paper-with-me

Papers

RAP: Retrieval-Augmented Planner for Adaptive Procedure Planning in Instructional Videos

2024-03-27 · Ali Zare, Yulei Niu, Hammad Ayyubi, Shih-Fu Chang

Procedure Planning in instructional videos entails generating a sequence of action steps based on visual observations of the initial and target states. Despite the rapid progress in this task, there remain several critical challenges to be solved: (1) Adaptive procedures: Prior works hold an unrealistic assumption that the number of action steps is known and fixed, leading to non-generalizable models in real-world scenarios where the sequence length varies. (2) Temporal relation: Understanding the step temporal relation knowledge is essential in producing reasonable and executable plans. (3) Annotation cost: Annotating instructional videos with step-level labels (i.e., timestamp) or sequence-level labels (i.e., action category) is demanding and labor-intensive, limiting its generalizability to large-scale datasets. In this work, we propose a new and practical setting, called adaptive procedure planning in instructional videos, where the procedure length is not fixed or pre-determined. To address these challenges, we introduce Retrieval-Augmented Planner (RAP) model. Specifically, for adaptive procedures, RAP adaptively determines the conclusion of actions using an auto-regressive model architecture. For temporal relation, RAP establishes an external memory module to explicitly retrieve the most relevant state-action pairs from the training videos and revises the generated procedures. To tackle high annotation cost, RAP utilizes a weakly-supervised learning manner to expand the training dataset to other task-relevant, unannotated videos by generating pseudo labels for action steps. Experiments on CrossTask and COIN benchmarks show the superiority of RAP over traditional fixed-length models, establishing it as a strong baseline solution for adaptive procedure planning.

📄 PDF Abstract BibTeX arXiv:2403.18600

Code (0)

등록된 구현이 없습니다.

Tasks

RelationRetrievalWeakly-supervised Learning

Similar Papers 제목 키워드 기반

Unveiling the Potential of Multimodal Retrieval Augmented Generation with Planning

2025-01-26 · Xiaohan Yu, Zhihan Yang, Chong Chen

Multimodal Retrieval Augmented Generation (MRAG) systems, while promising for enhancing Multimodal Large Language Models (MLLMs), often rely on rigid, single-step retrieval methods. This limitation hinders their ability …

RetrievalRetrieval-augmented Generation

Distilling Reasoning Without Knowledge: A Framework for Reliable LLMs

2026-03-15 · Auksarapak Kietkajornrit, Jad Tarifi, Nima Asgharbeygi arxiv

Fact-seeking question answering with large language models (LLMs) remains unreliable when answers depend on up-to-date or conflicting information. Although retrieval-augmented and tool-using LLMs reduce hallucinations, t…

Question Answering

RealDrive: Retrieval-Augmented Driving with Diffusion Models

2025-05-30 · Wenhao Ding, Sushant Veer, Yuxiao Chen, Yulong Cao 외

Learning-based planners generate natural human-like driving behaviors by learning to reason about nuanced interactions from data, overcoming the rigid behaviors that arise from rule-based planners. Nonetheless, data-driv…

DenoisingRAGRetrievalRetrieval-augmented Generation

Can We Further Elicit Reasoning in LLMs? Critic-Guided Planning with Retrieval-Augmentation for Solving Challenging Tasks

2024-10-02 · Xingxuan Li, Weiwen Xu, Ruochen Zhao, Fangkai Jiao 외

State-of-the-art large language models (LLMs) exhibit impressive problem-solving capabilities but may struggle with complex reasoning and factual correctness. Existing methods harness the strengths of chain-of-thought an…

MathNavigateRAGRetrieval+1

Adaptive Dynamics Planning for Robot Navigation

2025-10-06 · Yuanjie Lu, Mingyang Mao, Tong Xu, Linji Wang 외 arxiv

Autonomous robot navigation systems often rely on hierarchical planning, where global planners compute collision-free paths without considering dynamics, and local planners enforce dynamics constraints to produce executa…

Reinforcement LearningRobot Navigation