# Action QFormer: Structured Representation Shaping under Action Supervision in Vision-Language-Action Models

arXiv: https://arxiv.org/abs/2607.14635 · HTML: https://mlsanity.com/paper/2607.14635/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in VLA architectures and generalization, one of 13 directions in Vision-language-action models.
- Builds on 32 of its 34 references that are in the map.
- First posted on arXiv 16 Jul 2026.

## Placed in

- Vision-language-action models / [VLA architectures and generalization](https://mlsanity.com/vla/vla-architectures-and-generalization/)

## Builds on (most cited first, 30 of 32)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [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)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [RT-1: Robotics Transformer for Real-World Control at Scale](https://mlsanity.com/paper/2212.06817/) (2022)
- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [Align before Fuse: Vision and Language Representation Learning with Momentum Distillation](https://mlsanity.com/paper/2107.07651/) (2021)
- [Habitat: A Platform for Embodied AI Research](https://mlsanity.com/paper/1904.01201/) (2019)
- [Perceiver: General Perception with Iterative Attention](https://mlsanity.com/paper/2103.03206/) (2021)
- [FAST: Efficient Action Tokenization for Vision-Language-Action Models](https://mlsanity.com/paper/2501.09747/) (2025)
- [LM-Nav: Robotic Navigation with Large Pre-Trained Models of Language, Vision, and Action](https://mlsanity.com/paper/2207.04429/) (2022)
- [UniVLA: Learning to Act Anywhere with Task-centric Latent Actions](https://mlsanity.com/paper/2505.06111/) (2025)
- [Latent Action Pretraining from Videos](https://mlsanity.com/paper/2410.11758/) (2024)
- [Navigation World Models](https://mlsanity.com/paper/2412.03572/) (2024)
- [NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration](https://mlsanity.com/paper/2310.07896/) (2023)
- [ObjectNav Revisited: On Evaluation of Embodied Agents Navigating to Objects](https://mlsanity.com/paper/2006.13171/) (2020)
- [NaVid: Video-based VLM Plans the Next Step for Vision-and-Language Navigation](https://mlsanity.com/paper/2402.15852/) (2024)
- [ViNT: A Foundation Model for Visual Navigation](https://mlsanity.com/paper/2306.14846/) (2023)
- [GNM: A General Navigation Model to Drive Any Robot](https://mlsanity.com/paper/2210.03370/) (2022)
- [VQ-VLA: Improving Vision-Language-Action Models via Scaling Vector-Quantized Action Tokenizers](https://mlsanity.com/paper/2507.01016/) (2025)
- [ViKiNG: Vision-Based Kilometer-Scale Navigation with Geographic Hints](https://mlsanity.com/paper/2202.11271/) (2022)
- [Making Foresight Actionable: Repurposing Representation Alignment in World Action Models](https://mlsanity.com/paper/2606.12217/) (2026)
- [RotVLA: Rotational Latent Action for Vision-Language-Action Model](https://mlsanity.com/paper/2605.13403/) (2026)
- [DynaFLIP: Rethinking Robotics Perception via Tri-Modal-Dynamics Guided Representation](https://mlsanity.com/paper/2605.30350/) (2026)
- [HARP-VLA: Human-Robot Aligned Representation Learning for Vision-Language-Action Model](https://mlsanity.com/paper/2605.31234/) (2026)

## Built on by (most cited first, 1 of 1)

- [SCULPT-VLA: Learning Structured Control through Staged Action Grounding](https://mlsanity.com/paper/2609.23275/) (2026)

## Abstract (arXiv)

Action supervision in vision-language-action (VLA) models is often treated as a downstream objective for learning action prediction. In this paper, we study it instead as a force that shapes inherited multimodal representations. We show that this shaping has a dual effect: it is necessary for forming action-compatible representations, but when action supervision is applied too directly to the inherited multimodal pathway, it can also destabilize representations that support language-side processing and object grounding. To address this tension, we introduce Action QFormer, a query-based action-facing interface that uses instruction-conditioned queries to reorganize inherited multimodal information into action-facing representations before downstream action generation. In zero-shot sim-to-real navigation, Action QFormer improves average closed-loop task success from 18.8% to 56.3%, raises fixed-instruction action-generation correctness from 22.5% to 75.5%, and nearly eliminates out-of-distribution instruction generations. Further analyses show that Action QFormer changes how action supervision shapes inherited multimodal representations, reducing broad upstream rewriting while preserving targeted and sometimes constructive action-supervised adaptation. These results suggest that improving VLA performance requires not only stronger pretrained backbones, but also better ways of selecting and organizing inherited multimodal information while controlling how it is shaped under action supervision.