39K. The GitHub repository has the full method, data pipeline, and training scripts.
Accuracy (%) from Table 1 of the paper, with sampling at temperature 1.0. DAPO uses the same data and training recipe without EVAN.
| Model | Geo3K | MathVista | MathVerse | We-Math | DynaMath | MathVision | MMK12 | MMMU-Pro | LogicVista | Avg. |
|---|---|---|---|---|---|---|---|---|---|---|
| Qwen3-VL-8B-Instruct | 65.2 | 68.7 | 69.4 | 77.2 | 73.4 | 38.0 | 69.1 | 40.3 | 61.3 | 62.5 |
| DAPO | 74.6 | 71.9 | 84.0 | 83.9 | 76.9 | 56.7 | 84.1 | 52.3 | 62.4 | 71.9 |
| EVAN-8B | 76.6 | 74.9 | 85.0 | 84.9 | 77.6 | 57.0 | 85.0 | 52.0 | 63.6 | 73.0 |
Requires transformers>=4.57, accelerate, and torchvision. Add the instruction below after the question, as in training. The default generation config holds the sampling settings we used for evaluation.
python
from transformers import AutoModelForImageTextToText, AutoProcessor model_id = "DawnLiu/EVAN-8B"model = AutoModelForImageTextToText.from_pretrained(model_id, dtype="auto", device_map="auto")processor = AutoProcessor.from_pretrained(model_id) question = "Which object is the largest?"prompt = ( f"{question}\n\n" "You first think through the reasoning process as an internal monologue, enclosed within <think> </think> tags. Then, provide your final answer enclosed within \\boxed{}.")messages = [ { "role": "user", "content": [ {"type": "image", "image": "path/to/image.png"}, {"type": "text", "text": prompt}, ], }]inputs = processor.apply_chat_template( messages, tokenize=True, add_generation_prompt=True, return_dict=True, return_tensors="pt").to(model.device)output = model.generate(**inputs)print(processor.decode(output[0, inputs["input_ids"].shape[1]:], skip_special_tokens=True))bibtex
@misc{liu2026grounding, title = {Grounding Multimodal Reasoning with Evidence-Ablated Negatives}, author = {Chenxi Liu and Tianyi Xiong and Yanshuo Chen and Heng Huang}, year = {2026}, eprint = {2610.39237}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2610.39237},}Apache 2.0, the same as Qwen3-VL-8B-Instruct. Relative to the base model, the weights were changed by RL training with EVAN, generation_config.json sets our evaluation sampling defaults, and the other configuration, tokenizer, and processor files were re-saved (with dtype set to bfloat16 in config.json).
from transformers import AutoModelForImageTextToText, AutoProcessor model_id = "DawnLiu/EVAN-8B"model = AutoModelForImageTextToText.from_pretrained(model_id, dtype="auto", device_map="auto")processor = AutoProcessor.from_pretrained(model_id) question = "Which object is the largest?"prompt = ( f"{question}\n\n" "You first think through the reasoning process as an internal monologue, enclosed within <think> </think> tags. Then, provide your final answer enclosed within \\boxed{}.")messages = [ { "role": "user", "content": [ {"type": "image", "image": "path/to/image.png"}, {"type": "text", "text": prompt}, ], }]inputs = processor.apply_chat_template( messages, tokenize=True, add_generation_prompt=True, return_dict=True, return_tensors="pt").to(model.device)output = model.generate(**inputs)print(processor.decode(output[0, inputs["input_ids"].shape[1]:], skip_special_tokens=True))@misc{liu2026grounding, title = {Grounding Multimodal Reasoning with Evidence-Ablated Negatives}, author = {Chenxi Liu and Tianyi Xiong and Yanshuo Chen and Heng Huang}, year = {2026}, eprint = {2610.39237}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2610.39237},}