arXiv AI By Suhwan Choi, Jaeyoon Jung, Sungkyung Kim, Yunsung Lee, Youngjae Yu

PonderPounce: A Pretrained MLLM as an Episode Context Engine for Robot Control

Read the original on arXiv AI →

PonderPounce introduces a two‑stage system that leverages a multimodal large language model (MLLM) as an episode‑context engine for robot control. The System2 component, Ponder, accumulates observations, demonstrations, and prior cognition in its native causal context, producing subgoal text and reasoning. The System1 component, Pounce, uses the current observation, instruction, and proprioception, receiving only the newest cognition token and its age from Ponder; the pair is jointly trained end‑to‑end without a separate memory module, achieving real‑time action playback and outperforming baseline methods on RoboMME and RoboCasa‑DC benchmarks.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI
3d ago

MemBodied: Recurrent Associative Memory for Vision-Language-Action Models

MemBodied introduces a fixed‑size episodic memory for Vision‑Language‑Action models, comprising an associative state that tracks interactions across policy calls and an episode anchor that stores a compact representation of the initial scene. By conditioning action generation on these memory components instead of raw past observations, MemBodied reduces context bloat and inference latency. In five memory‑dependent RMBench tasks, it outperforms stateless and vanilla recurrent policies by significant margins, and achieves a 90.6% success rate on the LIBERO‑Long suite, improving over the baseline by 5.4%.

By Tej Deep Pala, Navonil Majumder, Bryce Goh, Raphael Yee, Jianfei Yang, Liming Chen, Soujanya Poria
arXiv AI
Aug 25

Act with Intent: Distilling Behavior Intent for Vision-Language-Action Models

The paper introduces Intention Distillation (INDI), a method that injects behavior-level intent into Vision‑Language‑Action (VLA) model decoders by leveraging a frozen teacher vision‑language model to interpret demonstrations. During training, the teacher processes the current observation, instruction, coarse action summary, and execution video, producing a multimodal intent representation that the VLA decoder uses alongside trajectory and execution features to predict actions. Experiments on SimplerEnv‑Bridge, RoboCasa Kitchen, and real‑world tasks show that INDI consistently improves success rates, especially on longer‑horizon tasks, demonstrating that explicit modeling of semantic intent benefits action decoders.

By Sangoh Lee, Sangwoo Mo, Wook-Shin Han
arXiv Machine Learning
Sep 10

LM-X: Explainable Vision--Language--Action Modeling via Progress, Event, and Uncertainty Prediction

arXiv:2608.25757v4 Announce Type: replace-cross Abstract: Large-scale vision--language--action (VLA) policies have advanced generalist robot control, yet most remain stimulus-to-action black boxes: a...

By Jin Lou, Zhiyuan Jing, Xupeng Wang, Andong Chen, Xingdong Zhu, Yuexuan Li, Yuan Xu, Zhijie Zhu, Yingwei Ji, Wenpeng Nie, Renxing Feng, Liangliang Chen, Ying Chu, Jingyi Li, Jinyan Liu, Zhiqi Song, Jingxuan Zhu, Jidong Zhang, Yufei Liu, Boyang Xing, Lei Jiang, Yan Cui, Hongming Li, Yuchen Zhu
arXiv AI
Jul 29

CoTinyVLA: Chain-of-Thought Distillation for a Sub-Billion-Parameter Vision-Language-Action Model

arXiv:2607. 25487v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models translate natural-language commands into robot action sequences, but leading systems on the LIBERO-Plus robustness benchmark use three- to seven-billion-parameter backbones whose memory demands can exceed embedded robotic budgets.

By Minhyeok Lee, Chiyoung Kim, Chanhoe Gu, Seongrok Kim, Sanghyuk Roy Choi, Donghwan Hwang, Donghun Ryu, Seokhyun Kim
Hugging Face Trending Papers
Jul 29

TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM

Vision-language-action (VLA) models commonly adopt an LLM-centric $V \to L \to A$ pathway, where visual observations are projected into the representation space of a large language model before being decoded into robot actions. Although effective, this design incurs substantial computation and memory overhead at every policy invocation.