The paper introduces an adaptive stiffness control framework for physical human‑robot collaboration that uses generative action‑chunk sampling conditioned on RGB images and joint‑torque estimates. By sampling multiple future action chunks, the system measures variation among them to adjust joint stiffness and damping in real time—higher variation leads to more compliance, while lower variation yields firmer assistance. In a collaborative transport experiment with four possible directions, the method achieved a 0.95 success rate, outperforming a fixed‑stiffness baseline (0.83) and a deterministic baseline (0.69).
By Aoi Otake, Ferdinand Hartmann, Ko Igari, Shingo Murata
Latent-CoT-Drive (LCDrive) is a vision‑language‑action model for autonomous driving that replaces natural‑language chain‑of‑thought reasoning with a latent language capturing possible outcomes of driving actions. The model interleaves action‑proposal tokens, aligned with the model’s output actions, and world‑model tokens grounded in a learned latent world model to reason about future outcomes. After a supervised cold‑start using ground‑truth future rollouts, LCDrive is further refined with closed‑loop reinforcement learning, achieving faster inference, higher‑quality trajectories, and greater gains from interactive RL than both non‑reasoning and text‑reasoning baselines on a large‑scale end‑to‑end driving benchmark.
By Shuhan Tan, Kashyap Chitta, Yuxiao Chen, Ran Tian, Yurong You, Yan Wang, Wenjie Luo, Yulong Cao, Philipp Krahenbuhl, Marco Pavone, Boris Ivanovic
TDFNet introduces a Tri-projection Deformable Fusion Network that uses equirectangular, cube map, and tangent projections to mitigate geometric distortions in panoramic salient object detection. It incorporates a cross-projection deformable attention module for geometry-aware sampling and a latitude-guided fusion module that balances ERP and CMP features using spherical latitude priors. The network’s three-branch encoding preserves global continuity, local detail, and boundary precision, improving detection performance over existing projection-based methods.
By Qiangqiang Zhou, Jiacong Yu, Jiawei Xu, Yong Chen, Xin Huang, Ping Li
Pose-Anchored Optical Flow for Low-Latency Human Action Anticipation in Human-Robot Teaming proposes PoseOFF, a representation that captures local motion around human joints by conditioning optical flow extraction on pose. This structured motion representation aligns with human kinematics and improves early action recognition accuracy across multiple datasets and backbones. PoseOFF achieves comparable or better performance while observing less of the action sequence, making it suitable for real‑time, resource‑constrained robotic systems.
By Lewis de Zoete Grundy, Chris McCarthy, Christopher Fluke
StreamPI introduces a streaming multimodal temporal modeling framework that enhances Vision‑Language‑Action models by adding temporal reasoning without extra parameters. It anchors each visual observation and language instruction pair as a temporal unit, using bidirectional attention for cross‑modal fusion and causal attention for autoregressive streaming inference. The method employs random‑interval streaming training to improve robustness and leverages the LLM backbone’s length extrapolation to inherit pretrained weights, achieving superior performance over pi0.5 on real‑robot and simulation tasks.
By Zhe Liu, Jinghua Hou, Yuxiang Lu, Zhenya Yang, Xianzhe Fan, Junwei Luo, Junyi Li, Ruihua Han, Zhi Hou, Hengshuang Zhao
Zero-WAM introduces a causal video-action model that enables robots to perform unseen manipulation tasks by following in-context human video guidance. The authors create HumanGen, a dataset of 74.2K human-robot ICL pairs across 8.6K tasks, and propose an in-context future chunk prediction objective to prevent shortcut learning. In simulation, Zero-WAM attains a 47.0% success rate on seven unseen tasks, outperforming the best video-action baseline by 29.5 percentage points, and demonstrates real‑world generalization to complex, long‑horizon, and fine‑grained tasks.
By Jiaming Zhou, Qihang Zhang, Gangwei Xu, Cunxin Fan, Yujie Zhao, Ruilin Wang, Yiming Luo, Shuai Yang, Xing Zhu, Yujun Shen, Junwei Liang, Yinghao Xu
GaussVLA is a Vision‑Language‑Action model that enhances spatial reasoning by converting flat 2D visual tokens into compact 3D Gaussian tokens using a Gaussian Spatial Tokenizer. It further employs a Depth‑Aware Chain‑of‑Thought module to perform structured, non‑autoregressive geometric reasoning conditioned on language and flow‑time. In both simulated and real‑world tests, GaussVLA achieves high spatial‑manipulation success rates—93.5% on LIBERO and 100% on the Spatial suite—while using only 200 M parameters, outperforming SpatialVLA by 19.7% relative success.
By Md Selim Sarowar, Md Tanvir Islam, Sungho Kim, Sangtae Ahn
LaGen is an autoregressive framework that generates long‑horizon LiDAR scenes frame by frame, using a single‑frame input and bounding‑box conditions to produce high‑fidelity 4D scenes. It introduces a scene decoupling estimation module for better object‑level interaction and a noise modulation module to reduce error accumulation over time. Evaluations on the nuScenes dataset show that LaGen outperforms existing methods, especially on later frames.
By Sizhuo Zhou, Xiaosong Jia, Fanrui Zhang, Junjie Li, Juyong Zhang, Yukang Feng, Jianwen Sun, Songbur Wong, Junqi You, Junchi Yan
PIVOT is a new multi‑trajectory dataset and evaluation framework that captures real‑world scenes with diverse camera paths, preserving both sensor‑derived measured poses and COLMAP‑optimized poses along with calibrated and optimized intrinsics. It defines three benchmark families—seen vs. unseen trajectory generalization, measured vs. optimized pose sensitivity, and calibrated vs. optimized intrinsics sensitivity—and introduces a directed pose‑space Chamfer distance to assess pose coverage. The first version of PIVOT includes five scenes recorded with a DJI Mini 4 Pro and offers an open processing and Nerfstudio‑based evaluation toolchain, revealing a consistent quality gap between held‑out and unseen trajectories and significant sensitivity to pose source and camera intrinsics.
By Mary Raymond
TacForcing is a streaming action‑generation framework that incorporates execution‑time tactile feedback into contact‑rich manipulation. It replaces a separate reactive controller with a streaming action expert that conditions actions on evolving tactile observations, and introduces Execution‑Aware Tactile Attention (EATA) to focus tactile conditioning on actions near execution. The method achieves 65% success in six simulated UniVTAC tasks and 69% in three real‑world contact‑rich manipulation tasks, outperforming strong baselines.
By Jianbo Zhou, Boyuan Zhao, Yuzheng Zhang, Yiyang Chen, Wenxin Chen, Qiuyue Li, Xiangyang Gu, Yuhan Cao, Xiao Xia, Yanzhe Hu, Zhijie Deng
The paper introduces the Hybrid Hierarchical Multi-Agent Framework (H$^{2}$MAF), which fuses decision-level outputs from EfficientNet-B3 and ConvNeXt-Tiny with semantic arbitration by multimodal large language models Gemma 4 E4B and Qwen3.5 4B to produce explainable plant disease diagnoses. Evaluated on 14,364 images from PlantDoc and two Cornell robotic field datasets, the framework achieves up to 99.3% accuracy, with Gemma improving PlantDoc accuracy from 63.9% to 68.5% and demonstrating low critical‑risk error. The results highlight the potential of MLLM arbitration for reliable, explainable agricultural AI under real‑world field conditions.
By Ranjan Sapkota, Konstantinos I. Roumeliotis, Pengyao Xie, Nikolaos D. Tselikas, Lirong Xiang, Manoj Karkee
The paper titled "Lost in Motion: Vision Language Models Fail the Dynamic Gauges Test" examines how current Vision‑Language Models (VLMs) perform on real‑time gauge reading tasks in industrial settings. Using a new dataset of video sequences featuring circular, linear, and Vernier gauges under varied motion, the study tests state‑of‑the‑art models from GPT‑5 and Gemini 3 families. Results show that these VLMs struggle to accurately interpret needle trajectories and scale semantics, falling short of the traceability and reliability required for safety‑critical monitoring and thus not meeting IEEE and ISO standards for trustworthy synthetic instruments.
By Tairan Fu, Francisco Javier Santos-Mart\'in, Javier Conde, Elena Merino-G\'omez, Pedro Reviriego
The paper introduces Activation Replay, a training‑free method that improves reasoning in post‑trained large multimodal models (LMMs) by replaying low‑entropy activations from the base model’s input context. It shows that Reinforcement Learning with Verifiable Rewards (RLVR) shifts low‑entropy activations and that modulating these activations enhances reasoning across tasks such as mathematics, visual agents, and video reasoning. Experiments demonstrate that Activation Replay outperforms alternatives like high‑entropy replay or direct cross‑model intervention, boosting Pass@K and broadening RLVR’s reasoning coverage.
By Yun Xing, Xiaobin Hu, Qingdong He, Jiangning Zhang, Shuicheng Yan, Shijian Lu, Yu-Gang Jiang
V-Link is a method designed to enhance Vision‑Language‑Action (VLA) models by recovering visual representations during the transfer from vision‑language (VL) features to action (A) features. It introduces complementary Spatial and Semantic Query representations that are injected into Action DiT through asymmetric pathways, providing both semantic augmentation and dedicated geometric conditioning for action generation. Experiments on LIBERO, LIBERO‑Plus, RoboTwin 2.0, and real‑world AGIBOT A3 Ultra tasks show significant performance gains over the base GR00T N1.6 model.
By Yehao Lu, Jiarui Yang, Yuning Su, Yufeng Xie, Yu Zhong, Yazhou Zhang, Haiyu Lan, Kaixiang Lu, Peiwen Lin, Chuang Wang, Zequn Qin, Enyu Li, Xi Li
PointRL introduces a verifiable reinforcement learning framework that learns point-level vision‑language grounding from heterogeneous annotation evidence such as bounding boxes, masks, and instance labels. The method converts these annotations into pointing instructions while preserving target supports, instance membership, and set constraints as hidden verifier evidence, which a deterministic checker uses to score predictions. Evaluation on PointArena shows that PointRL improves Qwen3.5‑4B’s accuracy from 56.11% to 65.58%, and similar gains are observed on RoboSpatial, BLINK, and Ref‑Adv benchmarks.
By Jingyang Su, Pu Cao, Xiuze Jin, Longyue Zhang, Qing Song, Lu Yang
The paper introduces FISER, a framework that explicitly infers human goals and intentions before planning actions for AI agents to follow natural language instructions in collaborative embodied tasks. It employs Transformer-based models and is evaluated on the HandMeThat benchmark, outperforming end-to-end approaches and strong baselines such as Chain of Thought prompting. FISER achieves state‑of‑the‑art performance on this embodied social reasoning task.
By Yanming Wan, Yue Wu, Yiping Wang, Jiayuan Mao, Natasha Jaques
LM‑X is a generalist vision‑language‑action policy that augments action prediction with three online, explicitly supervised signals: return‑to‑go (RTG) for task progress, event‑to‑go (ETG) for the next semantic transition, and heteroscedastic action flow for local reliability. By conditioning action generation on these signals, LM‑X embeds explainability directly into control rather than as a post‑hoc explanation. After a 20‑day pretraining run on 64 GPUs, LM‑X outperforms an action‑only backbone by 16.0 points and a single‑head variant by 10.8 points, and achieves 74.1 % success on 50 RoboTwin2.0 tasks and 68.6 % on seven real‑robot tasks, surpassing the GR00T N1.7 baseline.
By Jin Lou, Jingxuan Zhu, Andong Chen, Xupeng Wang, Yuan Xu, Yuexuan Li, Xingdong Zhu, Zhijie Zhu, Yingwei Ji, Wenpeng Nie, Jingyi Li, Liangliang Chen, Jinyan Liu, Zhiqi Song, Jidong Zhang, Hongming Li, Yuchen Zhu
TASE introduces a truncation‑aware embedding space that projects pretrained 2D semantic features into 3D scene representations, allowing flexible and controllable editing. The method optimizes feature channels so that fewer channels yield abstract semantics while more channels preserve detail, and it enforces multi‑view consistency with a scale‑ and translation‑equivariant loss. A finetuning stage for the editing diffusion model further reduces artifacts from geometric changes, achieving competitive performance and outperforming prior methods on large‑scale geometric edits.
By Tim-Felix Faasch, Jochen Kall, Lucas Nunes, Jens Behley, Cyrill Stachniss
The paper introduces $R^3$, a post‑training method that converts vision‑language models into robotic reasoners by first mid‑training on expert reasoning traces and then refining them with single‑step rubric‑based reinforcement learning. $R^3$ enables free‑form language reasoning to guide low‑level manipulation policies, improving exploration, generalization, and performance on long‑horizon tasks in Language Table and simulated bimanual grocery packing benchmarks. The approach outperforms instruction‑only imitation learning baselines and demonstrates that natural language reasoning can serve as a test‑time compute mechanism for steering robotic actions.
By Lehong Wu, Yuxiao Qu, Zheyuan Hu, Ivan Zhang, Limin Wei, Zackory Erickson, Aviral Kumar
Emergent Abilities in Large Language Models: A Survey reviews how scaling LLMs leads to previously unseen capabilities such as advanced reasoning, in-context learning, coding, and problem-solving. The paper critically examines definitions, inconsistencies, and the conditions that foster these abilities, including scaling laws, task complexity, pre‑training loss, quantization, and prompting strategies. It also discusses the extension to Large Reasoning Models and highlights safety concerns like deception, manipulation, and reward hacking, calling for improved evaluation and governance.
By Leonardo Berti, Flavio Giorgi, Gjergji Kasneci