arXiv subjects

Weijia Han

Publications and source records attributed to Weijia Han.

At least 19 recordsLinked to original sources

arXiv2026

Quantile Propagation Planning for Budgeted Tail-Risk Evaluation

Risk-aware planning must identify rare, costly outcomes using only a limited number of model evaluations. Monte Carlo rollouts spend this budget on noisy tail estimates, while planning from the mean can miss the tail entirely. We introduce quantile propagation planning (QPP), which propagates a small, fixed set of disturbance quantiles through the dynamics and integrates the resulting cost curve to evaluate both mean and tail risk. The method is exact at the propagated levels, converges as the quantile grid is refined, and admits a one-sided conservatism guarantee under stated structural conditions. In controlled overload and inventory tasks, QPP reduces realized tail cost and constraint violations relative to mean-rollout planning. On overload, it remains competitive with sampled evaluators that use substantially more trajectories. These results position QPP as a transparent, budget-controlled approach to tail-risk evaluation.

cs.LG↗

arXiv2026

Towards Self-Evolving Narrative Systems: Real-Time Adaptation and Long-Term Optimization with LLMs

Traditional interactive narrative systems offer strong authorial control and structural coherence but often restrict player agency. In contrast, LLM-driven narratives enable open-ended generation and greater flexibility, yet frequently suffer from inconsistency and diminished narrative control. To navigate this trade-off, we propose a hybrid narrative framework that dynamically balances pre-authored content with real-time LLM generation via semantic similarity alignment. Additionally, sustaining narrative quality over extended gameplay remains a core challenge in LLM-based storytelling. To address this, we introduce a self-evolving, feedback-driven optimization framework that leverages in-context learning to iteratively refine prompts, eliminating the need for manual prompt engineering. Through a case study and participant-based evaluation, we show that our framework enhances narrative coherence while preserving player agency, effectively aligning generated content with both authorial intent and evolving player behavior.

cs.CL↗

arXiv2026

Sparse Autoencoders as Implicit Diversification Regularisers for Cross-Sectional Cryptocurrency Returns

Factor models are central to asset pricing and portfolio construction, but neural factors are often difficult to interpret, and their effect on portfolio risk is unclear. A sparse autoencoder (SAE) may address both concerns by describing daily returns across many assets with only a few active features. We ask whether this sparse representation produces stable financial factors and better-diversified portfolios. We study cryptocurrency returns, use equities as a secondary test, and compare the SAE with common benchmarks in a time-ordered evaluation. We find consistently lower volatility in SAE portfolios and link this result to lower correlations among the selected assets. However, the SAE does not outperform a simple contrarian strategy on risk-adjusted return, and its factor identities are not stable across training runs. Our contribution is to show that sparse coding can support portfolio diversification, while clearly separating this risk effect from claims about alpha or stable factor identities.

cs.LG↗

arXiv2026

Checkpoint Reload Explains Most of the Measured State Gap in Two State Discovery Reproductions

Temporal models that maintain per-sequence latent states during training must reconstruct those states when the stored table is unavailable at inference. This creates a basic consistency question: do the reconstructed states agree with those formed during training? We study this question for State Discovery in World Machine, which updates stored states in parallel each epoch but performs inference by sequential rollout from an all-zero start. On Toy1D, we analyze two paired-checkpoint runs whose decoded-state MSE is 27 to 36 percent above the published value. We compare the stored table with rollouts under the best-validation checkpoint reloaded after training and the last-epoch weights that wrote the table. Matching the rollout to the table-writing weights removes 90 to 91 percent of the mean per-dimension Wasserstein-1 distance. Causal distillation and frozen SD updates do not satisfy their pre-specified criteria under either checkpoint. Overall, checkpoint reload accounts for most of the measured gap in these runs.

cs.LG↗

arXiv2026

The Label Complexity of Useful Class-Conditional Prediction Sets under Distribution Shift

Prediction sets can make deployed classifiers safer by returning several plausible labels when a single prediction is uncertain. Their value depends on classwise reliability: average coverage can meet its target while rare or difficult classes fail repeatedly. This concern is sharper after distribution shift, when calibration labels come from a source environment but reliability is needed on the target. We ask what labeled source data and unlabeled target inputs reveal about class-conditional prediction sets, and when target labels are necessary. Under unrestricted joint shift, two target laws can produce the same observable data while requiring different classwise thresholds; any label-free rule covering both must enlarge its sets on one law. We give a labeled target audit that estimates the missing quantiles with a simultaneous guarantee. Probability-scale error is invariant to increasing score transformations, and threshold recovery follows under local regularity. At fixed confidence, achieving threshold tolerance $\varepsilon$ with fixed, nonadaptive class-stratified labeled pairs has total complexity $Θ(K\varepsilon^{-2}\log K)$, or $Θ(\varepsilon^{-2}\log K)$ labels per class under equal allocation. Class imbalance creates a separate acquisition cost; for foreground class probabilities of order $1/K$, the mixed-stream label complexity is also $Θ(K\varepsilon^{-2}\log K)$ at fixed confidence. Experiments on action-recognition and image shifts show that marginal coverage can conceal severe class failures and that source classwise calibration depends on the shift. The results connect the information available at deployment to the target labels needed for useful class-conditional prediction.

cs.LG↗

arXiv2026

The Premise Is the Problem: Exchangeability Failure in Self-Monitored Test-Time Adaptation

Modern forecasting models are often updated after deployment so they can respond to changing data. These updates can also make predictions worse, so practical systems need a reliable monitor that can detect harmful changes and trigger protection. A natural design is to monitor the same prediction errors that guide the updates. This paper asks whether the statistical guarantee behind such a monitor remains valid when monitoring and adaptation use the same feedback. We study this question in multi-step time-series forecasting. We show that overlapping targets and dependence in forecast errors can break a key assumption required by the guarantee. The monitor may then raise alarms even when no harmful change has occurred, and its response can further damage prediction quality. We also find that adaptation can hide sustained changes from its own monitor, while the original frozen model retains a clearer signal. These results expose a basic failure mode in self-monitored adaptation. They show why reliable deployment requires checking the monitor's assumptions, comparing adaptation with the frozen model under realistic feedback, and limiting the effect of every protective response.

cs.LG↗

arXiv2026

What Does a Token Cost? A Mixture-of-Agents Measurement of Sufficient Per-Token Compute

Large language models spend the same amount of computation on every token they generate, regardless of how difficult each token is to produce. Methods such as speculative decoding and model routing are built on the premise that much of this computation is unnecessary, yet the computation an individual token actually requires has not been measured. We measure it through a Mixture-of-Agents (MoA) lens: a panel of fifteen language models of increasing capacity, drawn from three families, in which every agent attempts to reproduce a reference sequence token by token, conditioned on the correct preceding tokens. We define the inference cost of the smallest agent that succeeds as the token's sufficient compute, which upper-bounds what the token requires. On three core benchmarks, a 0.5B agent reproduces 92--95\% of reference tokens. Across Qwen, OLMo, and R1-distilled panels, the most expensive 10\% account for 64--80\% of estimated FLOPs. On all 500 MATH-500 problems, the MoA-derived map helps model routing reduce projected latency from 7.59 to 5.12 seconds while slightly improving accuracy, relative to the best confidence-routing baseline. The MoA-map helps drafting use 32.6\% fewer draft tokens and approximately 20\% lower projected latency than fixed-window drafting at similar accuracy. These comparisons reveal remaining allocation headroom, motivating controllers that exploit sufficient-compute structure.

cs.AI↗

arXiv2026

Robust Resource Management for SAGIN using DNN-Driven Channel Uncertainty Learning

This paper focuses on the joint robust beamforming and resource allocation for space-air-ground integrated networks (SAGIN) under uncertain channel state information (CSI). In SAGIN, uncertain CSI undermines the precise adjustment of beamforming and resource allocation, posing a major challenge to meeting heterogeneous users' strict quality of service (QoS) requirements. To address this challenge, we first formulate a chance-constrained optimization problem to minimize the total transmit power while satisfying QoS requirements under a predefined outage probability. By leveraging semidefinite relaxation (SDR), the objective function is transformed into a linear function of the traces of the beamforming matrices. Then, we propose a deep neural network (DNN)-driven channel uncertainty learning to dynamically learn and model the uncertain CSI as an asymmetric uncertainty set. Under the constructed CSI uncertainty set, a robust counterpart method based on pre-trained network parameters is developed. It characterizes the uncertainty set as a finite union of convex sets, thereby providing a tractable approximation for the original chance constraints. Finally, we design an adaptive iterative algorithm to jointly optimize the resource allocation and beamforming vectors. Simulation results show that our proposed DNN-driven method outperforms traditional robust and non-robust methods, achieving a superior energy efficiency and robust reliability in SAGIN.

cs.IT↗

arXiv2026

When the Martingale Never Stops Firing: Anytime-Valid Gating on Real Forecast Streams

Machine learning systems are increasingly corrected while they run, and the decision of when to intervene is increasingly delegated to statistical monitors. Anytime-valid inference promises evidence that can be acted on at any moment, exactly the guarantee this setting needs, and it is moving from theory into deployed monitoring. Conformal test martingales are the change-detection instrument, and Ville's inequality caps their false-alarm probability on exchangeable data. The guarantee is conditional. A deployment inherits it only if the stream it monitors behaves exchangeably. The premise is hardest to satisfy where these monitors are most useful, on dependent data and inside loops where the monitor modifies the learner whose scores it reads. It is also rarely measured. We measure it in a pre-specified case study, where such a monitor gates the online updates of a Kalman adapter correcting frozen time-series foundation models on five forecasting streams. On exchangeable synthetic streams, the same implementation fires in at most 1 of 60 runs. On the real streams, at alpha = 0.05, 135 of 135 clean-stream runs fired. The construction does not explain the firing; the failure comes from the deployed score stream itself. Repeated fires hold the gate's drift response active, and the gated filter amplifies the very transient it was designed to prevent. The component worth keeping makes no validity claim. Huber-style gating of the filter's own updates cuts isolated-spike degradation by an order of magnitude with no dataset specific tuning. Anytime-valid methods proposed for dependent data should therefore be accompanied by null-calibration controls and mechanism traces.

cs.LG↗

arXiv2026

Decomposing Runtime, Kernel, and Quantization Speedups via a Matched FP16 Intermediate: A Hardware-Conditioned Case Study on Four NVIDIA RTX A5000 GPUs

Reported serving speedups from quantized kernels typically bundle the weight format, the kernel, and the inference runtime into one number. We present an attribution study on four NVIDIA RTX A5000 GPUs, 24 GiB each, on a single host with NVLink-bridged pairs. A matched intermediate stack that keeps the faster runtime without the quantized kernel splits the full speedup into a runtime part and a kernel and quantization part. Under matched greedy decoding the full stack reaches $2.58\times$ end to end, with the runtime change accounting for about two thirds of that gain on a logarithmic scale; across three similar model families the kernel and quantization part moves by at most 1.5%. Sharding one instance across all four cards falls well below doubling: a profiler trace attributes about 80% of the per token shortfall to coordination, and an NVLink versus PCIe control on the same hardware shows similar realized bandwidth on both links, pointing away from link bandwidth as the cause. Whether to run one sharded instance or several independent ones depends on the workload and the model, with the ranking reversing on the larger model: the smaller model splits between sharding and multiple instances by workload, while the larger model favors two paired instances on every workload. Quantization extends sustainable concurrent users roughly four times past a reproducible half precision memory cliff. Differences in sampling mode and prompt pool between the two stacks are documented as threats to validity.

cs.DC↗

arXiv2026

Hyperon-Nucleon Spectrometer

Chirality lies at the heart of low-energy QCD, governing the symmetry structure that shapes hadron masses and strong interaction dynamics. Among the most compelling open questions tied to chiral dynamics and spontaneous chiral symmetry breaking is the longstanding $Λ$ polarization puzzle, in which $Λ$ hyperons produced in unpolarized hadronic collisions exhibit a surprisingly large transverse polarization that remains theoretically unexplained. This whitepaper presents the proposal for the Hyperon-Nucleon Spectrometer (H-NS) at the High-Intensity heavy-ion Accelerator Facility (HIAF). Leveraging the high energy and high intensity of HIAF's proton and heavy-ion beams, the H-NS experiment will perform systematic studies of hyperon polarization phenomena and their underlying mechanisms in proton-proton ($pp$), proton-nucleus ($pA$), and nucleus-nucleus ($AA$) collisions in the fixed target mode. A wide-range beam energy scan, including proton beams from 3 GeV up to 9.3 GeV (HIAF) and up to 32 GeV (upgraded HIAF), will be conducted to examine the dependence of polarization on collision energy. The spectrometer is designed with specialized detectors capable of high-precision reconstruction of final-state baryon polarizations. Among its many interesting and important measurements, H-NS will simultaneously measure hyperon and proton spin observables to explore the polarization mechanism in hadronic interactions and the spin structure of baryons. Furthermore, the use of $pA$ and $AA$ collisions will enable detailed investigations of cold and hot nuclear matter effects on spin polarization. Its physics program and detector development will significantly benefit the future Electron-ion Collider in China.

physics.ins-det↗

arXiv2025

Huizhou Hadron Spectrometer -- a Proposed High-rate Experimental Setup at the High Intensity Heavy-ion Accelerator Facility

The High-Intensity Heavy-Ion Accelerator Facility (HIAF), currently under construction in Huizhou, Guangdong Province, China, is projected to be completed by 2025. This facility will be capable of producing proton and heavy-ion beams with energies reaching several GeV, thereby offering a versatile platform for advanced fundamental physics research. Key scientific objectives include exploring physics beyond the Standard Model through the search for novel particles and interactions, testing fundamental symmetries, investigating exotic hadronic states such as di-baryons, pentaquark states and multi-strange hypernuclei, conducting precise measurements of hadron and hypernucleus properties, and probing the phase boundary and critical point of nuclear matter. To facilitate these investigations, we propose the development of a dedicated experimental apparatus at HIAF - the Huizhou Hadron Spectrometer (HHaS). This paper presents the conceptual design of HHaS, comprising a solenoid magnet, a five-dimensional silicon pixel tracker, a Low-Gain Avalanche Detector (LGAD) for time-of-flight measurements, and a Cherenkov-scintillation dual-readout electromagnetic calorimeter. The design anticipates an unprecedented event rate of 1-100 MHz, extensive particle acceptance, a track momentum resolution at 1% level, an electromagnetic energy resolution of ~3% @ 1 GeV and multi-particle identification capabilities. Such capabilities position HHaS as a powerful instrument for advancing experimental studies in particle and nuclear physics. The successful realization of HHaS is expected to significantly bolster the development of medium- and high-energy physics research within China.

hep-ex↗

arXiv2023

A Cooperative Content Dissemination Framework for Fog-Based Internet of Vehicles

As the fog-based internet of vehicles (IoV) is equipped with rich perception, computing, communication and storage resources, it provides a new solution for the bulk data processing. However, the impact caused by the mobility of vehicles brings a challenge to the content scheduling and resource allocation of content dissemination service. In this paper, we propose a time-varying resource relationship graph to model the intertwined impact of the perception, computation, communication and storage resources across multiple snapshots on the content dissemination process of IoV. Based on this graph model, the content dissemination process is modeled as a mathematical optimization problem, where the quality of service of both delay tolerant and delay sensitive services are considered. Owing to its NP-completeness, the optimization problem is decomposed into a joint link and subchannel scheduling subproblem and as well a joint power and flow control subproblem. Then, a cascaded low complexity scheduling algorithm is proposed for the joint link and subchannel scheduling subproblem. Moreover, a robust resource management algorithm is developed for the power and flow control subproblem, where the channel uncertainties in future snapshots are fully considered in the algorithm. Finally, we conduct simulations to show that the effectiveness of the proposed approaches outperforms other state-of-art approaches.

cs.NI↗

arXiv2021

Slippery Polymer Monoliths: Surface Functionalization with Ordered MoS2 Microparticle Arrays

Components of technical systems and devices often require self-lubricating properties, which are implemented by means of dry lubricants. However, continuous lubricant coatings on the components' surfaces often suffer from poor adhesion, delamination and crack propagation. The replacement of continuous coatings with dense ordered arrays of microparticles consisting of dry lubricants may overcome these drawbacks. Using the well-established solid lubricant MoS2 as model system, we demonstrate that the sliding capability of polymeric monoliths can be significantly enhanced by integration of arrays of micron-sized dry lubricant microparticles into their contact surfaces. To synthesize the MoS2 microparticle arrays, we first prepared ordered hexagonal arrays of ammonium tetrathiomolybdate (ATM) microparticles on Si wafers by molding against poly(dimethylsiloxane) templates followed by high-temperature conversion of the ATM microparticles into MoS2 microparticles under Ar/H2 atmosphere in the presence of elemental sulfur. Finally, the obtained large-scale hexagonal MoS2 microparticle arrays were transferred to the surfaces of polymer monoliths under conservation of the array ordering. Self-lubrication of components of technical systems by incorporation of dry lubricant microparticle arrays into their contact surfaces is an example for overcoming the drawbacks of continuous functional coatings by replacing them with microparticle arrays.

physics.app-ph↗

arXiv2019

Evaluation of 3D gold nanodendrite layers obtained by templated galvanic displacement reactions for SERS sensing and heterogeneous catalysis

Dense layers of overlapping three-dimensional (3D) gold nanodendrites characterized by high specific surfaces as well as by abundance of sharp edges and vertices creating high densities of SERS hotspots are promising substrates for SERS-based sensing and catalysis. We have evaluated to what extent structural features of 3D gold nanodendrite layers can be optimized by the initiation of 3D gold nanodendrite growth at gold particles rationally positioned on silicon wafers. For this purpose, galvanic displacement reactions yielding 3D gold nanodendrites were guided by hexagonal arrays of parent gold particles with a lattice constant of 1.5 micrometers obtained by solid-state dewetting of gold on topographically patterned silicon wafers. Initiation of the growth of dendritic features at edges of the gold particles resulted in the formation of 3D gold nanodendrites while limitation of dendritic growth to the substrate plane was prevented. The regular arrangement of the parent gold particles supported the formation of dense layers of overlapping 3D gold nanodendrites that were sufficiently homogeneous within the resolution limits of Raman microscopes. Consequently, SERS mapping experiments revealed a reasonable degree of uniformity. The proposed preparation algorithm comprises only bottom-up process steps that can be carried out without use of costly instrumentation.

physics.app-ph↗

arXiv2018

Free-sustaining Three-dimensional S235 Steel-based Porous Electrocatalyst for Highly Efficient and Durable Oxygen Evolution

A novel oxygen evolution reaction (OER) catalyst (3D S235-P steel) based on steel S235 substrate has been successfully prepared via a facile one-step surface modification. The standard Carbon Manganese steel was phosphorizated superficially leading to the formation of a unique 3D interconnected nanoporous surface with high specific area which facilitates the electrocatalytically initiated oxygen evolution reaction. The prepared 3D S235-P steel exhibits enhanced electrocatalytic OER activities in alkaline regime confirmed by a low overpotential (η=326 mV at j=10 mA cm-2) and a small Tafel slope of 68.7 mV dec-1. Moreover, the catalyst was found to be stable under long-term usage conditions functioning as oxygen evolving electrode at pH 13 as evidenced by the sufficient charge to oxygen conversion rate (Faradaic efficiency: 82.11% and 88.34% at 10 mA cm-2 and 5 mA cm-2, respectively). In addition, it turned out that the chosen surface modification renders steel S235 into an OER electrocatalyst sufficiently and stable to work in neutral pH condition. Our investigation revealed that the high catalytic activities are likely to stem from the generated Fe/(Mn) hydroxide/oxo-hydroxides generated during the OER process. The phosphorization treatment is therefore not only an efficient way to optimize the electrocatalytic performance of standard Carbon-Manganese steel, but also enables for the development of low cost and abundant steels in the field of energy conversion.

physics.chem-ph↗

arXiv2018

Steel-Based Electrocatalysts for Efficient and Durable Oxygen Evolution in Acidic Media

High overpotentials, particularly an issue of common anode materials, hamper the process of water electrolysis for clean energy generation. Thanks to immense research efforts up to date oxygen evolution electrocatalysts based on earth-abundant elements work efficiently and stably in neutral and alkaline regimes. However, non-noble metal-based anode materials that can withstand low pH regimes are considered to be an indispensable prerequisite for the water splitting to succeed in the future. All oxygen evolving electrodes working durably and actively in acids contain Ir at least as an additive. Due to its scarcity and high acquisition costs noble elements like Pt, Ru and Ir need to be replaced by earth abundant elements. We have evaluated a Ni containing stainless steel for use as an oxygen-forming electrode in diluted H2SO4. Unmodified Ni42 steel showed a significant weight loss after long term OER polarization experiments. Moreover, a substantial loss of the OER performance of the untreated steel specimen seen in linear sweep voltammetry measurements turned out to be a serious issue. However, upon anodization in LiOH, Ni42 alloy was rendered in OER electrocatalysts that exhibit under optimized synthesis conditions stable overpotentials down to 445 mV for 10 mA cm-2 current density at pH 0. Even more important: The resulting material has proven to be robust upon long-term usage (weight loss: 20 mug/mm2 after 50 ks of chronopotentiometry at pH 1) towards OER in H2SO4. Our results suggest that electrochemical oxidation of Ni42 steel in LiOH (sample Ni42Li205) results in the formation of a metal oxide containing outer zone that supports solution route-based oxygen evolution in acidic regime accompanied by a good stability of the catalyst.

physics.chem-ph↗

arXiv2018

Nanostructured submicron block copolymer dots by sacrificial stamping: a potential preconcentration platform for locally resolved sensing, chemistry and cellular interactions

Classical contact lithography involves patterning of surfaces by embossing or by transfer of ink. We report direct lithographic transfer of parts of sacrificial stamps onto counterpart surfaces. Using sacrificial stamps consisting of the block copolymer polystyrene-block-poly(2-pyridine) (PS-b-P2VP), we deposited arrays of nanostructured submicron PS-b-P2VP dots with heights of about 100 nm onto silicon wafers and glass slides. The sacrificial PS-b-P2VP stamps were topographically patterned with truncated-pyramidal contact elements and penetrated by spongy-continuous nanopore systems. The spongy nature of the sacrificial PS-b-P2VP stamps supported formation of adhesive contact to the counterpart surfaces and the rupture of the contact elements during stamp retraction. The submicron PS-b-P2VP dots generated by sacrificial stamping can be further functionalized, examples include loading submicron PS-b-P2VP dots with dyes and attachment of gold nanoparticles to their outer surfaces. The arrays of submicron PS-b-P2VP dots can be integrated into setups for advanced optical microscopy, total internal reflection fluorescence microscopy or Raman microscopy. Arrays of nanostructured submicron block copolymer dots may represent a preconcentration platform for locally resolved sensing and locally resolved monitoring of cellular interactions or might be used as microreactor arrays in lab-on-chip configurations.

physics.app-ph↗