arXiv subjects

Li Zhao

Publications and source records attributed to Li Zhao.

At least 19 recordsLinked to original sources

arXiv2026

EPOFusion: Exposure aware Progressive Optimization Method for Infrared and Visible Image Fusion

Overexposure caused by strong daylight and oncoming headlights frequently overwhelms visible sensors, resulting in critical information loss in visual perception. Infrared and visible image fusion can compensate for such degradation via multimodal complementarity. However, most fusion methods lack region-aware optimization for overexposed areas and cannot effectively exploit infrared cues in saturated regions, resulting in insufficient infrared detail preservation or redundant information in the fused results. To address this, we propose EPOFusion, an exposure-aware fusion framework. It employs a spatial guidance module to identify regions requiring infrared compensation, together with a region-aware fusion loss to strengthen informative infrared structures. In addition, an iterative feature refinement head equipped with a multiscale context fusion module progressively refines fused representations, enabling effective integration of complementary infrared information while maintaining visual consistency in normally exposed regions. The infrared and visible overexposure (IVOE) dataset consists of a synthetic training subset providing infrared-compensation supervision and a real-world subset for fusion and downstream perception evaluation under authentic overexposure. EPOFusion demonstrates superior VIF and $Q^{AB/F}$ performance with favorable visual quality, improving $Q^{AB/F}$ by 10.7% over the existing overexposure-oriented fusion baseline, while further improving downstream mIoU and mAP50 by 5.6% and 6.5%, respectively. Code, results, and the IVOE dataset will be made available at https://warren-wzw.github.io/EPOFusion/.

cs.CV↗

arXiv2026

Bringing Value Models Back: Generative Critics for Value Modeling in LLM Reinforcement Learning

Credit assignment is a central challenge in reinforcement learning (RL). Classical actor-critic methods address this challenge through fine-grained advantage estimation based on a learned value function. However, learned value models are often avoided in modern large language model (LLM) RL because conventional discriminative critics are difficult to train reliably. We revisit value modeling and argue that this difficulty is partly due to limited expressiveness. In particular, representation complexity theory suggests that value functions can be hard to approximate under the one-shot prediction paradigm used by existing value models, and our scaling experiments show that such critics do not improve reliably with scale. Motivated by this observation, we propose Generative Actor-Critic (GenAC), which replaces one-shot scalar value prediction with a generative critic that performs chain-of-thought reasoning before producing a value estimate. We further introduce In-Context Conditioning, which helps the critic remain calibrated to the current actor throughout training. GenAC improves value approximation, ranking reliability, and out-of-distribution generalization, and these gains translate into stronger downstream RL performance than both value-based and value-free baselines. Overall, our results suggest that stronger value modeling is a promising direction for improving credit assignment in LLM reinforcement learning.

cs.LG↗

arXiv2026

Search for neutrinoless quadruple beta decay of $^{136}$Xe in PandaX-4T detector

The observation of neutrinoless quadruple beta decay (0$ν$4$β$) in the absence of neutrinoless double beta decay (0$ν$2$β$) has been argued to provide a strong indication that neutrinos are Dirac particles. We report a search for 0$ν$4$β$ decay of $^{136}\text{Xe}$ using a total $^{136}\text{Xe}$ exposure of 148.4 kg$\cdot$yr, collected during the commissioning and the first science runs of the PandaX-4T experiment. No significant excess of events over the background is observed. A lower limit on the 0$ν$4$β$ decay half-life of $^{136}\text{Xe}$ is set at 6.01 x $10^{24}$ yr at the 90% confidence level. This result establishes the most stringent constraint on this process in xenon, demonstrating the unique capability of the PandaX-4T detector in probing lepton number violation and shedding light on the fundamental nature of neutrinos.

nucl-ex↗

arXiv2026

Alpha-R1: Alpha Screening with LLM Reasoning via Reinforcement Learning

Signal decay and regime shifts pose recurring challenges for data-driven investment strategies in non-stationary markets, where conventional time-series and machine learning approaches often struggle to generalize beyond historical correlations. While large language models (LLMs) offer strong capabilities for processing unstructured information, their potential to support quantitative factor screening through explicit economic reasoning remains underexplored. Existing factor-based methods typically reduce alphas to numerical time series, overlooking the semantic rationale that determines when a factor is economically relevant. We present Alpha-R1, an RL-aligned LLM framework for context-aware alpha screening. Its core mechanism, semantic gating, evaluates each candidate factor's semantic profile against a dynamically constructed market state description, selecting a sparse subset of factors whose economic rationale aligns with current market conditions. The selection model is trained via group relative policy optimization (GRPO), using realized portfolio returns as the primary reward signal. Under a 12-month out-of-sample evaluation, Alpha-R1 achieves annualized returns of 47.87% on S&P 500 and 40.57% on CSI 300 with Sharpe ratios of 1.62 and 2.23. These results, obtained under a bounded candidate-pool evaluation protocol, provide evidence for second-stage semantic factor reranking in non-stationary markets. The full implementation and resources are available at https://github.com/FinStep-AI/Alpha-R1.

q-fin.TR↗

arXiv2026

ContestTrade: A Multi-Agent Trading System Based on Internal Contest Mechanism

In financial trading, large language model (LLM)-based agents demonstrate significant potential, but their decisions can be sensitive to noisy and non-stationary market information. We propose ContestTrade, a multi-agent trading system with an internal competitive mechanism inspired by institutional investment workflows. The system consists of two specialized teams: (1) a Data Team that processes and condenses massive market data into diversified textual factors optimized for constrained LLM context windows, and (2) a Research Team that produces parallelized multipath trading decisions via tool-augmented deep research. The core design is a "Quantify-Predict-Allocate" contest mechanism within each team: agent outputs are scored only after market outcomes become observable, future utility is predicted from historical scores, and resources are allocated to agents with positive predicted utility. In a post-2024 A-share backtest, ContestTrade achieves higher backtested return and risk-adjusted performance than the evaluated baselines. We further describe the temporal protocol, implementation choices, and limitations to clarify the scope of these results.

q-fin.TR↗

arXiv2026

ProtoHGF-Net: Prototype HyperGraph Fusion with Intra-modal Calibration for RGBT Object Detection

RGB-Thermal (RGBT) object detection enables robust perception in complex scenes by leveraging the complementary strengths of visible textures and thermal cues. However, existing methods mainly rely on dense cross-modal interactions over full-resolution features, which inevitably introduce background interference and hinder the learning of target-relevant representations. In this paper, we propose the Prototype HyperGraph Fusion Network (ProtoHGF-Net), a novel framework that redefines cross-modal fusion as prototype-level semantic interaction rather than the dense cross-modal interaction paradigm. Specifically, we design Prototype HyperGraph Fusion to perform cross-modal interaction in a compact prototype-level semantic space. This design enables more selective fusion among target-relevant prototypes. To support this prototype-level fusion, we propose Teacher-Mask Calibration Distillation, which calibrates modality features before fusion using modality-specific teachers and target-aware masks. This strategy suppresses backgrou- nd-dominant responses and produces more target-focused features. Extensive experiments on DroneVehicle, DVTOD, and FLIR demonstrate that ProtoHGF-Net achieves state-of-the-art performance with 85.9\% $mAP_{50}$, 88.2\% $mAP_{50}$, and 79.1\% $mAP_{50}$, respectively. Our code is available at \href{https://github.com/ZiMo-Chen/ProtoHGF}{GitHub}.

cs.CV↗

arXiv2026

REATS: LLM Reasoning-based Ensemble Learning for Adaptive Time Series Forecasting

Due to the diversity of real-world time series, no single forecasting model consistently dominates across all samples. Ensemble learning addresses this by combining complementary model strengths, yet existing methods rely on fixed rules or black-box models based solely on numerical inputs, failing to leverage LLM reasoning for interpretable weighting decisions. We propose REATS, which leverages LLM reasoning capabilities as an intelligent ensemble router that jointly processes textual temporal pattern descriptions and numerical features to produce interpretable, sample-adaptive ensemble weights through chain-of-thought reasoning. To enable effective LLM-based ensembling, we study its key design choices and propose: (i) a structured input pipeline that transforms raw time series into hybrid textual--numerical representations with fixed token cost, enabling rule-based chain-of-thought construction without API dependency, augmented with retrieved similar-sample priors; (ii) a diverse multi-row weight supervision scheme coupled with a token-efficient percentage-table format that reduces numerical complexity and mitigates LLM hallucinations; and (iii) a two-stage fine-tuning framework combining SFT with GRPO, where a reciprocal reward mapping transforms the continuous unbounded MSE gap into bounded signals with amplified near-oracle sensitivity, addressing the uniform sensitivity and outlier-dominated advantage compression inherent in naive reward designs for regression-based GRPO. Experiments on eight benchmarks demonstrate that REATS outperforms competitive ensemble baselines while providing natural language explanations and demonstrating strong transfer learning and out-of-domain generalization to unseen candidate models.

cs.LG↗

arXiv2026

OATS: Online Data Augmentation for Time Series Foundation Models

Time Series Foundation Models (TSFMs) are a powerful paradigm for time series analysis and are often enhanced by synthetic data augmentation to improve the training data quality. Existing augmentation methods, however, typically rely on heuristics and static paradigms. Motivated by dynamic data optimization, which shows that the contribution of samples varies across training stages, we propose OATS (Online Data Augmentation for Time Series Foundation Models), a principled strategy that generates synthetic data tailored to different training steps. OATS leverages valuable training samples as principled guiding signals and dynamically generates high-quality synthetic data conditioned on them. We further design a diffusion-based framework to produce realistic time series and introduce an explore-exploit mechanism to balance efficiency and effectiveness. Experiments on TSFMs demonstrate that OATS consistently outperforms regular training and yields substantial performance gains over static data augmentation baselines across six validation datasets and two TSFM architectures. The code is available at the link https://github.com/microsoft/TimeCraft.

cs.LG↗

arXiv2026

UniSteer: Unified Noise Steering for Efficient Human-Guided VLA Adaptation

Diffusion-based vision-language-action (VLA) models have emerged as strong priors for robotic manipulation, yet adapting them to real-world distributions remains challenging. In particular, on-robot reinforcement learning (RL) is expensive and time-consuming, so effective adaptation depends on efficient policy improvement within a limited budget of real-world interactions. Noise-space RL lowers the cost by keeping the pretrained VLA fixed as a denoising generator while updating only a lightweight actor that predicts the noise. However, its performance is still limited due to inefficient autonomous exploration. Human corrective interventions can reduce this exploration burden, but they are naturally provided in action space, whereas noise-space finetuning requires supervision over noise variables. To address these challenges, we propose UniSteer, a Unified Noise Steering framework that combines human corrective guidance with noise-space RL through approximate action-to-noise inversion. Given a human corrective action, UniSteer inverts the frozen flow-matching decoder to recover a noise target, which provides supervised guidance for the same noise actor that is simultaneously optimized via reinforcement learning. Real-world experiments on diverse manipulation tasks show that UniSteer adapts more efficiently than strong noise-space RL and action-space human-in-the-loop baselines, improving the success rate from 20% to 90% in 66 minutes on average across four real-world adaptation tasks.

cs.RO↗

arXiv2026

Beyond Monotonic Progress: Retry-Supervised Value Learning for Robot Imitation

Human demonstrations for robot imitation learning often contain mistakes and corrective behaviors, such as imprecise grasps, object misalignment, unstable contact, and repeated attempts. While these segments are commonly treated as noisy or suboptimal data, they provide valuable evidence about when execution deviates from a desirable path and how task feasibility can be restored. However, existing reward and value models often rely on monotonic progress assumptions, which capture coarse task advancement but may overlook local execution errors and corrective behaviors in imperfect demonstrations. In this work, we propose ReTVL (ReTry-Supervised Value Learning), a framework for learning mistake-sensitive value functions from mixed-quality robot demonstrations by leveraging retry events as sparse supervision. ReTVL captures the local degradation-and-recovery structure around mistakes by combining global progress calibration with local pairwise preference learning induced by sparsely annotated retry keypoints. The learned value model is then used to reweight demonstration chunks for downstream behavior cloning, reducing the influence of harmful execution errors while preserving useful corrective behaviors. Experiments on real-robot manipulation tasks show that ReTVL produces more fine-grained value estimates than progress-based baselines and improves imitation learning from imperfect demonstrations.

cs.RO↗

arXiv2026

Measurement of solar $pp$ neutrino flux with the new PandaX-4T data

We report a new measurement of the solar proton--proton ($pp$) neutrino flux via neutrino--electron elastic scattering using the PandaX-4T Run 2 data set collected between 2024 and 2026, corresponding to an exposure of 1.9 tonne$\cdot$yr. Before Run 2 data taking, the detector underwent a series of upgrades to improve its response and background conditions. Time variations of radioactive noble-gas impurities are constrained using the physics data themselves, complemented by measurements from the gas-assay system. The analysis introduced improvements in the data processing chain, detector response characterization, and background models. A blind spectral analysis was then performed on the electronic-recoil data across a wide energy range from 20 to 1000 keV. In combination with the Run 0 data published earlier, the fitted $pp$ flux is $(8.5 \pm 3.5)\times 10^{10}$ $\mathrm{cm^{-2}s^{-1}}$, consistent with the prediction of the Standard Solar Model. With a statistical significance of $2.2σ$ above background, this marks the first positive indication of solar $pp$ neutrino--electron scattering below an electronic-recoil energy of 165 keV.

hep-ex↗

arXiv2026

Design, construction, and testing of the PandaX-xT cryogenics system

The PandaX-xT is a next-generation experiment with broad scientific goals, including the search for dark matter, Neutrinoless Double Beta Decay, and astrophysical neutrinos, using a dual-phase time projection chamber with about 43 tons of liquid xenon. A new cryogenics system of the PandaX-xT is described in this paper. It is developed to handle large mass of liquid xenon efficiently and safely, including two cooling towers for normal operation and one liquid-nitrogen coil for emergency case. Each cooling tower equipped with an AL600 Gifford-McMahon cryocooler features a 1300 W heater, specifically designed to maintain the cold finger's temperature at the desired setpoint. The performance of the cooling tower and the coil has been tested. The cryogenics system with two cooling towers has achieved about 1900~W cooling power at 178~K. The liquid nitrogen coil provides emergency cooling power of more than 1500~W at liquid xenon temperature. For the prototype of a 1-tonne liquid xenon detector, the fluctuation of xenon saturated vapor pressure remains below 1 kPa over one month, while the pressure is around 210~kPa.

physics.ins-det↗

arXiv2026

MVP-LAM: Learning Action-Centric Latent Action via Cross-Viewpoint Reconstruction

Latent actions learned from diverse human videos serve as pseudo-labels for vision-language-action (VLA) pretraining, but provide effective supervision only if they remain informative about the underlying ground-truth actions. For effective supervision, latent actions should contain information about the underlying actions even though they are inaccessible. We propose Multi-ViewPoint Latent Action Moel (MVP-LAM), which learns latent actions that are highly informative about ground-truth actions from multi-view videos. MVP-LAM trains latent actions with a cross-viewpoint reconstruction objective, so that a latent action from one view must explain the future in another view, reducing reliance on viewpoint-specific cues. On Bridge V2, MVP-LAM produces more action-centric latent actions, achieving higher mutual information with ground-truth actions and improved action prediction, including under out-of-distribution evaluation. Finally, pretraining VLAs with MVP-LAM latent actions improves downstream manipulation performance on various benchmarks. The code and trained checkpoints are available at https://jmsnu.github.io.

cs.RO↗

arXiv2026

Enabling High-Accuracy Data Assimilation with Limited Ensembles via Machine Learning-Based Covariance Correction

Data assimilation (DA) integrates numerical model forecasts with observations to achieve the optimal state estimation. Ensemble-based methods, such as the ensemble Kalman filter (EnKF), are widely used for state estimation for high-dimensional and nonlinear dynamic systems. However, their performance strongly depends on the ensemble size, therefore causing a tradeoff problem between analysis accuracy and computational cost. To address this problem, this study presents a machine learning-based EnKF framework that maintains high accuracy with a relatively small ensemble size. Specifically, a multilayer perceptron (MLP) function is built to predict the difference between the forecast error covariances estimated from a limited ensemble and a sufficiently large ensemble, with the latter being assumed to be an accurate approximation of the underlying truth. This predicted covariance difference term is then incorporated into the EnKF algorithm via an element-wise scaling strategy, resulting in an amended forecast covariance matrix that better approximates the true uncertainty level and sequentially produces more accurate analysis results. To demonstrate the feasibility and robustness of the proposed algorithm, we perform a set of numerical experiments with the Lorenz-63 and Lorenz-96 systems under various configurations, and the results consistently indicate that the proposed algorithm can significantly outperform the standard EnKF with the same limited ensemble size, by achieving notably higher analysis accuracy while remaining computationally efficient. This approach provides a practical and feasible pathway to accurate and computationally efficient data assimilation for high-dimensional and nonlinear dynamic systems.

physics.ao-ph↗

arXiv2026

Reinforcing VLAs in Task-Agnostic World Models

Post-training Vision-Language-Action (VLA) models via reinforcement learning (RL) in learned world models has emerged as an effective strategy to adapt to new tasks without costly real-world interactions. However, while using imagined trajectories reduces the sample complexity of policy training, existing methods still heavily rely on task-specific data to fine-tune both the world and reward models, fundamentally limiting their scalability to unseen tasks. To overcome this, we argue that world and reward models should capture transferable physical priors that enable zero-shot inference. We propose RAW-Dream (Reinforcing VLAs in task-Agnostic World Dreams), a new paradigm that completely disentangles world model learning from downstream task dependencies. RAW-Dream utilizes a world model pre-trained on diverse task-free behaviors for predicting future rollouts, and an off-the-shelf Vision-Language Model (VLM) for reward generation. Because both components are task-agnostic, VLAs can be readily finetuned for any new task entirely within this zero-shot imagination. Furthermore, to mitigate world model hallucinations, we introduce a dual-noise verification mechanism to filter out unreliable rollouts. Extensive experiments across simulation and real-world settings demonstrate consistent performance gains, proving that generalized physical priors can effectively substitute for costly task-dependent data, offering a highly scalable roadmap for VLA adaptation.

cs.AI↗

arXiv2026

From Human Videos to Robot Manipulation: A Survey on Scalable Vision-Language-Action Learning with Human-Centric Data

Recent progress in generalizable embodied control has been driven by large-scale pretraining of Vision-Language-Action (VLA) models. However, most existing approaches rely on large collections of robot demonstrations, which are costly to obtain and tightly coupled to specific embodiments. Human videos, by contrast, are abundant and capture rich interactions, providing diverse semantic and physical cues for real-world manipulation. Yet, embodiment differences and the frequent absence of task-aligned annotations make their direct use in VLA models challenging. This survey provides a unified view of how human videos are transformed into effective knowledge for VLA models. We categorize existing approaches into four classes based on the action-related information they derive: (i) latent action representations that encode inter-frame changes; (ii) predictive world models that forecast future frames; (iii) explicit 2D supervision that extracts image-plane cues; and (iv) explicit 3D reconstruction that recovers geometry or motion. Beyond this taxonomy, we highlight three key open challenges in this area: structuring unstructured videos into training-ready episodes, grounding video-derived supervision into robot-executable actions under embodiment and viewpoint heterogeneity, and designing evaluation protocols that better predict real-world deployment performance and transfer efficiency, thereby informing future research directions. A curated list of papers and resources is available at https://github.com/AaronFengZY/HumanCentricToVLA-Survey.

cs.RO↗

arXiv2026

Causal Time Series Generation via Diffusion Models

Time series generation (TSG) synthesizes realistic sequences and has achieved remarkable success. Among TSG, conditional models generate sequences given observed covariates, however, such models learn observational correlations without considering unobserved confounding. In this work, we propose a causal perspective on conditional TSG and introduce causal time series generation as a new TSG task family, formalized within Pearl's causal ladder, extending beyond observational generation to include interventional and counterfactual settings. To instantiate these tasks, we develop CaTSG, a unified diffusion-based framework with backdoor-adjusted guidance that causally steers sampling toward desired interventions and individual counterfactuals while preserving observational fidelity. Specifically, our method derives causal score functions via backdoor adjustment and the abduction-action-prediction procedure, thus enabling principled support for all three levels of TSG. Extensive experiments on both synthetic and real-world datasets show that CaTSG achieves superior fidelity and also supporting interventional and counterfactual generation that existing baselines cannot handle. Overall, we propose the causal TSG family and instantiate it with CaTSG, providing an initial proof-of-concept and opening a promising direction toward more reliable simulation under interventions and counterfactual generation.

cs.LG↗

arXiv2026

GEAR: Granularity-Adaptive Advantage Reweighting for LLM Agents via Self-Distillation

Reinforcement learning has become a widely used post-training approach for LLM agents, where training commonly relies on outcome-level rewards that provide only coarse supervision. While finer-grained credit assignment is promising for effective policy updates, obtaining reliable local credit and assigning it to the right parts of the long-horizon trajectory remains an open challenge. In this paper, we propose Granularity-adaptivE Advantage Reweighting (GEAR), an adaptive-granularity credit assignment framework that reshapes the trajectory-level GRPO advantage using token- and segment-level signals derived from self-distillation. GEAR compares an on-policy student with a ground-truth-conditioned teacher to obtain a reference-guided divergence signal for identifying adaptive segment boundaries and modulating local advantage weights. This divergence often spikes at the onset of a semantic deviation, while later tokens in the same autoregressive continuation may return to low divergence. GEAR therefore treats such spikes as anchors for adaptive credit regions: where the student remains aligned with the teacher, token-level resolution is preserved; where it departs, GEAR groups the corresponding continuation into an adaptive segment and uses the divergence at the departure point to modulate the segment' s advantage. Experiments across eight mathematical reasoning and agentic tool-use benchmarks with Qwen3 4B and 8B models show that GEAR consistently outperforms standard GRPO, self-distillation-only baselines, and token- or turn-level credit-assignment methods. The gains are especially strong on benchmarks with lower GRPO baseline accuracy, reaching up to around 20\% over GRPO, suggesting that the proposed adaptive reweighting scheme is especially useful in more challenging long-horizon settings.

cs.LG↗