CodeDiffuser: Attention-Enhanced Diffusion Policy via VLM-Generated Code for Instruction Ambiguity

• URL: http://arxiv.org/abs/2506.16652v1
• Date: Thu, 19 Jun 2025 23:42:03 GMT
• Title: CodeDiffuser: Attention-Enhanced Diffusion Policy via VLM-Generated Code for Instruction Ambiguity
• Authors: Guang Yin, Yitong Li, Yixuan Wang, Dale McConachie, Paarth Shah, Kunimatsu Hashimoto, Huan Zhang, Katherine Liu, Yunzhu Li,
• Abstract summary: We introduce a novel robotic manipulation framework that can accomplish tasks specified by potentially ambiguous natural language.<n>This framework employs a Vision-Language Model (VLM) to interpret abstract concepts in natural language instructions.<n>We show that our approach excels across challenging manipulation tasks involving language ambiguity, contact-rich manipulation, and multi-object interactions.
• Score: 23.77040677368575
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Natural language instructions for robotic manipulation tasks often exhibit ambiguity and vagueness. For instance, the instruction "Hang a mug on the mug tree" may involve multiple valid actions if there are several mugs and branches to choose from. Existing language-conditioned policies typically rely on end-to-end models that jointly handle high-level semantic understanding and low-level action generation, which can result in suboptimal performance due to their lack of modularity and interpretability. To address these challenges, we introduce a novel robotic manipulation framework that can accomplish tasks specified by potentially ambiguous natural language. This framework employs a Vision-Language Model (VLM) to interpret abstract concepts in natural language instructions and generates task-specific code - an interpretable and executable intermediate representation. The generated code interfaces with the perception module to produce 3D attention maps that highlight task-relevant regions by integrating spatial and semantic information, effectively resolving ambiguities in instructions. Through extensive experiments, we identify key limitations of current imitation learning methods, such as poor adaptation to language and environmental variations. We show that our approach excels across challenging manipulation tasks involving language ambiguity, contact-rich manipulation, and multi-object interactions.

Related papers

• Improving Generalization of Language-Conditioned Robot Manipulation [29.405161073483175]
  We present a framework that learns object-arrangement tasks from just a few demonstrations.<n>We validate our method on both simulation and real-world robotic environments.
  arXiv  Detail & Related papers  (2025-08-04T13:29:26Z)
• SimStep: Chain-of-Abstractions for Incremental Specification and Debugging of AI-Generated Interactive Simulations [16.00479720281197]
  Chain-of-Abstractions (CoA) is a way to recover programming's core affordances.<n>CoA decomposes the synthesis process into a sequence of cognitively meaningful, task-aligned representations.<n>SimStep is an authoring environment for teachers that scaffolds simulation creation through four intermediate abstractions.
  arXiv  Detail & Related papers  (2025-07-13T14:54:17Z)
• Interpretable Robotic Manipulation from Language [11.207620790833271]
  We introduce an explainable behavior cloning agent, named Ex-PERACT, specifically designed for manipulation tasks. At the top level, the model is tasked with learning a discrete skill code, while at the bottom level, the policy network translates the problem into a voxelized grid and maps the discretized actions to voxel grids. We evaluate our method across eight challenging manipulation tasks utilizing the RLBench benchmark, demonstrating that Ex-PERACT not only achieves competitive policy performance but also effectively bridges the gap between human instructions and machine execution in complex environments.
  arXiv  Detail & Related papers  (2024-05-27T11:02:21Z)
• Linguacodus: A Synergistic Framework for Transformative Code Generation in Machine Learning Pipelines [0.0]
  We introduce a dynamic pipeline that transforms natural language task descriptions into code through high-level data-shaping instructions. This paper details the fine-tuning process, and sheds light on how natural language descriptions can be translated into functional code. We propose an algorithm capable of transforming a natural description of an ML task into code with minimal human interaction.
  arXiv  Detail & Related papers  (2024-03-18T08:58:47Z)
• Learning with Language-Guided State Abstractions [58.199148890064826]
  Generalizable policy learning in high-dimensional observation spaces is facilitated by well-designed state representations. Our method, LGA, uses a combination of natural language supervision and background knowledge from language models to automatically build state representations tailored to unseen tasks. Experiments on simulated robotic tasks show that LGA yields state abstractions similar to those designed by humans, but in a fraction of the time.
  arXiv  Detail & Related papers  (2024-02-28T23:57:04Z)
• SAGE: Bridging Semantic and Actionable Parts for GEneralizable Manipulation of Articulated Objects [9.500480417077272]
  We propose a novel framework that bridges semantic and actionable parts of articulated objects to achieve generalizable manipulation under natural language instructions. A part-grounding module maps the semantic parts into so-called Generalizable Actionable Parts (GAParts), which inherently carry information about part motion. An interactive feedback module is incorporated to respond to failures, which closes the loop and increases the robustness of the overall framework.
  arXiv  Detail & Related papers  (2023-12-03T07:22:42Z)
• InstructSeq: Unifying Vision Tasks with Instruction-conditioned Multi-modal Sequence Generation [59.24938416319019]
  InstructSeq is an instruction-conditioned multi-modal modeling framework. It unifies diverse vision tasks through flexible natural language control and handling of both visual and textual data.
  arXiv  Detail & Related papers  (2023-11-30T18:59:51Z)
• PADL: Language-Directed Physics-Based Character Control [66.517142635815]
  We present PADL, which allows users to issue natural language commands for specifying high-level tasks and low-level skills that a character should perform. We show that our framework can be applied to effectively direct a simulated humanoid character to perform a diverse array of complex motor skills.
  arXiv  Detail & Related papers  (2023-01-31T18:59:22Z)
• LISA: Learning Interpretable Skill Abstractions from Language [85.20587800593293]
  We propose a hierarchical imitation learning framework that can learn diverse, interpretable skills from language-conditioned demonstrations. Our method demonstrates a more natural way to condition on language in sequential decision-making problems.
  arXiv  Detail & Related papers  (2022-02-28T19:43:24Z)
• Neural Abstructions: Abstractions that Support Construction for Grounded Language Learning [69.1137074774244]
  Leveraging language interactions effectively requires addressing limitations in the two most common approaches to language grounding. We introduce the idea of neural abstructions: a set of constraints on the inference procedure of a label-conditioned generative model. We show that with this method a user population is able to build a semantic modification for an open-ended house task in Minecraft.
  arXiv  Detail & Related papers  (2021-07-20T07:01:15Z)
• A Persistent Spatial Semantic Representation for High-level Natural Language Instruction Execution [54.385344986265714]
  We propose a persistent spatial semantic representation method to bridge the gap between language and robot actions. We evaluate our approach on the ALFRED benchmark and achieve state-of-the-art results, despite completely avoiding the commonly used step-by-step instructions.
  arXiv  Detail & Related papers  (2021-07-12T17:47:19Z)

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.