Tag
提出 TD-Aug 方法，通过从几何 horizon 模型（基于 TD-Flows 训练）采样来想象新的目标状态，作为目标条件行为克隆的数据增强，显著提升了策略在 OGBench 拼接任务中的零样本组合泛化能力。
提出了 Keyframe Mnemonics，一种自监督关键帧发现方法，用于解决非马尔可夫环境下的行为 cloning 长时程依赖问题，在合成记忆域达到 100% 成功率，并在机器人操作基准上比最强基线平均绝对提升 13.9%。
A new paper, 'Split the Differences, Pool the Rest: Provably Efficient Multi-Objective Imitation Learning', introduces MA-BC, which pools expert demonstrations only where observed actions do not disagree, and provides upper and lower bounds on sample complexity for learning from experts with differing objectives.
This paper evaluates Disagreement-Regularized Imitation Learning (DRIL) for image-based continuous control in CarRacing, comparing Gaussian and Beta learner policies. Score-selected DRIL shows the largest gains in few-demonstration settings, while Beta behavior cloning remains competitive with more demonstrations.
This paper presents a reinforcement learning agent trained to generate provably optimal quadrilateral block decompositions for planar domains, demonstrating superior performance over existing tools like Gmsh in terms of usability and optimality.
Proposes decoupling policy extraction from actor training in offline reinforcement learning to address issues like out-of-distribution action amplification and support-value trade-offs, showing improved performance over coupled methods.
This paper proposes Intention Distillation (INDI) to distill behavior intent into the action decoder of Vision-Language-Action models, improving performance on benchmarks like SimplerEnv-Bridge and real-world tasks.
This paper proposes 'Mirror Learning', a framework for imitation learning from third-person observation that uses a fine-tuned video diffusion model for perspective transformation and an inverse dynamics model to synthesize pseudo first-person expert data, showing that this mirror data alone can train effective policies and improve behavior cloning.
DeepEpoch is a no-code desktop app that learns to play 2D games by watching users play via behavior cloning, with human-in-the-loop fine-tuning.
Anchor-Align augments behavioral cloning with vision-language anchoring to preserve pretrained representations and language-action alignment, improving real-robot success rates by over 20% on xArm7 and showing consistent gains in simulation benchmarks.
Task-Agnostic Pretraining (TAP) decomposes VLA training into self-supervised motor skill learning from unlabeled interaction data, then lightweight language grounding, achieving strong performance with minimal expert demonstrations. It matches or outperforms models trained on millions of expert trajectories while being robust to real-world perturbations.
The paper proposes Skill-Guided Continuation Distillation (SGCD), an iterative self-improvement framework that uses skill-guided policies to generate supervision for off-trajectory states during closed-loop execution, improving GUI agent success rates on OSWorld-Verified from around 30% to over 50%.
This paper investigates whether per-phase metric selection improves demonstration curation for behavior cloning policies. The authors find that phase-gated curation never outperforms global or uniform metric application, and the dilution of defect signals across phases explains the failure.
This paper introduces Parameterized Diffusion Policy (PDP), a framework that makes diffusion policies controllable by conditioning on low-dimensional latent parameters, enabling smooth behavior interpolation and adaptation without retraining. It demonstrates improved performance on complex multimodal robot tasks in simulation and real-world experiments.
This paper introduces PRO-CUA, a process-reward optimization framework for training Computer Use Agents (CUAs) using iterative step-level reinforcement learning. The method decouples on-policy environment interaction from policy optimization, enabling dense credit assignment without relying on expert trajectories, and demonstrates effectiveness on live web benchmarks.
This paper introduces the Frequency Guidance Operator (FGO), a method for diffusion policies that smooths action generation by steering noisy samples through intermediate sub-frequency manifolds, improving performance on robotic manipulation tasks.
Leaked audio reveals Meta is using engineers' work traces to train coding AI through behavior cloning, while cutting thousands of jobs.