MLR Executor for Qwen2.5-1.5B

This repository contains the low-level executor for Multi-Level Reasoning (MLR) in the paper Enhancing Language Model Reasoning with Structured Multi-Level Modeling (ICLR 26).

MLR decomposes long-horizon reasoning into an alternating plan--execute loop: the planner proposes a structured, abstract subgoal and the executor produces the detailed reasoning conditioned on it.

Model details

• Base architecture: Qwen2.5-1.5B

Loading the checkpoint

Install compatible versions of PyTorch and Transformers, then load the checkpoint explicitly:

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "sxiong/MLR_executor_Qwen-1.5B"

tokenizer = AutoTokenizer.from_pretrained(model_id)
executor = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)
executor.eval()

For end-to-end MLR inference, use the accompanying MLR inference code. It applies the planner and executor with the required prompts, parser, stopping rule, and alternating plan--execute control flow. Direct free-form generation from this adapter is not the intended interface.

Citation

@inproceedings{xiong2026enhancing,
  title={Enhancing language model reasoning with structured multi-level modeling},
  author={Xiong, Siheng and Payani, Ali and Fekri, Faramarz},
  booktitle={International Conference on Learning Representations},
  volume={2026},
  pages={36557--36610},
  year={2026}
}