arXiv:2504.18522v3 Announce Type: replace-cross
Abstract: We consider the problem of modeling the effects of perturbations like gene knockouts on measurements such as single-cell RNA counts. Given data for some perturbations, we aim to predict the distribution of measurements for new combinations of perturbations. To address this challenging extrapolation task, we posit that perturbations act additively in a suitable, unknown embedding space. We formulate the data-generating process as a latent variable model, in which perturbations amount to mean shifts in latent space and can be combined additively. We then prove that, given sufficiently diverse training perturbations, the representation and perturbation effects are identifiable up to orthogonal transformation and use this to derive extrapolation guarantees for unseen perturbations that can be expressed as linear combinations of seen ones. To estimate the model from data, we propose the perturbation distribution autoencoder (PDAE), which is trained by maximizing the distributional similarity between true and simulated perturbation distributions. The trained model can then be used to predict previously unseen perturbation distributions. In support of our theoretical results, we demonstrate through simulations that PDAE can accurately predict the effects of unseen but identifiable perturbations, and showcase the method on combinatorial gene perturbation data.
Science Journals
arXiv:2506.00182v3 Announce Type: replace-cross
Abstract: Will further scaling up of machine learning models continue to bring success? A significant challenge in answering this question lies in understanding generalization gap, which is the impact of overfitting. Understanding generalization gap behavior of increasingly large-scale machine learning models remains a significant area of investigation, as conventional analyses often link error bounds to model complexity, failing to fully explain the success of extremely large architectures. This research introduces a novel perspective by establishing a model-independent upper bound for generalization gap applicable to algorithms whose outputs are determined solely by the data's histogram, such as empirical risk minimization or gradient-based methods. Crucially, this bound is shown to depend only on the R\'enyi entropy of the data-generating distribution, suggesting that a small generalization gap can be maintained even with arbitrarily large models, provided the data quantity is sufficient relative to this entropy. This framework offers a direct explanation for the phenomenon where generalization performance degrades significantly upon injecting random noise into data, where the performance degrade is attributed to the consequent increase in the data distribution's R\'enyi entropy. Furthermore, we adapt the no-free-lunch theorem to be data-distribution-dependent, demonstrating that an amount of data corresponding to the R\'enyi entropy is indeed essential for successful learning, thereby highlighting the tightness of our proposed generalization bound.
arXiv:2605.09403v2 Announce Type: replace
Abstract: Architectural choices inside the Transformer feedforward network (FFN) block do not merely affect the block itself; they reshape the computations learned by the rest of the model. We study this effect in one-layer Transformers trained on digit addition with carry, modular arithmetic, and histogram counting. Comparing dense FFNs, gated linear units (GLUs), mixture-of-experts (MoE), and MoE-GLUs, we find that sparse MoE routing can shift computation from FFN to attention, with the strongest ablation-visible effect on carry-based addition. We decompose this redistribution into reduced per-token FFN capacity and sparse partitioning across experts. Critically, frozen random routing nearly matches learned routing, suggesting that redistribution is driven largely by architectural sparsity rather than router-learned specialization. As a secondary finding, GLU-style multiplicative gating rotates task-relevant Fourier structure out of the per-neuron basis and into distributed subspaces, making neuron-level interpretability less informative while preserving structured computation. We validate these conclusions with random-routing, narrow-FFN, and top-2 MoE controls, plus parameter-matching, activation-function, and width-scaling analyses. Together, these results show that local FFN design choices can have nonlocal consequences for Transformer computation.
arXiv:2605.15603v1 Announce Type: new
Abstract: Model-based reinforcement learning (RL) offers a compelling approach to offline RL by enabling value learning on imagined on-policy trajectories. However, it often suffers from compounding errors due to repeated model inference on self-generated states. While geometric horizon models (GHM) alleviate this issue through direct prediction over a discounted infinite-horizon future, they remain challenged in accurately modeling distant future states. To this end, we introduce universal horizon models (UHM), a generalization of GHM that directly predicts future states under arbitrary horizons. Leveraging this flexibility, we propose a scalable value learning method that employs a winsorized horizon distribution to stabilize training by capping excessively large horizons. Experimental results on 100 challenging OGBench tasks demonstrate that the proposed method outperforms competitive baselines, particularly on tasks with highly suboptimal datasets and those requiring long-horizon reasoning. Project page: https://rllab-snu.github.io/projects/UHM/
arXiv:2605.15615v1 Announce Type: new
Abstract: Efficient transfer learning of vision-language models (VLMs) commonly suffers from a Base-New Trade-off (BNT): improving performance on unseen (new) classes often degrades accuracy on known (base) classes. Addressing how to boost recognition of unseen classes without sacrificing known-class performance remains a central challenge. Existing work often simplistically attributes the BNT to overfitting on known classes. We observe an interesting phenomenon: VLMs frequently exhibit asymmetric confusion on certain downstream data, i.e., samples of class A are systematically mispredicted as class B, while the reverse confusion (B to A) rarely occurs. For known classes, this kind of bias can be mitigated by tuning using a cross-entropy loss, but for unseen classes, such pretraining-induced bias persists and harms generalization. Motivated by this, we propose NeRP, a plug-and-play prompting correction strategy that improves discrimination on unseen classes without modifying model parameters. NeRP leverages neutral text prompts and reference images to measure class-wise prior preferences along the pre-trained inter-class geometry, and combines them with the sample likelihood to obtain the model's surrogate score. If, for a given sample, the prior strongly favors the current prediction while the observed evidence is clearly insufficient, we perform a local flip between easily confusable class pairs, thereby correcting prior-dominated mispredictions. Extensive experiments across multiple backbones and 15 few-shot and cross-domain benchmarks show that NeRP substantially improves accuracy on unseen classes while preserving known-class prediction performance.
arXiv:2605.15855v1 Announce Type: new
Abstract: Despite strong image-generation performance, diffusion models' reconstruction objectives limit alignment with human preferences. RL enables such alignment through explicit rewards. However, most studies apply RL to the full denoising trajectory, making it computationally costly and weakening preference alignment, i.e., doing more but achieving less. We observe that the impact of RL fine-tuning varies significantly across denoising stages. In the early stage, image structures are unstable and distant from the final reward signal. Applying RL at this stage leads to delayed rewards and action-reward mismatching, resulting in high variance and inefficient updates. Conversely, in the later stage, reward gains saturate, and continued training tends to overfit local details, intensifying reward hacking. To tackle these challenges, we propose AdaScope, an RL-enhanced plug-in that improves generation quality while reducing computational cost. Specifically, AdaScope adaptively identifies the optimal intervention timing for RL by perceiving the structural evolution and semantic consistency during denoising, and dynamically terminates training once the denoising converges and reward gains saturate. As a result, it achieves a rare 'dual benefit': a reduction in computational costs alongside a significant performance improvement. We offer theoretical grounds for the design of AdaScope. Compared with state-of-the-art methods, AdaScope improves performance by 66% while cutting computational cost by 59%.
arXiv:2605.15498v1 Announce Type: new
Abstract: From a new perspective, this paper rederives Lagrange's equations. By applying the chain rule of differentiation, the intrinsic relationship between the momentum theorem and the kinetic energy theorem is first established. Subsequently, expressing the differential form of energy conservation in an arbitrary coordinate system and performing suitable differential operations yields Lagrange's equations. Generalized forces and generalized displacements are shown to be component representations of forces and displacements in a chosen coordinate system. Consequently, the essence of Lagrange's equations is identified as the transformation of the kinetic energy theorem into the momentum theorem via the chain rule for composite functions, thereby revealing how energy conservation constructs momentum conservation.
arXiv:2605.15370v1 Announce Type: cross
Abstract: Accurate salt-body delineation is essential for seismic interpretation because salt structures distort wave propagation, complicate velocity-model building, obscure reservoir geometry, and increase uncertainty in exploration and drilling decisions. Although hybrid quantum-classical models have shown competitive performance on small-scale image-classification tasks, their value for dense, pixel-level geophysical prediction remains largely untested. This work introduces quantum feature gating, a hybrid segmentation architecture that embeds a parameterized quantum circuit (PQC) at feature-fusion points within an encoder-decoder pipeline. A 4-qubit, 2-layer PQC with data re-uploading computes a learned convex combination of lateral and top-down features at each Feature Pyramid Network merge point. A global-average-pooling layer maps encoder features to a fixed 4-dimensional quantum input, decoupling the 72-parameter quantum budget from backbone size and image resolution. The method is evaluated on the 2018 TGS Salt Identification Challenge using 4,000 seismic images at 101 x 101 resolution, across two integration topologies, eight circuit variants, and six encoders with 8M to 118M parameters under five-fold cross-validation. In a controlled EfficientNetV2-L ablation at 256 x 256 resolution, replacing the three Quantum FPN Gates with element-wise addition while holding the encoder, loss schedule, splits, and threshold search fixed reduces mean IoU from 0.9389 to 0.8404, a 9.85 percentage-point gap. Inserting the same circuit as skip-connection attention in a custom U-Net improves IoU by 0.88 points over the SolidUNet baseline, showing that the PQC contribution depends on where and what it gates. These results provide controlled evidence that quantum feature fusion can improve dense seismic segmentation.
arXiv:2605.15366v1 Announce Type: cross
Abstract: We present hybrid quantum-classical pipelines for solving the Duffing equation that leverage Carleman linearization and the Variational Quantum Linear Solver (VQLS). First, we demonstrate that Carleman linearization accurately approximates the weakly nonlinear Duffing equation, with errors diminishing as the truncation order increases. Next, across IBM and Xanadu platforms, we deploy VQLS with symmetry-grouped Hadamard Test evaluations under both global and local cost formulations, compare distinct Hermitianization within a common cost framework, and benchmark hardware-efficient ansatz architectures under a fixed Hermitianization. Across block-banded test cases, each method achieves near-unity fidelity and vanishing relative residuals. These results show that topology-agnostic ansatz, optimized Hermitianization, and efficient cost formulation enable VQLS to recover quantum states proportional to classical solutions for Carleman-structured systems, providing a portable recipe for quantum-in-the-loop simulation of nonlinear dynamics.
arXiv:2605.15350v1 Announce Type: cross
Abstract: Stochastic compositional optimization minimizes objectives of the form $\min_{\bm{x} \in \mathcal{X}} F(\bm{f}(\bm{x}), \bm{x})$, where $\bm{f}$ is accessible only through noisy stochastic queries. Existing methods for this problem assume that the outer function $F$ is continuously differentiable, which excludes many practically important applications such as robust max-of-losses, Conditional Value-at-Risk, and norm regularizers. We propose the Hybrid Momentum Stochastic Frank--Wolfe algorithm, which drops the smoothness assumption on $F$. By combining a momentum-based Jacobian tracker with a Taylor-corrected function tracker, the algorithm feeds an entire stochastic linearization -- rather than a single gradient -- into a generalized linear minimization oracle. We establish an $\mathcal{O}(K^{-1/4})$ convergence rate in the generalized Frank--Wolfe gap for non-convex objectives with $L_F$-Lipschitz outer functions, matching the optimal complexity for projection-free single-sample stochastic methods under expected smoothness. The analysis extends to heavy-tailed noise oracles with bounded $r$-th moments for $r \in (1, 2]$ and recovers the deterministic rates of Vladarean et al (2023) as the noise vanishes.
arXiv:2605.15496v1 Announce Type: new
Abstract: Neural distance fields offer a compact and continuous representation of 3D geometry, making them attractive for incremental LiDAR mapping. However, their online optimization is vulnerable to catastrophic forgetting, where new observations can degrade previously reconstructed geometry. Replay-based training is commonly used to address this issue, but existing methods typically rely on passive replay buffers and uniform sampling, which can waste memory on redundant observations and under-train poorly constrained regions. We propose LAPS, a replay management framework for incremental neural mapping that improves both replay retention and replay allocation during online updates. LAPS combines reliability-based active pooling to retain reliable historical samples under limited memory with uncertainty-guided active sampling to focus optimization on under-constrained regions. Experiments on synthetic and real-world benchmarks show that LAPS consistently improves reconstruction completeness while maintaining competitive geometric accuracy. On Oxford Spires, it improves recall by 4.66 pp and F1-score by 3.79 pp over PIN-SLAM on the Blenheim Palace 05 sequence. We release our open source implementation at: https://github.com/dongjae0107/LAPS.
arXiv:2605.15492v1 Announce Type: new
Abstract: Generative models such as diffusion and flow matching have become dominant paradigms for visuomotor policy learning, yet their reliance on iterative denoising incurs high inference latency incompatible with real-time robotic control. We present Fast Legendre-polynomial Action policy via Sparse History-anchored flow (FLASH Policy), which replaces discrete action-chunk generation with continuous Legendre polynomial trajectory representation. Specifically, by fitting expert demonstrations under sparse temporal sampling, FLASH enables a single inference to cover a significantly extended action horizon. To further accelerate generation, FLASH initiates the flow matching process from history polynomial coefficients rather than uninformative Gaussian noise, shortening the transport distance and enabling accurate single-step inference. Moreover, analytic polynomial differentiation directly provides desired velocity feed-forward signals to the torque controller without numerical approximation. Extensive experiments on five simulated and two real-world manipulation tasks demonstrate that FLASH achieves state-of-the-art success rates ($\ge 92\%$ across all tasks), a per-episode inference time of $31.40\,ms$ (up to $175\times$ faster than diffusion policies and $18\times$ faster than prior flow matching policies), up to $4\times$ faster training convergence than ACT, and $5\times$ to $7\times$ reduction in controller tracking error compared to discrete-action baselines.
arXiv:2605.15266v1 Announce Type: cross
Abstract: Preparing arbitrary logical states is a central primitive for universal fault-tolerant quantum computation and the cost of encoded-state preparation contributes directly to the overall resource overhead. This makes the synthesis of efficient general-state encoding circuits an important problem, particularly with respect to two-qubit gate count and circuit depth. Yet the synthesis of such encoders has been studied less extensively than general Clifford circuit synthesis or the preparation of specific logical Pauli-eigenstates. In this work, we develop methods for synthesizing efficient encoders for arbitrary stabilizer codes. We formulate encoder synthesis as a search over stabilizer tableaus and introduce greedy and rollout-based algorithms that exploit the freedom among stabilizer-equivalent realizations of the same encoding isometry. For code families with a modular structure, such as generalized concatenated and holographic codes, we show how large encoders can be assembled from optimized local constituent encoders, and we use SMT-based exact synthesis to obtain optimal local circuits for small instances. We further evaluate the proposed methods on a broad set of stabilizer codes, including holographic and quantum low-density parity-check (qLDPC) codes, and compare them against recent encoder-synthesis methods and existing constructions from the literature, obtaining improvements of up to 43% in two-qubit gate count and up to 70% in depth. Our results support the optimization of encoded-state preparation in several fault-tolerant quantum-computing schemes, and all methods are openly available as part of the Munich Quantum Toolkit.
arXiv:2605.15616v1 Announce Type: new
Abstract: In this work, we consider the One-Fluid Two-Temperature Euler (OFTT-Euler) equations used for modeling non-equilibrium hydrodynamics. The model comprises a system of nonlinear hyperbolic partial differential equations with non-conservative products. The model decomposed the total pressure into two scalar components: one for electrons and one for ions. Our aim in this work is to design entropy-stable finite difference numerical schemes for the model. This is achieved by reformulating the equations such that the reformulated non-conservative part does not contribute to the entropy. Then, we design higher-order entropy-conservative numerical schemes by using Tadmor's relation for the conservative part and higher-order central differences for the non-conservative parts. Finally, we design the entropy-dissipation terms using the entropy-scaled right eigenvectors of the conservative part, thereby deriving the entropy inequality for the entire system. We present several test cases in one and two dimensions to demonstrate the accuracy and stability of the proposed schemes.
arXiv:2605.15617v1 Announce Type: new
Abstract: Large language model (LLM) training today runs on clusters spanning thousands of GPUs. While this scale enables rapid model advances, developing, debugging, and performance-tuning the training framework inevitably becomes complex and costly. This is because engineers often need to reproduce production behaviors to diagnose failures or evaluate optimizations, thereby demanding frequent and even exclusive access to production-scale clusters -- which becomes increasingly hard given that the majority of GPUs are already committed to production workloads. Simulation relies on complex performance models that are difficult to maintain, and downscaled experiments often fail to capture scale-dependent behaviors.
We present PrismLLM to decouple large-scale execution from the need to access large clusters, enabling engineers to run and observe ranks of interest under faithful large-scale behavior using only a few GPUs. PrismLLM constructs a high-fidelity execution graph via a slicing-based approach that captures computation, communication, and dependencies of the target scale. Then, PrismLLM performs hybrid emulation where selected ranks execute the original program while the remaining ranks are replayed as virtual participants.
Experiments on large-scale LLM training workloads show that PrismLLM accurately reproduces performance and memory behavior, achieving only 0.58\% average error in iteration time and less than 0.01\% error in peak GPU memory usage. PrismLLM can emulate clusters of up to 8192 GPUs using fewer than 1\% of the physical GPUs required by the original deployment.
arXiv:2605.15887v1 Announce Type: new
Abstract: Robust aggregation is the core operation in Byzantine-tolerant federated learning. To ensure the quality of aggregation independently of data distribution or attacks, validity conditions are needed. They provide geometric guarantees of where the output of the aggregation must lie. The widespread convex validity requires the output to lie in the convex hull of the honest vectors. Although this guarantee is strong in theory, it is poorly suited to modern federated learning systems, as it has dimension-dependent resilience and excludes many practical aggregation rules.
We introduce the minimum enclosing ball (MEB) validity condition for robust aggregation, as well as its multiplicative relaxation, $c$-MEB validity, where $c$ is a constant. We show that exact MEB validity still suffers from limited resilience, while relaxed $c$-MEB validity is achievable if a majority of clients is honest, i.e. $n>2t$. We give an optimal MinMax-MEB rule for the relaxed condition with the bound $c<\sqrt{2}$ and prove explicit relaxed-MEB guarantees for standard aggregators including minimum-diameter averaging, medoid and geometric median. Finally, we relate MEB validity to convex, relaxed-convex and box validity studied in prior literature, thus providing a systematic map of geometric validity conditions for Byzantine-robust aggregation. Our results show that relaxed MEB validity connects validity conditions in distributed computing and Byzantine-tolerant aggregation rules, and offers a practical alternative to convex validity.
arXiv:2505.21535v4 Announce Type: replace
Abstract: While transformers dominate modern vision and language models, their attention mechanism remains poorly suited for in-memory computing (IMC) devices due to intensive activation-to-activation multiplications and non-local memory access, leading to substantial latency and bandwidth overhead on ReRAM-based accelerators. To address this mismatch, we propose FAR, a Function-preserving Attention Replacement framework that substitutes all attention in pretrained DeiTs with sequential modules inherently compatible with IMC dataflows. Specifically, FAR replaces self-attention with a multi-head bidirectional LSTM architecture via block-wise distillation to retain functional equivalence while enabling linear-time computation and localized weight reuse. We further incorporate structured pruning on FAR models, enabling flexible adaptation to resource-constrained IMC arrays while maintaining functional fidelity. Evaluations on the DeiT family demonstrate that FAR maintains comparable accuracy to the original attention-based models on ImageNet and multiple downstream tasks with reduced parameters and latency. Further analysis shows that FAR preserves the semantic token relationships learned by attention while improving computational efficiency, highlighting its potential for energy-efficient transformer inference on IMC-based edge accelerators.
arXiv:2605.15240v1 Announce Type: cross
Abstract: This paper investigates the critical role of eigenalignments between the kernel matrix and learning targets in achieving robust generalization in learning problems. We establish a direct connection between generalization performance in kernel methods and the estimation of eigenvectors and eigenvalues of matrices, offering a more intuitive understanding compared to prior work with minimal assumptions. We also show that, since the prediction task in KRR is essentially the weighted sum of eigenvectors/singular vectors, by analyzing how much error can be caused by perturbations to the kernel matrix, we can then derive a bound on this generalization error using the estimation stability of matrix eigenvalues and eigenvectors. Compared with previous work, our analysis concentrates on finite-sample settings and on the generalization error arising from having a suboptimal finite training set. Our findings reveal that in kernel methods, as long as the kernel is of high rank, the near-zero reconstruction error can be trivially obtained, implying that the reconstruction error will have limited predictive power for generalization. Finally, we establish a generalization bound from an eigenvalues/eigenvectors estimation perspective, showing that strong generalization requires increasing eigenvector alignment, eigenvalue magnitude, or gaps between consecutive eigenvalues.
arXiv:2605.15225v1 Announce Type: cross
Abstract: Biologically-inspired AI agent frameworks claim reliability benefits through structural guarantees adapted from gene regulatory networks, immune systems, and metabolic control. These claims are rarely tested empirically against simpler alternatives. We present three deep benchmarks: metabolic priority gating, autoinducer-based quorum sensing, and Bayesian stagnation detection, each comparing a biologically-grounded implementation against a naive non-biological alternative and an ablated control, across 1,000 trials per seed and 10 seeds (10M+ data points total).
arXiv:2605.12581v1 Announce Type: cross
Abstract: Synthesising autonomous agents that can navigate uncertain environments while adhering to complex temporal constraints remains a fundamental challenge. While Linear Temporal Logic (LTL) provides a rigorous language for specifying such tasks, the inherent undecidability of qualitatively verifying LTL satisfaction in partially observable Markov decision processes renders quantitative synthesis difficult, especially when designing reliable reward signals for approximate solvers. In this paper, we bridge this gap with a novel, sound reward-shaping mechanism that dynamically generates belief-dependent rewards grounded in certified LTL satisfaction. By integrating this mechanism into an enhanced Monte Carlo Planning framework, we empower agents to navigate the `fog' of partial observability with a search process focused on maximising verifiable success. Our experiments demonstrate that this approach not only thrives in scenarios where existing solvers fail but also maintains effectiveness and scalability across diverse benchmark domains.
arXiv:2605.13246v2 Announce Type: replace
Abstract: Reference counting bugs in Linux kernel drivers can lead to severe resource mismanagement and security vulnerabilities. We introduce DrvHorn, a novel automated tool to detect these bugs by reducing reference counting verification to an assertion checking problem leveraging the Linux driver interface. Through efficient modeling of the Linux kernel and aggressive program slicing, DrvHorn discovered 545 bugs, of which 424 were previously unknown, across all platform drivers in v6.6 Linux kernel, with a lower false positive rate of 29.9% compared to prior studies. To address the root causes of these newly discovered bugs, we submitted patches to the Linux kernel, and 45 of them were merged.
arXiv:2605.13788v2 Announce Type: replace
Abstract: Active learning for machine-learning interatomic potentials (MLIPs) must address several challenges to be practical: scaling to large candidate pools, leveraging energy-force supervision, and maintaining robustness when candidate pools are biased relative to the target distribution. In this work, we jointly address these challenges. We first introduce a linearly scaling acquisition framework based on chunked feature-space posterior-variance shortlisting. By avoiding materialisation of the candidate and train set kernels, this approach enables screening of ~200k structures within hours and applies broadly to acquisition strategies that score candidates based on molecular similarity metrics. We then extend the Neural Tangent Kernel (NTK) to a force-aware setting via mixed parameter-coordinate derivatives, yielding a force NTK and a joint energy-force NTK that provide natural similarity metrics for vector-field prediction. We demonstrate the effectiveness of the joint energy-force NTK on the OC20 dataset, where force-aware acquisition is crucial: it achieves the lowest energy and force MAE and RMSE across all metrics and distribution splits. Across T1x, PMechDB, and RGD benchmarks, our force NTK methods remain competitive with established baselines while being significantly more efficient than committee-based approaches. Under a controlled candidate-pool shift case study on T1x, acquisition based on pretrained MLIP embeddings and NTKs remains robust, whereas committee-based methods exhibit higher variance. Overall, these results show that a single pretrained MLIP can enable scalable, force-aware, and distribution-robust active learning for foundation-model fine-tuning.
arXiv:2605.15534v1 Announce Type: cross
Abstract: In this work, we study stochastic one-shot games where agents' utilities depend on the collective strategy profiles of other agents as well as on some well-behaved randomness. While each decision-maker is agnostic to the random variable's underlying distribution, they have access to finitely many i.i.d. samples generated from it. We consider two cases: one where samples are shared; and another, more special one, where samples are individually accessible. To hedge against the unknown uncertainty, each agent plays a distributionally robust game and aims to maximize the worst-case expected utility over a Wasserstein ball around the sample average distribution. In this setting, we provide conditions under which the game has a non-empty set of distributionally robust Nash equilibria (DRoNE) and then characterize the closeness of the DRoNE set to the Nash equilibria (NE) of the associated stochastic game. We then propose an inertial, supported, better response, ascending supergradient dynamics ISBRAG that seeks the DRoNE's when the distributionally robust game possesses what we term as amicable supergradients. This forms the basis of a distributed version (d-ISBRAG) where agents estimate others' strategies by means of a dynamic consensus subroutine over a directed communication network. While initially the distributed algorithm works in the case where agents have individual samples, we later extend this to the case of shared observations under certain simplifying assumptions. This involves analyzing a tractable reformulation of the distributionally robust optimization problem and solving it in a distributed manner to compute the required supergradients. Simulations illustrate our results.
arXiv:2604.04539v2 Announce Type: replace
Abstract: Reinforcement learning (RL) is a core approach for robot control when expert demonstrations are unavailable. On-policy methods such as Proximal Policy Optimization (PPO) are widely used for their stability, but their reliance on narrowly distributed on-policy data limits accurate policy evaluation in high-dimensional state and action spaces. Off-policy methods can overcome this limitation by learning from a broader state-action distribution, yet suffer from slow convergence and instability, as fitting a value function over diverse data requires many gradient updates, causing critic errors to accumulate through bootstrapping. We present FlashSAC, a fast and stable off-policy RL algorithm built on Soft Actor-Critic. Motivated by scaling laws observed in supervised learning, FlashSAC sharply reduces gradient updates while compensating with larger models and higher data throughput. To maintain stability at increased scale, FlashSAC explicitly bounds weight, feature, and gradient norms, curbing critic error accumulation. Across over 60 tasks in 10 simulators, FlashSAC consistently outperforms PPO and strong off-policy baselines in both final performance and training efficiency, with the largest gains on high-dimensional tasks such as dexterous manipulation. In sim-to-real humanoid locomotion, FlashSAC reduces training time from hours to minutes, demonstrating the promise of off-policy RL for sim-to-real transfer.
arXiv:2605.09869v2 Announce Type: replace
Abstract: Zero-shot object navigation has advanced rapidly with open-vocabulary detectors, image--text models, and language-guided exploration. However, even after current methods detect a plausible target hypothesis, the agent may still oscillate between exploration and pursuit, or abandon the object near success. We identify this failure mode as an action consistency gap: semantic evidence is repeatedly reinterpreted at each step without persistent commitment across the episode. We introduce ConsistNav, a training-free zero-shot ObjectNav framework built around a semantic executive composed of three coordinated modules: Finite-State Executive Controller stages target pursuit through guarded semantic phases; Persistent Candidate Memory accumulates cross-frame target evidence into stable object hypotheses; and Stability-Aware Action Control suppresses rotational stagnation, ineffective pursuit, and unverified stopping. This design changes neither the detector nor the low-level planner; instead, it controls when semantic evidence should influence navigation and when it should be suppressed or revisited. We conduct extensive experiments on HM3D and MP3D, where ConsistNav achieves state-of-the-art results among compared zero-shot ObjectNav methods and improves SR by 11.4% and SPL by 7.9% over the controlled baseline on MP3D. Ablation studies and real-world deployment experiments further demonstrate the effectiveness and robustness of the proposed executive mechanism.