Neural Solver Synthesis
Collection
Models, datasets, certified evaluations, and final evidence. Interactive reports: https://wandb.ai/neural-solver-synthesis/neural-solver-synthesis • 55 items • Updated
How to use IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-Prompt-seed101 with Transformers:
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-Prompt-seed101")
model = AutoModelForCausalLM.from_pretrained("IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-Prompt-seed101", device_map="auto")Prompt ablation (seed 101).
This model is part of the Neural Solver Synthesis project (ICML 2026). It acts as a compiled solver for the SDS problem.
Base model
Qwen/Qwen2.5-14B