{"total":33,"items":[{"citing_arxiv_id":"2607.01067","ref_index":44,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Human-Centric Transferable Tactile Pre-Training for Dexterous Robotic Manipulation","primary_cat":"cs.RO","submitted_at":"2026-07-01T15:26:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Introduces H-Tac human tactile-action dataset and TTP pre-training that unifies spaces and predicts future tactile signals to improve robotic dexterous manipulation transfer.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2607.00678","ref_index":40,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"ABot-M0.5: Unified Mobility-and-Manipulation World Action Model","primary_cat":"cs.CV","submitted_at":"2026-07-01T09:21:20+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ABot-M0.5 proposes a unified mobility-and-manipulation world action model using three alignment strategies that achieves state-of-the-art performance on mobile and fine-grained manipulation benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.29941","ref_index":51,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Seeing Touch from Motion: A Unified Modality-Aware Visuo-Tactile Policy with Tactile Motion Correlation","primary_cat":"cs.RO","submitted_at":"2026-06-29T08:20:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A visuo-tactile policy learning method that exploits tactile motion correlation for contact state distinction and Mixture-of-Transformers for cross-modal fusion.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.28133","ref_index":38,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Translation as a Bridging Action: Transferring Manipulation Skills from Humans to Robots","primary_cat":"cs.RO","submitted_at":"2026-06-26T14:34:04+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A relative wrist translation bridging action with a vision-language-action model using interleaved tokens and attention masking transfers human manipulation skills to robots more effectively than 6DoF actions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.22136","ref_index":9,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Wh0: Generative World Models as Scalable Sources of Egocentric Human Hand Manipulation Data","primary_cat":"cs.RO","submitted_at":"2026-06-20T16:31:40+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Wh0 generates scalable egocentric human manipulation videos with world models and converts them to boost pretrained VLA models' zero-shot dexterous task success from 8.3% to 38.9% on 18 real-world tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.20521","ref_index":25,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"HumanScale: Egocentric Human Video Can Outperform Real-Robot Data for Embodied Pretraining","primary_cat":"cs.CV","submitted_at":"2026-06-18T17:37:34+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Processed egocentric human video outperforms teleoperated real-robot trajectories as pretraining data for embodied foundation models, delivering 24% lower validation loss and 52.5-90% higher task success rates under matched post-training protocols.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.17846","ref_index":27,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Models","primary_cat":"cs.RO","submitted_at":"2026-06-16T12:14:39+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Qwen-RobotManip applies unified alignment across representation, motion, and behavior to enable large-scale training on heterogeneous manipulation data, yielding emergent generalization on out-of-distribution robotic benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.16533","ref_index":94,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Kairos: A Native World Model Stack for Physical AI","primary_cat":"cs.AI","submitted_at":"2026-06-15T10:37:42+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Kairos is a native world model stack using cross-embodiment pretraining, hybrid linear temporal attention with theoretical error bounds, and deployment-aware co-design, reporting top performance on embodied benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.09615","ref_index":3,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"DexPIE: Stable Dexterous Policy Improvement from Real-World Experience","primary_cat":"cs.RO","submitted_at":"2026-06-08T15:21:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"DexPIE improves dexterous manipulation success rates by 37% over demo policies via real-world experience collection with adapted intervention, multi-stage DAgger, asynchronous relative-action inference, and optimality conditioning.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.09457","ref_index":25,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"$\\omega$-EVA: Envision, Verify, and Act with Latent Interactive World Models","primary_cat":"cs.RO","submitted_at":"2026-06-08T13:12:56+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ω-EVA is a three-stage latent world model framework that trains action-conditioned dynamics, a language-conditioned flow policy, and a tri-branch refiner to improve embodied action generation in simulation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.06194","ref_index":22,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"ActiveMimic: Egocentric Video Pretraining with Active Perception","primary_cat":"cs.RO","submitted_at":"2026-06-04T14:01:01+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ActiveMimic pretrains on egocentric human video by recovering and modeling active camera motion as viewpoint actions, matching robot-data pretraining performance on real-world tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.06033","ref_index":7,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"RealDexUMI: A Wearable Universal Manipulation Interface for Dexterous Robot Learning","primary_cat":"cs.RO","submitted_at":"2026-06-04T11:28:23+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A wearable interface with a shared dexterous hand module enables retargeting-free teleoperation and matched data collection, yielding policies with 88.75% average success across eight real-robot tasks that generalize and transfer across embodiments.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.03868","ref_index":22,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Unified Video-Action Joint Denoising for Dexterous Action and Data Generation","primary_cat":"cs.CV","submitted_at":"2026-06-02T16:39:23+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Donk is a unified video-action denoising model that generates dexterous hand trajectories and videos under language, image, and state conditioning while also serving as a text-conditioned data engine.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.03177","ref_index":31,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"ConTrack: Constrained Hand Motion Tracking with Adaptive Trade-off Control","primary_cat":"cs.RO","submitted_at":"2026-06-02T05:31:32+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"ConTrack introduces a constrained RL method with online dual-variable adaptation and adaptive resets for improved long-horizon hand tracking in simulation and on real robots.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.31286","ref_index":26,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"DeMaVLA: A Vision-Language-Action Foundation Model for Generalizable Deformable Manipulation","primary_cat":"cs.RO","submitted_at":"2026-05-29T13:20:08+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"DeMaVLA is a VLA foundation model using a pruned action expert and flow matching, pre-trained on 5000 hours of real demonstrations and post-trained on multi-task folding data with human-in-the-loop correction, reporting competitive benchmark and real-world folding performance.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.30226","ref_index":16,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"BORA: Bridging Offline Reinforcement Learning and Online Residual Adaptation for Real-World Dexterous VLA Models","primary_cat":"cs.RO","submitted_at":"2026-05-28T16:57:47+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"BORA combines offline RL critic training with online chunk-wise residual adaptation to raise average success rates of real-world dexterous VLA policies by 33% and up to 43% on unseen objects across five tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.24890","ref_index":19,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"QuoVLA: Quotient Space for Vision-Language-Action Models","primary_cat":"cs.CV","submitted_at":"2026-05-24T06:28:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"QuoVLA introduces a quotient-space framework that compresses VLM latents into action-sufficient representations via quantization and dual-branch design for better VLA generalization.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.23733","ref_index":39,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Any2Any: Efficient Cross-Embodiment Transfer for Humanoid Whole-Body Tracking","primary_cat":"cs.RO","submitted_at":"2026-05-22T15:10:42+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Any2Any transfers humanoid whole-body tracking models across embodiments via kinematic alignment followed by targeted PEFT, matching full-training performance with 1% of the data and compute on tested platforms.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.21414","ref_index":44,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"PointACT: Vision-Language-Action Models with Multi-Scale Point-Action Interaction","primary_cat":"cs.RO","submitted_at":"2026-05-20T17:10:31+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"PointACT proposes a 3D-aware dual-system VLA policy using multi-scale point-action interaction with bottleneck window self-attention, achieving 10% higher success rates on RLBench-10Tasks over prior pretrained VLAs.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.20752","ref_index":2,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"GaussianDream: A Feed-Forward 3D Gaussian World Model for Robotic Manipulation","primary_cat":"cs.RO","submitted_at":"2026-05-20T05:51:30+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"GaussianDream is a feed-forward 3D Gaussian world model plug-in that conditions VLA policies on learned 3D spatial and future evolution representations for improved robotic manipulation performance.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.16743","ref_index":35,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"LACE: Latent Visual Representation for Cross-Embodiment Learning","primary_cat":"cs.RO","submitted_at":"2026-05-16T01:50:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"LACE aligns human-robot visual features via semantic distribution matching on corresponding body parts plus Gram loss, yielding 65% better zero-shot policy transfer than baseline DINO.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.14950","ref_index":27,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Evo-Depth: A Lightweight Depth-Enhanced Vision-Language-Action Model","primary_cat":"cs.CV","submitted_at":"2026-05-14T15:21:36+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Evo-Depth is a compact VLA model using a lightweight implicit depth encoder from RGB views plus progressive alignment to boost manipulation performance without added hardware.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.13083","ref_index":21,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"TouchAnything: A Dataset and Framework for Bimanual Tactile Estimation from Egocentric Video","primary_cat":"cs.RO","submitted_at":"2026-05-13T06:54:36+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"EgoTouch is a new multi-view egocentric dataset with dense bimanual tactile supervision, and TouchAnything is a baseline framework showing that wrist views improve vision-based tactile prediction over egocentric input alone.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Tactile saliency prediction for robust sim-to-real tactile control. In IROS, pages 10806-10812, 2023. doi: 10.1109/IROS55552.2023.10341888. URLhttps://doi.org/10.1109/IROS55552.2023.10341888. [20] Yunze Liu, Yun Liu, Che Jiang, et al. Hoi4d: A 4d egocentric dataset for category-level human-object interaction. In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022. [21] Hao Luo, Ye Wang, Wanpeng Zhang, Sipeng Zheng, Ziheng Xi, Chaoyi Xu, Haiweng Xu, Haoqi Yuan, Chi Zhang, Yiqing Wang, Yicheng Feng, and Zongqing Lu. Being-h0.5: Scaling human-centric robot learning for cross-embodiment generalization. CoRR, abs/2601.12993, 2026. doi: 10.48550/ARXIV.2601.12993. URL https://doi.org/10.48550/arXiv.2601.12993. [22] Maxime Oquab, Timothée Darcet, Théo Moutakanni, et al."},{"citing_arxiv_id":"2605.06747","ref_index":25,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"HumanNet: Scaling Human-centric Video Learning to One Million Hours","primary_cat":"cs.CV","submitted_at":"2026-05-07T15:21:58+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"HumanNet is a 1M-hour human-centric video dataset with interaction annotations that enables better vision-language-action model performance than equivalent robot data in a controlled test.","context_count":1,"top_context_role":"dataset","top_context_polarity":"background","context_text":"actions are executed, exposing contact dynamics, hand-object relations, temporal intent, and the visual consequences of motor decisions. Third-person video complements this signal by making full-body motion, posture, interaction context, surrounding agents, and scene-level dynamics easier to observe. Large-scale community resources such as Ego4D [13], EPIC-KITCHENS [7], Ego-Exo4D [14], and EgoSchema [25] have expanded recognition, forecasting, narration, and multimodal understanding from egocentric and paired exocentric video, while structured interaction resources such as HOI4D [21] show the value of dense hand-object supervision. Recent work has shown that human-centered data can improve robot learning and representation learning [9, 18, 28, 30, 38], but current corpora remain limited in duration, fragmented across collection efforts,"},{"citing_arxiv_id":"2605.00080","ref_index":36,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"World Model for Robot Learning: A Comprehensive Survey","primary_cat":"cs.RO","submitted_at":"2026-04-30T14:35:31+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"A comprehensive survey that organizes the literature on world models in robot learning, their roles in policy learning, planning, simulation, and video-based generation, with connections to navigation, driving, datasets, and benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.00078","ref_index":7,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Being-H0.7: A Latent World-Action Model from Egocentric Videos","primary_cat":"cs.RO","submitted_at":"2026-04-30T14:16:15+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Being-H0.7 adds future-aware latent reasoning to direct VLA policies via dual-branch alignment on latent queries, matching world-model benefits at VLA efficiency.","context_count":1,"top_context_role":"extension","top_context_polarity":"extend","context_text":"trajectories, keypoints, and bounding boxes [92-99]. Moving closer to direct policy supervision, another line recovers human actions from videos via hand pose, wrist motion, or retargeted manipulation trajectories for VLA pretraining [100-106], with Being-H0 [6] scaling this paradigm through motion-tokenized reconstructed human hand trajectories. Being-H0.5 [7] further generalizes this direction toward unified cross-embodiment pretraining. Our work follows this data-centric scaling route and introduces latent world-action modeling as a reasoning form to inject future-aware structure into human-robot VLA pretraining. 3 Method 3.1 Latent Reasoning: At the Crossroads of VLA and World-Action Model An effective embodied model should not only react to the instant context but also truly understand how the"},{"citing_arxiv_id":"2604.24681","ref_index":49,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Learning Human-Intention Priors from Large-Scale Human Demonstrations for Robotic Manipulation","primary_cat":"cs.RO","submitted_at":"2026-04-27T16:42:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"MoT-HRA learns embodiment-agnostic human-intention priors from a curated 2.2M-episode human video dataset via a three-expert hierarchical vision-language-action model to improve robotic manipulation under distribution shift.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Reconstructing hands in 3D with transformers. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024. 12 [48] Haotong Lin, Sili Chen, Junhao Liew, Donny Y Chen, Zhenyu Li, Guang Shi, Jiashi Feng, and Bingyi Kang. Depth anything 3: Recovering the visual space from any views.arXiv preprint arXiv:2511.10647, 2025. [49] Hao Luo, Ye Wang, Wanpeng Zhang, Sipeng Zheng, Ziheng Xi, Chaoyi Xu, Haiweng Xu, Haoqi Yuan, Chi Zhang, Yiqing Wang, et al. Being-h0. 5: Scaling human-centric robot learning for cross-embodiment generalization.arXiv preprint arXiv:2601.12993, 2026. [50] Andreas Steiner, André Susano Pinto, Michael Tschannen, Daniel Keysers, Xiao Wang, Yonatan Bitton, Alexey Gritsenko, Matthias Minderer, Anthony Sherbondy, Shangbang Long, et al."},{"citing_arxiv_id":"2604.23570","ref_index":26,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"EgoLive: A Large-Scale Egocentric Dataset from Real-World Human Tasks","primary_cat":"cs.RO","submitted_at":"2026-04-26T07:21:15+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"EgoLive is presented as the largest open-source annotated egocentric dataset for real-world task-oriented human routines, captured with a custom head-mounted device and multi-modal annotations exclusively in unconstrained environments.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.20100","ref_index":2,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"JoyAI-RA 0.1: A Foundation Model for Robotic Autonomy","primary_cat":"cs.RO","submitted_at":"2026-04-22T01:51:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"JoyAI-RA is a multi-source pretrained VLA model that claims to bridge human-to-robot embodiment gaps via data unification and outperforms prior methods on generalization-heavy robotic tasks.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"As a result, current corpora often underrepresent long-tail interactions, rare failure modes, and the diversity of scene layouts encountered in open-world settings. This lack of coverage restricts the breadth of behavior priors that policies can acquire and weakens robustness under distribution shift. Recent VLA systems have begun to explore broader forms of pretraining and cross-embodiment scaling [2, 5, 15, 30, 43], yet effective knowledge transfer across substantially different embodiments remains a central obstacle to scalable robot learning. To step forward for addressing these challenges, we proposeJoyAI-RA, a VLA embodied foundation model 1 arXiv:2604.20100v2 [cs.RO] 23 Apr 2026 for generalizable robotic manipulation. JoyAI-RA leverages a large-scale multi-source pretraining dataset"},{"citing_arxiv_id":"2604.18000","ref_index":6,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Unmasking the Illusion of Embodied Reasoning in Vision-Language-Action Models","primary_cat":"cs.RO","submitted_at":"2026-04-20T09:25:30+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"State-of-the-art vision-language-action models catastrophically fail dynamic embodied reasoning due to lexical-kinematic shortcuts, behavioral inertia, and semantic feature collapse caused by architectural bottlenecks, as shown by the new BeTTER benchmark with real-world validation.","context_count":1,"top_context_role":"baseline","top_context_polarity":"baseline","context_text":"existing evaluation protocols to distinguish whether this apparent proficiency stems from true reasoning or the mere exploitation of shortcut behaviors memorized from training data. To address this, we conduct a diagnostic study using our BeTTER benchmark, with the goal to identify the reasoning weakness overlooked in prior VLAs. We evaluate three representative VLAs:π0.5[7], GR00T-N1.6 [5], and Being-H0.5 [6]. To systematically probe their reasoning limitations, we introduce four categories of targeted semantic and physical interventions at test time:1) Spatial Layout Shift. We rearrange object positions or alter their spatial relationships with nearby distractors, testing the model's ability to adapt to out-of-distribution configurations.2) Primitive"},{"citing_arxiv_id":"2604.16592","ref_index":116,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Human Cognition in Machines: A Unified Perspective of World Models","primary_cat":"cs.RO","submitted_at":"2026-04-17T17:51:46+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"The paper introduces a unified framework for world models that fully incorporates all cognitive functions from Cognitive Architecture Theory, highlights under-researched areas in motivation and meta-cognition, and proposes Epistemic World Models as a new category for scientific discovery agents.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"✓ ✗ ✗ ✓ ✓ ✗ ✗Interleaves video and action to- kens for joint imagination and action decoding VLA-JEPA [164] 2026 Robot.✓ ✓ ✗ ✗ ✗ ✗ ✗Leakage-free JEPA grounding vi- sual encoder in action-relevant dynamics VAGEN [178] 2025 Robot.✗ ✗ ✓ ✓ ✗ ✓ ✗RL-structured World Model rea- soning into state estimation and transition modeling for VLM agents BeingH VLA [116] 2026 Robot✗ ✗ ✓ ✓ ✗ ✓ ✗A foundational VLA for robust cross-embodiment generalization across diverse robotic platforms PointWorld [70] 2026 Robot.✗ ✓ ✗ ✓ ✓ ✗ ✗Unifies state and action as 3D point flows with MPC over imag- ined scene deformations ManiGaussian [114] 2024 Robot.✗ ✓ ✗ ✓ ✓ ✗ ✗Dynamic 3DGS World Model predicting future Gaussian scenes under action for manipu-"},{"citing_arxiv_id":"2604.15483","ref_index":49,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"${\\pi}_{0.7}$: a Steerable Generalist Robotic Foundation Model with Emergent Capabilities","primary_cat":"cs.LG","submitted_at":"2026-04-16T19:18:07+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"π₀.₇ is a steerable generalist robotic model that uses rich multimodal prompts including language, subgoal images, and performance metadata to achieve out-of-the-box generalization across tasks and robot bodies.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"language-action models from egocentric human videos, 2025. URL https://arxiv.org/abs/2507.12440. [48] Hao Luo, Ye Wang, Wanpeng Zhang, Sipeng Zheng, Ziheng Xi, Chaoyi Xu, Haiweng Xu, Haoqi Yuan, Chi Zhang, Yiqing Wang, Yicheng Feng, and Zongqing Lu. Being-h0.5: Scaling human-centric robot learning for cross-embodiment generalization, 2026. URL https:// arxiv.org/abs/2601.12993. [49] Chubin Zhang, Jianan Wang, Zifeng Gao, Yue Su, Tianru Dai, Cai Zhou, Jiwen Lu, and Yansong Tang. Clap: Contrastive latent action pretraining for learn- ing vision-language-action models from human videos, 2026. URL https://arxiv.org/abs/2601.04061. 3 [50] Physical Intelligence Team.π ⋆ 0.6: a vla that learns from experience.arXiv preprint arXiv:2511."},{"citing_arxiv_id":"2604.07993","ref_index":29,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"HEX: Humanoid-Aligned Experts for Cross-Embodiment Whole-Body Manipulation","primary_cat":"cs.RO","submitted_at":"2026-04-09T09:01:43+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"HEX introduces a state-centric framework with humanoid-aligned representations and mixture-of-experts proprioceptive prediction for coordinated whole-body control on bipedal humanoids.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"International Conference on Robotics and Automation (ICRA), 2026. 15 [28] Chenhao Lu, Xuxin Cheng, Jialong Li, Shiqi Yang, Mazeyu Ji, Chengjing Yuan, Ge Yang, Sha Yi, and Xiaolong Wang. Mobile-television: Predictive motion priors for humanoid whole-body control. In2025 IEEE International Conference on Robotics and Automation (ICRA), pages 5364-5371. IEEE, 2025. [29] Hao Luo, Ye Wang, Wanpeng Zhang, Sipeng Zheng, Ziheng Xi, Chaoyi Xu, Haiweng Xu, Haoqi Yuan, Chi Zhang, Yiqing Wang, et al. Being-h0. 5: Scaling human-centric robot learning for cross-embodiment generalization.arXiv preprint arXiv:2601.12993, 2026. [30] Zhengyi Luo, Ye Yuan, Tingwu Wang, Chenran Li, Sirui Chen, Fernando Castaneda, Zi-Ang Cao, Jiefeng Li, David"}],"limit":50,"offset":0}