Decoupled Action Head: Confining Task Knowledge to Conditioning Layers

• URL: http://arxiv.org/abs/2511.12101v1
• Date: Sat, 15 Nov 2025 08:39:50 GMT
• Title: Decoupled Action Head: Confining Task Knowledge to Conditioning Layers
• Abstract summary: Behavior Cloning (BC) is a data-driven supervised learning approach.<n>We propose a decoupled training recipe that leverages observationfree data to pretrain a kinematics-generated action head.<n>Our experiments demonstrate the feasibility of this approach in both in-distribution and out-of-distribution scenarios.
• Score: 14.665308261922464
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Behavior Cloning (BC) is a data-driven supervised learning approach that has gained increasing attention with the success of scaling laws in language and vision domains. Among its implementations in robotic manipulation, Diffusion Policy (DP), with its two variants DP-CNN (DP-C) and DP-Transformer (DP-T), is one of the most effective and widely adopted models, demonstrating the advantages of predicting continuous action sequences. However, both DP and other BC methods remain constrained by the scarcity of paired training data, and the internal mechanisms underlying DP's effectiveness remain insufficiently understood, leading to limited generalization and a lack of principled design in model development. In this work, we propose a decoupled training recipe that leverages nearly cost-free kinematics-generated trajectories as observation-free data to pretrain a general action head (action generator). The pretrained action head is then frozen and adapted to novel tasks through feature modulation. Our experiments demonstrate the feasibility of this approach in both in-distribution and out-of-distribution scenarios. As an additional benefit, decoupling improves training efficiency; for instance, DP-C achieves up to a 41% speedup. Furthermore, the confinement of task-specific knowledge to the conditioning components under decoupling, combined with the near-identical performance of DP-C in both normal and decoupled training, indicates that the action generation backbone plays a limited role in robotic manipulation. Motivated by this observation, we introduce DP-MLP, which replaces the 244M-parameter U-Net backbone of DP-C with only 4M parameters of simple MLP blocks, achieving a 83.9% faster training speed under normal training and 89.1% under decoupling.

Related papers

• PAC-DP: PAC-Bayesian Diffusion Policy Learning [13.10166905442267]
  Diffusion Policies (DPs) are able to perform complex manipulation tasks.<n>We propose PAC-DP, an approach that increases the performance of DPs in robotic manipulation tasks.
  arXiv  Detail & Related papers  (2026-07-27T11:43:49Z)
• LAFP: Preserving Latent Action Structure in Latent Policy Learning via Flow Matching [15.922360971421654]
  We propose Latent Action Flow Policy (LAFP) for scalable latent policy learning.<n>LAFP consistently outperforms prior methods on downstream imitation learning tasks.<n>LAFP incurs less than 1x additional inference overhead.
  arXiv  Detail & Related papers  (2026-06-09T07:49:49Z)
• Preference-Calibrated Human-in-the-Loop Reinforcement Learning for Robotic Manipulation [40.17737666526493]
  We propose PACT, a Preference-calibrated Actor-Critic Training framework.<n>We first design a progress model that learns from human demonstration and identifies suboptimal segments for credit correction.<n>We then build preference pairs to define a counterfactual advantage that penalizes Bellman targets of the identified suboptimal segment.
  arXiv  Detail & Related papers  (2026-06-02T17:38:25Z)
• SENTINEL: Stagewise Integrity Verification for Pipeline Parallel Decentralized Training [54.8494905524997]
  Decentralized training introduces critical security risks when executed across untrusted, geographically distributed nodes.<n>We propose SENTINEL, a verification mechanism for pipeline parallelism (PP) training without duplication.<n>Experiments demonstrate successful training of up to 4B- parameter LLMs across untrusted distributed environments with up to 176 workers while maintaining model convergence and performance.
  arXiv  Detail & Related papers  (2026-03-03T23:51:10Z)
• Post-Training and Test-Time Scaling of Generative Agent Behavior Models for Interactive Autonomous Driving [3.8612647047433217]
  Group Relative Behavior Optimization improves safety performance by over 40% while preserving behavioral realism.<n>Warm-K is a warm-started Top-K sampling strategy that balances consistency and diversity in motion selection.
  arXiv  Detail & Related papers  (2025-12-15T12:18:50Z)
• Learning to Reason as Action Abstractions with Scalable Mid-Training RL [55.24192942739207]
  An effective mid-training phase should identify a compact set of useful actions and enable fast selection.<n>We propose Reasoning as Action Abstractions (RA3), a scalable mid-training algorithm.
  arXiv  Detail & Related papers  (2025-09-30T05:34:20Z)
• A Scalable Pretraining Framework for Link Prediction with Efficient Adaptation [16.82426251068573]
  Link Prediction (LP) is a critical task in graph machine learning.<n>Existing methods face key challenges including limited supervision from sparse connectivity.<n>We explore pretraining as a solution to address these challenges.
  arXiv  Detail & Related papers  (2025-08-06T17:10:31Z)
• Enhancing One-run Privacy Auditing with Quantile Regression-Based Membership Inference [22.843200081364873]
  Differential privacy (DP) auditing aims to provide empirical lower bounds on the privacy guarantees of DP mechanisms like DP-SGD.<n>Recent work introduces one-run auditing approaches that effectively audit DP-SGD in white-box settings while still being computationally efficient.<n>In this work, we study how incorporating approaches for stronger membership inference attacks (MIA) can improve one-run auditing in the black-box setting.
  arXiv  Detail & Related papers  (2025-06-18T11:03:39Z)
• Denoising Pre-Training and Customized Prompt Learning for Efficient Multi-Behavior Sequential Recommendation [69.60321475454843]
  We propose DPCPL, the first pre-training and prompt-tuning paradigm tailored for Multi-Behavior Sequential Recommendation. In the pre-training stage, we propose a novel Efficient Behavior Miner (EBM) to filter out the noise at multiple time scales. Subsequently, we propose to tune the pre-trained model in a highly efficient manner with the proposed Customized Prompt Learning (CPL) module.
  arXiv  Detail & Related papers  (2024-08-21T06:48:38Z)
• Differentially Private Zeroth-Order Methods for Scalable Large Language Model Finetuning [0.0]
  differentially private (DP) fine-tuning of pretrained LLMs has been widely used to safeguarding the privacy of task-specific datasets.<n>Despite pushing the scalability of DP-SGD to its limit, DP-SGD-based fine-tuning methods are unfortunately limited by the inherent inefficiency of SGD.
  arXiv  Detail & Related papers  (2024-02-12T17:24:15Z)
• Selectivity Drives Productivity: Efficient Dataset Pruning for Enhanced Transfer Learning [66.20311762506702]
  dataset pruning (DP) has emerged as an effective way to improve data efficiency. We propose two new DP methods, label mapping and feature mapping, for supervised and self-supervised pretraining settings. We show that source data classes can be pruned by up to 40% 80% without sacrificing downstream performance.
  arXiv  Detail & Related papers  (2023-10-13T00:07:49Z)
• Tuning Pre-trained Model via Moment Probing [62.445281364055795]
  We propose a novel Moment Probing (MP) method to explore the potential of LP. MP performs a linear classification head based on the mean of final features. Our MP significantly outperforms LP and is competitive with counterparts at less training cost.
  arXiv  Detail & Related papers  (2023-07-21T04:15:02Z)
• Large Language Models Can Be Strong Differentially Private Learners [70.0317718115406]
  Differentially Private (DP) learning has seen limited success for building large deep learning models of text. We show that this performance drop can be mitigated with the use of large pretrained models. We propose a memory saving technique that allows clipping in DP-SGD to run without instantiating per-example gradients.
  arXiv  Detail & Related papers  (2021-10-12T01:45:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.