# SeqVLA: Sequential Task Execution for Long-Horizon Manipulation with Completion-Aware Vision-Language-Action Model arXiv: https://arxiv.org/abs/2509.14138 · HTML: https://mlsanity.com/paper/2509.14138/ ## Key facts - Cited by 10 papers in the map (data through 14 Oct 2026). - Placed in Reasoning and chain-of-thought in VLA, one of 13 directions in Vision-language-action models. - Builds on 20 of its 23 references that are in the map. - First posted on arXiv 17 Sep 2025. ## Placed in - Vision-language-action models / [Reasoning and chain-of-thought in VLA](https://mlsanity.com/vla/reasoning-and-chain-of-thought-in-vla/) ## Builds on (most cited first, 20 of 20) - [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022) - [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023) - [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023) - [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024) - [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024) - [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022) - [RT-1: Robotics Transformer for Real-World Control at Scale](https://mlsanity.com/paper/2212.06817/) (2022) - [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023) - [SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics](https://mlsanity.com/paper/2506.01844/) (2025) - [CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models](https://mlsanity.com/paper/2503.22020/) (2025) - [VIMA: General Robot Manipulation with Multimodal Prompts](https://mlsanity.com/paper/2210.03094/) (2022) - [Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation](https://mlsanity.com/paper/2401.02117/) (2024) - [Cross-Entropy Loss Functions: Theoretical Analysis and Applications](https://mlsanity.com/paper/2304.07288/) (2023) - [A Review of Robot Learning for Manipulation: Challenges, Representations, and Algorithms](https://mlsanity.com/paper/1907.03146/) (2019) - [AHA: A Vision-Language-Model for Detecting and Reasoning Over Failures in Robotic Manipulation](https://mlsanity.com/paper/2410.00371/) (2024) - [SAFE: Multitask Failure Detection for Vision-Language-Action Models](https://mlsanity.com/paper/2506.09937/) (2025) - [GOPlan: Goal-conditioned Offline Reinforcement Learning by Planning with Learned Models](https://mlsanity.com/paper/2310.20025/) (2023) - [Exploring the limits of Hierarchical World Models in Reinforcement Learning](https://mlsanity.com/paper/2406.00483/) (2024) - [Value-Informed Skill Chaining for Policy Learning of Long-Horizon Tasks with Surgical Robot](https://mlsanity.com/paper/2307.16503/) (2023) - [Hierarchical Task Decomposition for Execution Monitoring and Error Recovery: Understanding the Rationale Behind Task Demonstrations](https://mlsanity.com/paper/2505.04565/) (2025) ## Built on by (most cited first, 10 of 10) - [CycleVLA: Proactive Self-Correcting Vision-Language-Action Models via Subtask Backtracking and Minimum Bayes Risk Decoding](https://mlsanity.com/paper/2601.02295/) (2026) - [See, Plan, Rewind: Progress-Aware Vision-Language-Action Models for Robust Robotic Manipulation](https://mlsanity.com/paper/2603.09292/) (2026) - [WeaveLA: Event Driven Cross-Subtask Latent Memory Weaving for Repetitive Robot Manipulation](https://mlsanity.com/paper/2606.17463/) (2026) - [Learning Robust Execution in Robotic Manipulation with Agentic Reinforcement Learning](https://mlsanity.com/paper/2607.13818/) (2026) - [SADP: Subgoal-Aware Diffusion Policy for Long-Horizon Manipulation Learned from Foundation Model Generated Demonstrations](https://mlsanity.com/paper/2605.16871/) (2026) - [FurnitureVLA: Learning Long-Horizon Bimanual Furniture Assembly with Vision-Language-Action Model](https://mlsanity.com/paper/2607.01212/) (2026) - [VILAS: A VLA-Integrated Low-cost Architecture with Soft Grasping for Robotic Manipulation](https://mlsanity.com/paper/2605.02037/) (2026) - [Threading Optimization for Vision-Language-Action Model Inference in Low-Cost Smart Agricultural Manipulation](https://mlsanity.com/paper/2606.00966/) (2026) - [SEES: A Self-Evolving Embodied System via Failure-Guided VLA Policy Adaptation](https://mlsanity.com/paper/2609.32698/) (2026) - [CogWAM: Aligning Semantic Cognition with World Action Modeling via Event-Driven Interfaces](https://mlsanity.com/paper/2609.37721/) (2026) ## Abstract (arXiv) Long-horizon robotic manipulation tasks require executing multiple interdependent subtasks in strict sequence, where errors in detecting subtask completion can cascade into downstream failures. Existing Vision-Language-Action (VLA) models such as π0 excel at continuous low-level control but lack an internal signal for identifying when a subtask has finished, making them brittle in sequential settings. We propose SeqVLA, a completion-aware extension of π0 that augments the base architecture with a lightweight detection head perceiving whether the current subtask is complete. This dual-head design enables SeqVLA not only to generate manipulation actions but also to autonomously trigger transitions between subtasks. We investigate four finetuning strategies that vary in how the action and detection heads are optimized (joint vs. sequential finetuning) and how pretrained knowledge is preserved (full finetuning vs. frozen backbone). Experiments are performed on two multi-stage tasks: salad packing with seven distinct subtasks and candy packing with four distinct subtasks. Results show that SeqVLA significantly outperforms the baseline π0 and other strong baselines in overall success rate. In particular, joint finetuning with an unfrozen backbone yields the most decisive and statistically reliable completion predictions, eliminating sequence-related failures and enabling robust long-horizon execution. Our results highlight the importance of coupling action generation with subtask-aware detection for scalable sequential manipulation.