{"total":29,"items":[{"citing_arxiv_id":"2607.00442","ref_index":77,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Learning Gait-Aware Quadruped Locomotion with Temporal Logic Specifications","primary_cat":"cs.RO","submitted_at":"2026-07-01T04:57:39+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Framework using parameterized Signal Temporal Logic specifications to shape rewards for PPO-based RL, yielding tighter velocity tracking and more stable training than hand-crafted rewards on Barkour quadruped in MuJoCo simulation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.31700","ref_index":9,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Diffusing Blame: Task-Dependent Credit Assignment in Biologically Plausible Dual-Stream Networks","primary_cat":"cs.LG","submitted_at":"2026-06-30T14:09:40+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Dual-stream excitatory/inhibitory networks trained with modulo error routing achieve 96.7% MNIST and 61.7% CIFAR-10 accuracy plus competitive RL performance, revealing task-dependent credit-assignment bottlenecks under Dale's principle.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.29201","ref_index":10,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Behavior Uncloning: Distilling Mode Redirection into Policy Weights without Inference-Time Steering","primary_cat":"cs.RO","submitted_at":"2026-06-28T05:01:27+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"MoRE improves robot policy success rates by 44 percentage points by distilling mode redirection into weights, matching filtered retraining performance without inference overhead.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.26341","ref_index":12,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Scaling Nonlinear Optimization: Many Problems One GPU","primary_cat":"cs.RO","submitted_at":"2026-06-24T19:34:41+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"jaxipm is the first GPU-batched IPOPT solver in JAX using heterogeneous iteration fusion and iteration-level batching, delivering up to 32.85x higher throughput than standard IPOPT on quadrotor NMPC benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.22447","ref_index":7,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A Differentiable Atari VCS:A Complex, Fully Known Ground Truth for Explainable AI","primary_cat":"cs.AI","submitted_at":"2026-06-21T11:46:00+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":8.0,"formal_verification":"none","one_line_summary":"Differentiable reimplementations of the Atari VCS provide a complex, fully known ground-truth system for testing gradient-based explainable AI methods.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.21525","ref_index":17,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Backpropagating Through Simulation: Analytic Policy Gradients for Sample and Learning Efficient Differentiable Continuous Control","primary_cat":"cs.LG","submitted_at":"2026-06-19T15:22:51+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Analytic Policy Gradients enable exact gradient computation via backpropagation through simulation for differentiable continuous control, with segmented backprop to mitigate degradation on long-horizon tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.06094","ref_index":95,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Integrating Mechanistic and Data-Driven Models for Neurological Disorders through Differentiable Programming","primary_cat":"cs.AI","submitted_at":"2026-06-04T12:32:24+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"This perspective paper categorizes hybrid architectures for combining mechanistic and data-driven models using residual learning, Neural ODEs, and solver-in-the-loop to model neurological disorder progression.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.04749","ref_index":25,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"COP-Q: Safety-First Reinforcement Learning for Robot Control via Cholesky-Ordered Projection","primary_cat":"cs.RO","submitted_at":"2026-06-03T11:30:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"COP-Q uses Cholesky-Ordered Projection on joint Q-values to incorporate inter-objective covariance, preserving safety conservatism while improving sample efficiency in robot locomotion and navigation tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.31120","ref_index":31,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"SWIM: Single-Instance Whole-Body Imitation for swiMming","primary_cat":"cs.GR","submitted_at":"2026-05-29T10:30:03+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"SWIM is a single-instance imitation method for learning and generalizing physically simulated swimming motions to new environments, bodies, and styles.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.30542","ref_index":21,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Physically Viable World Models: A Case for Query-Conditioned Embodied AI","primary_cat":"cs.AI","submitted_at":"2026-05-28T20:18:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Embodied AI requires query-conditioned world models that select the simplest physical abstraction sufficient to answer intervention queries.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.30313","ref_index":18,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"UniLab: A Heterogeneous Architecture for Robot RL Beyond GPU-Dominant Paradigms","primary_cat":"cs.RO","submitted_at":"2026-05-28T17:53:50+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"UniLab is a CPU/GPU heterogeneous system for robot RL training using MuJoCoUni and MotrixSim backends that reports 3-10x end-to-end efficiency improvements and cross-platform compatibility beyond CUDA.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.24922","ref_index":3,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"MuJoCoUni:Persistent Batched Runtime Primitives for MuJoCo","primary_cat":"cs.RO","submitted_at":"2026-05-24T07:57:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"MuJoCoUni introduces BatchEnvPool, a C++/pybind11-based executor providing persistent batched stateful MuJoCo environments with domain randomization, sensor queries, and short stepping.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.20576","ref_index":21,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"$\\Delta$ynamics: Language-Based Representation for Inferring Rigid-Body Dynamics From Videos","primary_cat":"cs.CV","submitted_at":"2026-05-20T00:23:56+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A vision-language framework generates text-based rigid-body scene configurations from videos using motion reasoning and optical flow, reporting 0.30 IoU on CLEVRER (7x over baselines) and transfer to 235 real videos.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.19503","ref_index":7,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"ARC-RL: A Reinforcement Learning Playground Inspired by ARC Raiders","primary_cat":"cs.RO","submitted_at":"2026-05-19T07:54:40+00:00","verdict":"ACCEPT","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ARC-RL is a new suite of four MuJoCo continuous-control environments featuring game-inspired hexapod and quadruped morphologies, a single closed-form multi-component reward function, CPG demonstrators, and empirical comparisons of online and offline-to-online RL algorithms.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.16520","ref_index":73,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Global Convergence of Sampling-Based Nonconvex Optimization through Diffusion-Style Smoothing","primary_cat":"cs.LG","submitted_at":"2026-05-15T18:14:38+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Recasts sampling-based nonconvex optimization as smoothed gradient descent to obtain non-asymptotic convergence guarantees and introduces the DIDA annealed algorithm that converges to the global optimum.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"As stated in the theorem,xt∈RSC(t), i.e. ∥x∗−xt∥≤CE min( √ λ β+t λ β , √ τ2 +t τ2 )Dτ (71) This also implies the corresponding bound on the effective meanµ′ in: ∥µ′ in−x∗∥= λ β t+ λ β ∥xt−x∗∥≤∥xt−x∗∥≤CEDτ. Moreover, ∥µ′ in−x∗∥=∥xt−x∗∥ λ β t+ λ β ≤CE √ λ β λ β+t Dτ≤CEDτ,(72) Most importantly, from our assumption, we have β Lt+λ∥xt−x∗∥2≤C2 E β λD2 τ (73) Note that ∫ Bτ N ( y|µ′ in,(σ′ in)2I ) dy=P(∥Z+µ′ in−x∗∥≤Dτ)≥P(∥Z∥≤Dτ−∥µ′ in−x∗∥), where Z∼N(0, (σ′ in)2I). In particular, if we chooseCE≤1 3, then Equation (72) implies∥µ′ in−x∗∥≤1 3Dτ, and henceDτ−∥µ′ in−x∗∥≥2 3Dτ. Therefore, ∫ Bτ N ( y|µ′ in,(σ′ in)2I ) dy≥PZ∼N(0,(σ′ in)2I) ( ∥Z∥≤2 3Dτ ) ≥PZ∼N(0,I) ( ∥Z∥2≤4D2 τ 9(σ′ in)2 ) ≥PZ∼N(0,I) ( ∥Z∥2≤4D2"},{"citing_arxiv_id":"2605.14911","ref_index":10,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Chrono-Gymnasium: An Open-Source, Gymnasium-Compatible Distributed Simulation Framework","primary_cat":"cs.RO","submitted_at":"2026-05-14T14:45:28+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"Chrono-Gymnasium is a Ray-based distributed wrapper that adds a Gymnasium interface to Project Chrono simulations, demonstrated on RL navigation and Bayesian lander optimization.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.14297","ref_index":49,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Policy Optimization in Hybrid Discrete-Continuous Action Spaces via Mixed Gradients","primary_cat":"cs.LG","submitted_at":"2026-05-14T02:59:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"HPO enables unbiased policy optimization in hybrid action spaces by mixing differentiable simulation gradients with score-function estimates, outperforming PPO as continuous dimensions increase.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.16395","ref_index":10,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"OrbiSim: World Models as Differentiable Physics Engines for Embodied Intelligence","primary_cat":"cs.RO","submitted_at":"2026-05-12T13:43:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"OrbiSim builds a differentiable physics engine from world models to support gradient-based policy optimization and contact modeling in robotics.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.09196","ref_index":11,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"RigidFormer: Learning Rigid Dynamics using Transformers","primary_cat":"cs.CV","submitted_at":"2026-05-09T22:31:09+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"RigidFormer learns mesh-free rigid dynamics from point clouds using object-centric anchors, Anchor-Vertex Pooling, Anchor-based RoPE, and differentiable Kabsch alignment to enforce rigidity.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"• We validate RigidFormer across diverse experiments, demonstrating fast inference, generalization, scalability, and a preliminary application to command-conditioned articulated bodies. 2 Related Work Classical numerical rigid-body simulators [ 2, 9, 24, 35] resolve contact by solving constrained optimization or complementarity problems. Differentiable simulators (e.g., DiffTaichi [16], Warp [23], and Brax [11]) enable gradient-based learning and inverse problems, but they rely on explicit physics engines and typically assume mesh-based geometry rather than mesh-free point inputs. Early learning-based dynamics models often targeted relatively simple systems with explicit, low- dimensional state representations, typically in 2D. Interaction Networks [ 3] and Neural Physics"},{"citing_arxiv_id":"2605.05373","ref_index":65,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Neural Co-state Policies: Structuring Hidden States in Recurrent Reinforcement Learning","primary_cat":"cs.LG","submitted_at":"2026-05-06T18:53:33+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Recurrent RL policies can have their hidden states aligned with PMP co-states through a derived loss, yielding robust performance on partially observable control tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.03288","ref_index":2,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Neural Control: Adjoint Learning Through Equilibrium Constraints","primary_cat":"cs.RO","submitted_at":"2026-05-05T02:19:37+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Neural Control uses adjoint differentiation of equilibrium conditions to compute trajectory-dependent proxy gradients for history-dependent implicit models in deformable object manipulation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.01865","ref_index":61,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Quality-Aware Exploration Budget Allocation for Cooperative Multi-Agent Reinforcement Learning","primary_cat":"cs.MA","submitted_at":"2026-05-03T13:20:00+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A quality-aware exploration method using return-conditioned sigmoid scheduling and per-agent RSQ metrics achieves top-tier returns on seven cooperative MARL benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.21456","ref_index":29,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Tempered Sequential Monte Carlo for Trajectory and Policy Optimization with Differentiable Dynamics","primary_cat":"cs.LG","submitted_at":"2026-04-23T09:13:59+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Tempered sequential Monte Carlo samples from a Boltzmann-tilted distribution over controllers to optimize trajectories and policies under differentiable dynamics.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"outperforms all baselines and is the only method that reaches the goal pose. Fig. 2(e) visualizes the best-cost trajectories. VI. E XPERIMENTS : P OLICY OPTIMIZATION We next turn to policy optimization benchmarks. We ini- tially aimed to evaluate on MuJoCo locomotion tasks ( e.g., HalfCheetah and Ant), which involve contact dynamics. While both MuJoCo JAX (MJX) [98] and Brax [29] provide JAX- based implementations, we found MJX's contact solver relies on a while loop that prevents reverse-mode differentiation, as documented in GitHub issues [1, 96]. Brax supports dif- ferentiable contact, but for HalfCheetah we observed gradient norms as large as 109 to 1013, which makes it unsuitable for our framework. We therefore benchmark on classical control"},{"citing_arxiv_id":"2604.20381","ref_index":15,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Distributional Value Estimation Without Target Networks for Robust Quality-Diversity","primary_cat":"cs.LG","submitted_at":"2026-04-22T09:31:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"QDHUAC is a distributional, target-free QD-RL method that enables stable high-UTD training and competitive performance on Brax locomotion tasks using far fewer environment steps than prior approaches.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.18161","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Does \"Do Differentiable Simulators Give Better Policy Gradients?'' Give Better Policy Gradients?","primary_cat":"cs.LG","submitted_at":"2026-04-20T12:23:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"In policy gradient RL, careful variance control and simple estimator switching frequently outperform explicit discontinuity detection even when using differentiable simulators.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.10974","ref_index":3,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Robust Adversarial Policy Optimization Under Dynamics Uncertainty","primary_cat":"cs.LG","submitted_at":"2026-04-13T04:23:54+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"RAPO uses a dual robust RL formulation with trajectory-level adversarial networks and model-level Boltzmann reweighting over dynamics ensembles to improve policy resilience and out-of-distribution generalization while keeping the problem tractable.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2603.12145","ref_index":1,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Automatic Generation of High-Performance RL Environments","primary_cat":"cs.LG","submitted_at":"2026-03-12T16:45:47+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Closed-loop prompt-based translation with hierarchical verification and iterative repair produces equivalent high-performance RL environments across five cases including new TCGJax.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2602.12633","ref_index":15,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Real-to-Sim for Highly Cluttered Environments via Physics-Consistent Inter-Object Reasoning","primary_cat":"cs.RO","submitted_at":"2026-02-13T05:24:58+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A differentiable optimization pipeline uses a contact graph and rigid-body simulation to jointly refine object poses and physical properties, producing physically valid 3D scene reconstructions from single-view RGB-D observations for cluttered environments.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2506.14009","ref_index":20,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"GRaD-Nav++: Vision-Language Model Enabled Visual Drone Navigation with Gaussian Radiance Fields and Differentiable Dynamics","primary_cat":"cs.RO","submitted_at":"2025-06-16T21:12:27+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"GRaD-Nav++ combines 3D Gaussian Splatting simulation and differentiable RL to train an onboard VLA policy that achieves 50-83% success on language-guided drone navigation tasks in simulation and real hardware.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}