Hy-Embodied-0.5-VLA: From Vision-Language-Action Models to a Real-World Robot Learning Stack

In this report, we present Hy-Embodied-0.5-VLA, abbreviated as HyVLA-0.5, an end-to-end system that spans the full robot learning stack: data collection, model design, continued pre-training and supervised fine-tuning, RL post-training, and real-world deployment. Each component serves a distinct role in this stack.

Direct/adaptive-mixture phase-gradient learning for neural-network quantum states with complex phase structure

Neural-network quantum states (NQS) are a leading variational tool for quantum many-body physics, yet their optimization is fragile whenever the ground state carries a non-trivial sign or complex phase structure, a situation generic to gauge fields, broken time-reversal symmetry, and fermionic statistics. We trace this fragility to the stochastic estimator of the phase gradient rather than to network expressiveness. The phase sector of the Monte Carlo energy gradient is a noisy score-function estimator; differentiating the local energy instead yields a direct estimator that is unbiased for the same phase force, has far lower variance, and requires only a separated amplitude--phase ansatz. Demonstrated on a 100-site flux ladder, a small network trained this way reaches $0.89\%$ median error, where tuned standard baselines plateau at $1.8\%$ and wider or deeper standard-gradient networks degrade from $8.4\%$ to $24.6\%$. The advantage carries over to chiral XXX chains: the direct estimator again converges to a markedly lower error than the standard one, across $α$ and size; it grows with flux and vanishes in zero-flux controls. An adaptive-mixture of the two estimators is provably never worse in variance than the better endpoint at the optimal mixing coefficient, with seed-resolved diagnostics tracing much of the gain to eliminating failed runs. Estimator design thus emerges as a first-class lever for complex-valued neural quantum states.

PBSD: Privileged Bayesian Self-Distillation for Long-Horizon Credit Assignment

Long-horizon agentic tasks pose a fundamental credit assignment challenge for outcome-base reinforcement learning: trajectory-level rewards verify final correctness but provide limited guidance on which intermediate reasoning steps or tool interactions contribute to the outcome. The difficulty is especially pronounced in multi-turn search agents, where successful trajectories may contain misleading actions and failed trajectories may contain valuable evidence-gathering steps. We propose PBSD (Privileged Bayesian Self-Distillation), a Bayes-calibrated self-distillation method for fine-grained credit assignment under sparse final rewards. PBSD measures trajectory quality through the posterior-to-prior probability ratio of the verified answer and applies Bayes' rule to convert this hard-to-estimate answer-side ratio into a tractable likelihood ratio between a standard student model and a privileged answer-conditioned teacher model. Autoregressive decomposition of this Bayesian evidence score yields turn-level signals that identify whether each intermediate turn supports or undermines the verified outcome. Consequently, PBSD provides a principled and elegant reweighting scheme that transforms sparse outcome supervision into Bayes-calibrated turn-level credit signals, while remaining fully compatible with standard policy optimization. Experiments demonstrate that PBSD consistently enhances performance across both in-domain and out-of-domain settings, and effectively transfers knowledge from short-context training to long-context inference, suggesting that its fine-grained credit assignment mechanism facilitates more effective policy learning and yields improved generalization.

AlignFed: Alignment-Aware Asynchronous Federated Fine-Tuning for Large Language Models in Heterogeneous Edge Environments

Large Language Models (LLMs) have significantly propelled the advancement of edge intelligence and have been widely deployed across various scenarios, including autonomous driving, industrial inspection, and personalized IoT services. However, the collaborative adaptation of LLMs on edge devices continues to face formidable challenges due to strict data privacy constraints, highly heterogeneous computing and communication resources, and the non-independent and identically distributed (non-IID) nature of local data. Federated Fine-Tuning (FFT) enables the collaborative optimization of distributed models without exposing raw data. Yet, traditional synchronous aggregation suffers from a severe straggler effect, resulting in high system latency and low resource utilization. Existing asynchronous federated learning methods are predominantly designed for small-to-medium-scale models and struggle to address the specific challenges inherent in LLM fine-tuning namely, model drift caused by stale updates, aggravated client drift stemming from data heterogeneity, and aggregation fairness imbalance resulting from the dominance of fast clients. To address these issues, this paper proposes AlignFed, an asynchronous federated fine-tuning framework for LLMs tailored to heterogeneous edge environments. AlignFed employs a lightweight multi-stage semantic alignment mechanism comprising three core modules: version-aware update grouping, cross-version semantic alignment based on a mini-batch calibration set, and fairness-aware aggregation that integrates both update freshness and client participation frequency. This framework effectively mitigates cross-version model drift and client drift while enhancing aggregation fairness, thereby achieving stable and efficient asynchronous federated optimization in scenarios characterized by high heterogeneity and significant update staleness.

A Sliced-Wasserstein Framework on Correlation Matrices for EEG Decoding

Electroencephalography (EEG) offers noninvasive, millisecond resolution recordings of neuronal activity and is widely used in neuroscience and healthcare. Many EEG decoding pipelines rely on covariance descriptors for their robustness to noise, but such representations are sensitive to channel-wise scaling. Recent studies have therefore advocated full-rank correlation matrices as a scale-invariant alternative for EEG decoding. In this paper, we propose a general framework for Sliced Wasserstein (SW) discrepancies on manifolds endowed with Pullback Euclidean Metrics (PEMs), termed Pullback Euclidean Metric Sliced Wasserstein (PEMSW). Within this framework, we instantiate two Correlation Sliced-Wasserstein (CorSW) discrepancies on the manifold of full-rank correlation matrices under two recently introduced correlation geometries, \textit{i.e.}, the Off-Log Metric (OLM) and Log-Scaled Metric (LSM). Building on CorSW, we further develop a domain generalization (DG) framework for EEG decoding. Experiments on three EEG datasets demonstrate improved generalization under distribution shifts, with low training overhead and no additional inference cost. The source code is available at https://github.com/ChenHu-ML/CorSW.

LLMCodec: Adapting Video Codecs for Efficient Weight Compression of Large Language Models

The rapid development of large language models(LLMs) has led to remarkable advances in natural language processing. However, the increasing scale of these models introduces substantial challenges in terms of storage, transmission, and deployment. Though great efforts have been devoted to model compression and quantization, existing methods often rely on fine-tuning or calibration data, which exhibit limited generalization across different tensor types. In this paper, we argue that video codecs offer a promising solution for LLM compression, due to their inherent compatibility with matrix structured data, configurable compression strategies, and the availability of highly optimized, off-the-shelf implementations. Therefore, we present LLMCodec, a video codec-based LLM compression method that integrates affine quantization with the recent VVC/H.266 video codec. Beyond VVC, we further compare a range of video codecs and encoding profiles to evaluate their impact on compression performance. Experiments on different models demonstrate the robustness and generality of LLMCodec. Notably, on LLaMA-3-8B at 2-bit precision, LLMCodec reduces perplexity by over 1.5x and improves downstream task accuracy by 21% compared with the existing method.

TPA-AD: A Two-Stage Pseudo Anomaly-Guided Method for Bearing Time-Series Anomaly Detection

This paper proposes a two-stage pseudo anomaly-guided anomaly detection method (\textbf{T}wo-stage \textbf{P}seudo \textbf{A}nomaly-guided \textbf{A}nomaly \textbf{D}etection, \textbf{TPA-AD}) for axle-box bearing time-series anomaly detection (time series anomaly detection, TSAD) under the setting where only normal samples are available for training. The method first generates pseudo-anomalous windows near the normal boundary using a reconstruction model and per-feature target-error control. It then learns anomaly-sensitive representations through contrastive learning between normal and pseudo-anomalous windows, and finally produces window-level and point-level anomaly scores using k-nearest neighbors (KNN). Compared with existing methods that rely on known fault categories, real anomaly priors, or random anomaly injection, TPA-AD improves the separability of the normal boundary by constructing pseudo-anomalies in boundary neighborhoods and can jointly handle continuous and discrete features in mixed-variable scenarios. The main experiments are conducted on bearing fault detection datasets and degradation-process datasets, with an additional exploratory extension on $13$ public TSAD datasets. The results show that the proposed method yields relatively stable anomaly responses, is sensitive to degradation evolution, and demonstrates a certain degree of broader applicability on public TSAD benchmarks and real high-speed-train-related bearing data.

AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting

Time series forecasting models are increasingly scaled through large Transformer backbones, yet most existing approaches process all series through a shared dense computation path despite substantial heterogeneity in temporal structure. Mixture-of-Experts (MoE) offers a natural alternative by enabling conditional computation, but standard MoE routing leaves expert specialization weakly identified and often unstable during downstream adaptation. We propose AME-TS, a structure-guided sparse time series foundation model that aligns expert routing with interpretable temporal structure. AME-TS first uses a lightweight regime predictor to estimate series-level descriptors, including forecastability, seasonality, trend, and sparsity, and maps them to a soft structural prior over experts. This series-level prior guides token-level routing during training, encouraging structure-aligned specialization. On the GIFT-Eval benchmark, AME-TS delivers a strong accuracy-efficiency tradeoff across model scales: it substantially outperforms existing time series foundation models at small model scales and remains competitive with the strongest models at larger scales, while activating substantially fewer parameters through sparse routing. We further show that AME-TS learns more interpretable routing geometry and substantially more stable expert specialization than standard MoE during fine-tuning on the M5 dataset. These results suggest that structure-aware routing is an effective and reliable way to realize the benefits of sparse expert models for time series forecasting.

StepAudio 2.5 Technical Report

Unified audio-language modeling has emerged as a prominent trend in modern speech systems, promising to bring the reasoning capabilities of large language models to auditory tasks. However, existing unified foundations often struggle to match the depth of specialized systems across automatic speech recognition (ASR), text-to-speech synthesis (TTS), and realtime spoken interaction. Bridging this gap remains an open challenge. This report presents StepAudio 2.5, a unified audio-language foundation model that matches or exceeds specialized systems across all three capabilities. Rather than treating these tasks as architecturally distinct, we operate on the premise that once text and audio share a multimodal representational space, task specialization becomes a matter of operational regimes: data construction, optimization targets, and decoding constraints. Guided by this insight, we advance the post-training paradigm from standard supervised learning to task-tailored Reinforcement Learning from Human Feedback (RLHF), using it as the primary mechanism to define complex optimization targets. We leverage this RLHF-centric alignment, alongside specialized decoding, to shape a shared backbone into three distinct operational modes. Concretely, the ASR branch advances transcription efficiency via verifiable multi-token decoding; the TTS branch achieves controllable, expressive synthesis through preference-based RLHF and context-rich supervision; and the Realtime branch realizes low-latency, persona-consistent dialogue via generative reward modeling within an RLHF framework. On standard benchmarks, StepAudio 2.5 achieves state-of-the-art results across ASR, TTS, and Realtime, demonstrating that a singular audio-language foundation can successfully internalize the distinct deployment objectives of speech understanding, generation, and live interaction.

Hybrid Quantum-Classical Corrective Diffusion Modeling for Meteorological Downscaling

Statistical downscaling is a crucial component of the weather modeling field, where high-resolution outputs must be reconstructed from coarse-resolution inputs with the full cost of dynamical refinement. In this work, we investigate a hybrid quantum-classical corrective diffusion model for probabilistic statistical downscaling of weather fields. The proposed model inserts variational quantum circuit layers into the most compressed bottleneck of the diffusion UNet while leaving the regression branch fully classical. This placement tests whether quantum circuits can act as compact nonlinear feature maps for latent-channel mixing. We evaluate intra-channel and cross-channel ansätze on 10m wind components. On the 2020 validation set, the hybrid models remain stable, preserve the large-scale spatial organization of the generated wind fields, and improve both MAE and CRPS relative to a classical corrective diffusion model in several configurations. Structural diagnostics further show that the hybrid variants preserve kinetic-energy spectra and windspeed distributions similar to its classical counterpart while producing controlled changes in tail behavior, extreme-windspeed localization, and joint wind field components structure. Backend studies on the 2020 validation set show negligible impact from simulated device noise at the tested circuit scale, whereas real-hardware deployment remains limited by qubit availability and execution fidelity. The 2021 out-of-distribution test shows that these in-distribution gains do not transfer uniformly under temporal shift, revealing a generalization gap that motivates future mitigation through stabilization and regularization. These results show that bottleneck-level quantum hybridization can make a nontrivial contribution to weather statistical downscaling, while also highlighting that circuit scale and hardware deployment remain key limiting factors.