{"total":18,"items":[{"citing_arxiv_id":"2607.02195","ref_index":51,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Bridge-WA: Predicting Where and How the World Changes for Robotic Action","primary_cat":"cs.RO","submitted_at":"2026-07-02T14:03:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Bridge-WA introduces a lightweight distillation-based world-action model that uses future-change priors to improve robotic task success and robustness without deployment-time dense rollouts.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.20781","ref_index":197,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"World Action Models: A Survey","primary_cat":"cs.RO","submitted_at":"2026-06-18T17:05:19+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"A survey that clarifies boundaries and organizes World Action Models by generation requirements and predictive substrates, identifying a trend toward generating less of the future.","context_count":1,"top_context_role":"dataset","top_context_polarity":"background","context_text":"Simulation sits at the high-throughput end, where benchmarks probe different competences. The LIBERO family [34, 105, 159, 211] and RoboMME [22] focus on language-conditioned manipulation and multimodal evaluation. robomimic [116], MetaWorld [190], ManiSkill 2 [47], and ManiSkill 3 [150] provide manipulation suites with different levels of control diversity and physics coverage. HomeRobot [189], VLABench [197], RoboEval [165], and RoboVerse [43] broaden the benchmark space toward navigation, embodied reasoning, and multi-backend evaluation. Real-robot arenas such as RoboArena [5] and RoboChallenge [179] trade throughput for physical validity by running the policy on hardware. DWorldEval [97] closes the loop inside a learned world model instead, scoring a policy from imagined rollouts whose success correlates with real execution."},{"citing_arxiv_id":"2606.19297","ref_index":79,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Does VLA Even Know the Basics? Measuring Commonsense and World Knowledge Retention in Vision-Language-Action Models","primary_cat":"cs.LG","submitted_at":"2026-06-17T17:20:46+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Act2Answer protocol reveals VLA models retain simple concepts but show larger gaps on complex semantics than source VLMs, with VQA co-training linked to better retention and knowledge signals peaking in middle layers.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.10382","ref_index":20,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"UMI-Bench 1.0: An Open and Reproducible Real-World Benchmark for Tabletop Robotic Manipulation with UMI Data","primary_cat":"cs.RO","submitted_at":"2026-06-09T03:47:54+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"UMI-Bench 1.0 is presented as the first open benchmark dedicated to reproducible real-world evaluation of Universal Manipulation Interface policies.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.10366","ref_index":14,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A Practical Recipe Towards Improving Sim-and-Real Correlation for VLA Evaluation","primary_cat":"cs.RO","submitted_at":"2026-06-09T03:25:02+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Authors perform a cross-simulator, cross-policy empirical study of sim-to-real correlation for VLA policies and distill guidance on using simulation for policy improvement.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.09669","ref_index":87,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"SpatialWorld: Benchmarking Interactive Spatial Reasoning of Multimodal Agents in Real-World Tasks","primary_cat":"cs.AI","submitted_at":"2026-06-08T15:51:51+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"SpatialWorld is a new multi-simulator benchmark showing top multimodal agents achieve under 18% success on interactive spatial tasks requiring active exploration and long-horizon planning.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.07723","ref_index":78,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"VoLo: A Physical Orchestrator for Open-Vocabulary Long-Horizon Manipulation","primary_cat":"cs.RO","submitted_at":"2026-06-05T16:21:27+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"VoLoAgent uses a VLM to steer heterogeneous robot capabilities as interruptible tools for long-horizon manipulation and introduces the RoboVoLo benchmark, claiming substantial outperformance over single VLA/VLM or tool-based systems with real-robot validation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.03784","ref_index":60,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Revisiting Embodied Chain-of-Thought for Generalizable Robot Manipulation","primary_cat":"cs.RO","submitted_at":"2026-06-02T15:37:59+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ERVLA trains on a 978k-trajectory embodied CoT corpus using reasoning as supervision with dropout, then predicts actions without CoT at test time, reaching 86.9% on LIBERO-Plus and 53.2% on VLABench.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.02277","ref_index":41,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"RoboSemanticBench: Diagnosing Semantic Grounding in Action Prediction for VLA Models","primary_cat":"cs.RO","submitted_at":"2026-06-01T14:02:37+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"RoboSemanticBench reveals that representative VLA models grasp blocks successfully but select the semantically correct answer at near-random rates, indicating a gap between backbone semantics and action prediction.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.27759","ref_index":14,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Colosseum V2: Benchmarking Generalization for Vision Language Action Models","primary_cat":"cs.RO","submitted_at":"2026-05-26T23:17:39+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Introduces Colosseum V2 benchmark for evaluating VLA model generalization in robotic manipulation with 28 tasks, revealing limitations in current methods and sim-real correlations.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.18727","ref_index":60,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"DexHoldem: Playing Texas Hold'em with Dexterous Embodied System","primary_cat":"cs.RO","submitted_at":"2026-05-18T17:51:34+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DexHoldem is a new benchmark providing 1,470 teleoperated demonstrations across 14 manipulation primitives, plus standardized tests for dexterous policy execution and agentic perception in a physical Texas Hold'em setting.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12090","ref_index":240,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"World Action Models: The Next Frontier in Embodied AI","primary_cat":"cs.RO","submitted_at":"2026-05-12T13:10:52+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"The paper introduces World Action Models as a new paradigm unifying predictive world modeling with action generation in embodied foundation models and provides a taxonomy of existing approaches.","context_count":1,"top_context_role":"dataset","top_context_polarity":"background","context_text":"RoboMME [224], GenManip [ 225], VLABench [ 226], RoboSuite [227], RoboLab [228] SimplerEnv [229], ARNOLD [230], GemBench [231] Bimanual and Humanoid Form Robo T win [153], BiGym [232], HumanoidBench [ 233] HumanoidGen [234] Mobile Manipulation ManipulaTHOR [235], HomeRobot [236], BEHA VIOR-1K [237] Contact and Deformation Manipulation SoftGym [238], PlasticineLab [239], DaXBench [240] TacSL [241], ManiFeel [242] Real-Device RoboArena [243], RoboChallenge [244], Maniparena [245] Figure 2 The comprehensive roadmap and taxonomy of World Action Models (W AMs) reviewed in this survey. The literature is systematically categorized into four core dimensions: background ( Sec. 3 ), architecture ( Sec. 4 ), training data (Sec. 5), and evaluation protocols ( Sec."},{"citing_arxiv_id":"2605.10408","ref_index":60,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"VISOR: A Vision-Language Model-based Test Oracle for Testing Robots","primary_cat":"cs.SE","submitted_at":"2026-05-11T11:46:57+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"VISOR is a VLM-based automated test oracle that evaluates robot task correctness and quality from videos while reporting its own uncertainty, tested on GPT and Gemini across four tasks and over 1000 videos with Gemini showing higher recall and GPT higher precision but low uncertainty-correctness tie","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.06311","ref_index":48,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Toward Visually Realistic Simulation: A Benchmark for Evaluating Robot Manipulation in Simulation","primary_cat":"cs.RO","submitted_at":"2026-05-07T14:13:05+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"VISER is a new visually realistic simulation benchmark for robot manipulation tasks that uses PBR materials and MLLM-assisted asset generation, achieving 0.92 Pearson correlation with real-world policy performance.","context_count":1,"top_context_role":"baseline","top_context_polarity":"baseline","context_text":"org/abs/2412.18194. [47] Jinliang Zheng, Jianxiong Li, Zhihao Wang, Dongxiu Liu, Xirui Kang, Yuchun Feng, Yinan Zheng, Jiayin Zou, Yilun Chen, Jia Zeng, Ya-Qin Zhang, Jiangmiao Pang, Jingjing Liu, Tai Wang, and Xianyuan Zhan. X-vla: Soft-prompted transformer as scalable cross-embodiment vision-language-action model, 2025. URLhttps://arxiv.org/abs/2510.10274. [48] Liming Zheng, Feng Yan, Fanfan Liu, Chengjian Feng, Zhuoliang Kang, and Lin Ma. Robocas: A benchmark for robotic manipulation in complex object arrangement scenarios, 2024. URL https://arxiv.org/abs/2407.06951. 13 A Technical appendices and supplementary material A.1 3D Asset Dataset Overview We present the overview of our 3D asset dataset in Fig."},{"citing_arxiv_id":"2603.13966","ref_index":17,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"vla-eval: A Unified Evaluation Harness for Vision-Language-Action Models","primary_cat":"cs.AI","submitted_at":"2026-03-14T14:38:53+00:00","verdict":"ACCEPT","verdict_confidence":"MODERATE","novelty_score":5.0,"formal_verification":"none","one_line_summary":"vla-eval decouples VLA model inference from benchmark execution via WebSocket and Docker, supporting 14 benchmarks with up to 47x speedup and reproducing published scores across six codebases.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2507.21046","ref_index":119,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence","primary_cat":"cs.AI","submitted_at":"2025-07-28T17:59:05+00:00","verdict":"ACCEPT","verdict_confidence":"MODERATE","novelty_score":4.0,"formal_verification":"none","one_line_summary":"The paper delivers the first systematic review of self-evolving agents, structured around what components evolve, when adaptation occurs, and how it is implemented.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2507.01925","ref_index":276,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A Survey on Vision-Language-Action Models: An Action Tokenization Perspective","primary_cat":"cs.RO","submitted_at":"2025-07-02T17:34:52+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"The survey frames VLA models as pipelines that generate progressively grounded action tokens and classifies those tokens into eight types to guide future development.","context_count":1,"top_context_role":"dataset","top_context_polarity":"use_dataset","context_text":"Recent works further exploit weak supervision to extract actionable representations from large-scale videos. Frame-level captioning and temporal alignment provide indirect supervision signals for generating trajectory- based and goal state action tokens. For instance, Magma [179] introduces Set-of-Mark and Trace-of-Mark abstractions to anchor action grounding within video streams. Ego-Exo4D [276] augments egocentric data with third-person views for 3D motion grounding, facilitating embodiment transfer. These approaches enable VLA models to build temporal grounding and language-conditioned policy priors in open-world settings. 12.2. Middle Layer: Synthetic and Simulation Data To provide a critical bridge between the human video and the high cost of real-world data collection, VLA"},{"citing_arxiv_id":"2503.03480","ref_index":67,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"SafeVLA: Towards Safety Alignment of Vision-Language-Action Model via Constrained Learning","primary_cat":"cs.RO","submitted_at":"2025-03-05T13:16:55+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"SafeVLA applies constrained reinforcement learning via CMDP min-max optimization to VLAs, cutting safety violation costs by 83.58% while preserving task success on long-horizon mobile manipulation tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}