PAC-ACT: Post-training Actor-Critic for Action Chunking Transformers
Precision industrial contact manipulation requires reliable robot policies under pose perturbations and contact-force constraints. Vision-language-action models offer broad generalization but often introduce high inference latency and GPU-memory cost, while vision-action chunking policies are more suitable for real-time industrial control. However, these policies are usually trained by behavior cloning and suffer from distribution shift in contact-rich tasks. This paper proposes PAC-ACT, a reinforcement-learning post-training framework for pretrained Action Chunking Transformer policies. PAC-A
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 53%AgentCore-8B →
- LinkedLinked via arxiv author · 85%Yujie Pang →
“PAC-ACT: Post-training Actor-Critic for Action Chunking Transformers”
- LinkedLinked via arxiv author · 85%Zudong Li →
“PAC-ACT: Post-training Actor-Critic for Action Chunking Transformers”
- FuzzySimilar title/name (fuzzy) · 87%lucidrains/x-transformers →
“Fuzzy title match (0.94): “PAC-ACT: Post-training Actor-Critic for Action Chunking Tran” ≈ “lucidrains/x-transformers””
- FuzzySimilar title/name (fuzzy) · 84%huggingface/transformers →
“Fuzzy title match (0.92): “PAC-ACT: Post-training Actor-Critic for Action Chunking Tran” ≈ “huggingface/transformers””
- FuzzySimilar title/name (fuzzy) · 84%liguodongiot/llm-action →
“Fuzzy title match (0.92): “PAC-ACT: Post-training Actor-Critic for Action Chunking Tran” ≈ “liguodongiot/llm-action””