TeleEmbodied/AssemLM2.0

AssemLM 2.0 Model

Related links: AssemLM 1.0 paper (arXiv) · AssemLM official website · Model repository · Dataset repository

This directory contains the AssemLM 2.0 checkpoint (epoch_0215_steps_2111515) and the matching training/evaluation configuration. The model predicts a 9D pose for the moving part of a two-part assembly from two point clouds, two rendered manual images, and an object-category instruction.

Files

File	Description
config.yaml	Active AssemLM 2.0 model and training configuration
epoch_0215_steps_2111515/pytorch_model/mp_rank_00_model_states.pt	Full model checkpoint of the release

Checkpoint details:

Item	Value
Checkpoint directory	epoch_0215_steps_2111515
Recorded epoch / step	215 / 2,111,515
Checkpoint size	4,482,346,012 bytes (about 4.17 GiB)
Stored format	PyTorch state dictionary
SHA-256	ae2030284436e500ff090ccebb9c6135493c7ddc6f522d3ce03125d3f72aa9fc

Model architecture

Component	Configuration
Framework	AssemLM2
Base VLM	Qwen3-VL-2B-Instruct
Attention implementation	eager
Point encoder	vn_dgcnn_patch
Point feature dimension	512
Patch construction	128 patches, patch size 4, FPS-KNN
Pooling / normalization	mean pooling, RMSNorm
Patch bridge	4 transformer layers, hidden size 1024, 8 heads
Bridge layout	four_branch_equiv_invariant
Point projector	hidden sizes [1024, 2048], output size 2048
Point M-RoPE	grid size 64, joint_isotropic
Point attention	custom causal attention
Pose output	9D: 3D translation + 6D rotation
Pose head	hidden size 4096, 2 residual blocks

Licensing

This checkpoint is released for research use. It was trained on the AssemLM 2.0 data collection, which aggregates several upstream 3D assembly datasets with different terms (IKEA-Manual, PartNet, PartNeXt, BiAssemble, Two by Two). This release does not supersede any upstream license, and users remain responsible for complying with the terms of the original datasets.

Citation

If this project is useful to you, please consider citing:

@article{jing2026assemlm,
  title={AssemLM: A Spatial Reasoning Multimodal Large Language Model for Robotic Assembly},
  author={Jing, Zhi and Qiao, Jinbin and Lu, Ouyang and Ao, Jicong and Qiu, Shuang and Xu, Huazhe and Jiang, Yu-Gang and Bai, Chenjia},
  journal={arXiv preprint arXiv:2604.08983},
  year={2026}
}