arXiv subjects

Shuhan Liu

Publications and source records attributed to Shuhan Liu.

At least 19 recordsLinked to original sources

arXiv2026

Weighting Scheme and Construction in IPCW Survival Estimation

Inverse probability of censoring weighting (IPCW) is widely used to address dependent censoring in survival analysis. Beyond estimation of the censoring distribution, its implementation requires specifying how inverse probability weights are incorporated into survival estimators. IPCW does not uniquely determine this construction, and different choices can produce different finite-sample behavior even under the same censoring model. In this work, we examine a class of IPCW estimators for the survival function, distinguishing hazard-based from margin-based formulations and time-varying from fixed weighting schemes. We show how these choices affect structural validity, stability, and finite-sample performance. In particular, the weight-changing hazard (WCH) estimator applies time-varying censoring weights within a product-limit construction with risk-set normalization, ensuring monotonicity and boundedness when the weighted risk-set denominators are positive. Through simulation studies under varying levels of covariate-dependent and administrative censoring, we demonstrate that WCH achieves favorable bias--variance tradeoffs and improved finite-sample stability relative to the alternative IPCW constructions examined, particularly in settings with few observed events or heavy censoring. We further illustrate these differences using data from the Women's Interagency HIV Study (WIHS), in which alternative IPCW constructions produced meaningfully different survival curves. Overall, our findings emphasize that the effectiveness of IPCW depends on both the weighting scheme and the estimator formulation. Careful consideration of these choices is therefore important for reliable survival estimation under dependent censoring.

stat.ME↗

arXiv2026

Global minimax rates for heterogeneous treatment effect estimation

Recent work established minimax rates for pointwise estimation of the conditional average treatment effect (CATE), but the fundamental limits under global losses remained unresolved. We derive matching minimax lower and upper bounds under the $L_2$ and $L_\infty$ risks on certain covariate-density and support classes. For the lower bounds, we develop a many-fuzzy-hypotheses method that replaces each fixed alternative in a multiple-hypothesis construction with a mixture over nuisance functions that leaves its CATE unchanged. A localized product-mixture $χ^2$ bound is used to control the divergence of each resulting mixture from a common baseline, enabling multiple well-separated CATE hypotheses to be handled simultaneously. For the upper bounds, we analyze a modified local polynomial R-learner. The $L_2$ argument extends the pointwise bias--variance analysis, while the $L_\infty$ argument combines concentration inequalities for U-statistics with a continuity argument enabled by the use of a continuous kernel and additional regularity of the second-order basis. The resulting rates show that the pointwise elbow phenomenon persists globally: the $L_2$ transition threshold is unchanged, while the $L_\infty$ rate has an additional logarithmic factor and a transition threshold adjusted by $1-\log\log n/\log n$.

math.ST↗

arXiv2026

Doubly Robust Estimation and Inference for Longitudinal Win Probabilities with Ordinal Outcomes in Observational Studies

Repeated ordinal outcomes capture how treatment benefits and harms evolve over time. We develop doubly robust estimation and inference for visit-specific win probabilities and their prespecified weighted summaries, with Desirability of Outcome Ranking as a motivating example. Baseline confounding is addressed through augmented inverse-probability weighting (AIPW) of treatment-specific marginal category probabilities or joint multinomial targeted maximum likelihood estimation (TMLE). We consider fold-specific and pooled targeting and establish conditions for double-robust consistency and semiparametric efficiency under participant-level cross-fitting. When only one nuisance model is correctly specified, inference based only on the efficient influence function may be invalid. For smooth parametric nuisance models, we derive covariance corrections from the complete estimating procedure, including targeting. When a correctly specified propensity model is estimated by maximum likelihood, a simplified score correction accommodates convergent flexible outcome learners without requiring their derivatives. The corrected covariance estimators retain within-participant dependence and support visit-specific, weighted-summary, and joint inference across repeated assessments. Simulations supported double robustness, demonstrated the importance of cross-fitting with overfit learners, and showed improved marginal and joint confidence coverage with the proposed corrections. TMLE, particularly with pooled targeting, often had lower root mean squared error under adequate treatment overlap, whereas AIPW was more stable under weak overlap. An electronic health record study of hospitalized adults with an acute respiratory illness illustrates how the trajectory and weighted summary reveal complementary aspects of the treatment comparison.

stat.ME↗

arXiv2026

SEER: Self-Enhancing Chain-of-Thought Compression for Reasoning Models

Chain-of-Thought (CoT) prompting can substantially improve the reasoning ability of large language models (LLMs), but it often comes with high inference cost due to long and poorly controlled reasoning traces. This overhead is particularly problematic in software engineering tasks (e.g., code generation), where both latency and output reliability matter. To better understand this trade-off, we conduct an empirical study on widely used code generation benchmarks and observe that many modern reasoning models produce excessively verbose CoTs (often thousands of tokens), which frequently leads to truncation and unstable generation. Using a strict n-gram repetition detector, we find that most observed truncations are associated with degenerate looping behaviors. In addition, a HumanEval/129 case study shows that failed generations can be longer than successful ones, suggesting limited returns from overlong reasoning. Motivated by these findings, we propose SEER (Self-Enhancing Efficient Reasoning), a self-enhancing framework for adaptive CoT compression. SEER improves the conciseness of reasoning while preserving output quality, without relying on external compression tools. SEER refines self-generated CoT data via Best-of-N sampling to suppress looping and redundant traces, then applies a lightweight, data-driven filter to encourage concise yet correct reasoning. It then fine-tunes the model on the filtered data to internalize concise reasoning behaviors. Across four software engineering benchmarks on the evaluated DeepSeek-R1-Distill-Qwen-7B backbone, SEER reduces CoT length by 34.6% on average while improving task performance, with reduced truncation and fewer reasoning loops.

cs.SE↗

arXiv2026

Adversarial Training for Deep Hedging in Nonstationary Markets

Deep hedging learns trading policies from historical or simulated market trajectories, yet under nonstationarity these training paths may not represent future market conditions. We propose WRAP (Wasserstein-Reweighting Adversarial Perturbation), a drift-aware adversarial training framework derived from a two-budget distributionally robust optimization (DRO) formulation. The formulation is anchored to a weighted empirical reference distribution whose fixed baseline weights are chosen to balance sampling uncertainty against temporal drift. Around this reference distribution, the ambiguity set addresses two complementary forms of distributional misspecification by allowing an adversary to reweight the observed trajectories subject to a $ϕ$-divergence constraint and perturb their paths subject to an optimal-transport (OT) constraint. We derive a joint first-order expansion in which the leading-order increase over the nominal expected loss decomposes into a reweighting contribution determined by the dispersion of hedging losses across trajectories and a transport contribution determined by the sensitivity of the loss to path perturbations. This expansion yields an explicit finite-dimensional adversarial attack that replaces the distributional inner supremum with a tractable first-order approximation. Across stationary and nonstationary Heston dynamics and a generalized affine diffusion (GAD), the experiments show complementary benefits from reweighting and transport, with joint adversarial training providing the largest gains under nonstationarity.

cs.LG↗

arXiv2026

Adaptive 3D-RoPE: Physics-Aligned Rotary Positional Encoding for Wireless Foundation Models

Wireless foundation models (WFMs) have emerged as a promising paradigm for unified channel state information (CSI) acquisition across diverse tasks in sixth-generation (6G) networks. Although WFMs significantly outperform task-specific small models, their zero-shot cross-scenario generalization still remains limited for real-world applications. Existing positional embeddings, the sole interface through which self-attention perceives the temporal-frequency-antenna 3D physical coordinates of CSI, fail to capture the highly dynamic and axis-dependent coherence inherent in wireless channels. This paper proposes Adaptive 3D-RoPE, a channel-driven 3D rotary positional embedding framework for WFMs to dynamically align the 3D positional embeddings with the instantaneous coherence state of heterogeneous CSI. The design proceeds in three stages: first, an axis-wise learnable rotary prior independently preserves the temporal, frequency, and antenna coordinate structures; second, a feature-guided rotary modulation module maps the feature-wise standard deviation of visible CSI tokens to compact, sample-adaptive scales; third, identical coordinate offsets induce dynamically adjusted query-key interactions tailored to the instantaneous channel state. Extensive experiments on both simulated and measured datasets validate the effectiveness of Adaptive 3D-RoPE across three complementary dimensions. It reduces NMSE by 10.14, 6.25, and 4.61 dB relative to baselines under antenna, temporal, and frequency scaling, respectively. It transfers effectively to real-world measured CSI and remains robust under imperfect CSI. Finally, it transfers to the independently designed LWM backbone and beam-prediction task, improving zero-shot Top-1 accuracy by 8.03 percentage points.

eess.SP↗

arXiv2026

AdaLens: Interactive Storyline for Monitoring and Steering Long-Running Agentic Data Analysis

Large language models are pushing data science toward increasingly autonomous and agentic workflows, with recent systems already supporting multi-step and long-running analyses. As these workflows become more autonomous, conventional interfaces no longer provide adequate support for two critical requirements: observability for understanding an agent's evolving reasoning and evidence, and steerability for redirecting low-value directions or deepening promising ones during execution. Existing interactive approaches improve process visibility and open intervention points, but they remain largely designed for discrete, turn-by-turn exchanges rather than the parallel branches and evolving decision structures of long-running agentic analysis. We study this need as interactive oversight in long-running agentic data analysis and present AdaLens, an interactive system for monitoring and steering ongoing runs. AdaLens combines a storyline-based representation that unifies analytical plans, execution progress, intermediate findings, and data-column involvement with steering interactions grounded in these analytical elements for directional guidance and execution control. We evaluate AdaLens through two case studies and a user study, examining how it supports analysts in monitoring and steering long-running agentic data analysis.

cs.HC↗

arXiv2026

RACE-Bench: A Reasoning-Augmented Benchmark for Repository-Level Code Agents on Feature Addition

Repository-level code agents have shown strong promise in real-world feature addition tasks, making reliable evaluation of their capabilities increasingly important. However, existing benchmarks primarily evaluate these agents as black boxes based on final test correctness, providing limited insight into how they reason and where failures arise. To address this limitation, we introduce RACE-bench, a reasoning-augmented benchmark for evaluating code agents on repository-level feature addition tasks. RACE-bench contains 528 real-world feature addition instances from 12 open-source repositories. Each instance is paired with executable patch verification and structured intermediate reference reasoning covering issue understanding, file localization, implementation tasks, and step decomposition. Based on this design, we introduce a dual-track evaluation framework that jointly measures patch correctness and intermediate reasoning alignment with developer-accepted reference trajectories. We evaluate three representative repository-level code agents on RACE-bench. On the full benchmark, Resolved Rate ranges from 29% to 70% across different agents. Our reasoning-level analysis further shows that while current agents perform well at understanding high-level intent, their performance degrades substantially when translating intent into concrete implementation steps. We also find patches that can be applied but still fail the tests cover fewer reference-reasoning elements (35.7% lower recall) and contain more unsupported reasoning elements (94.1% higher over-prediction) than successful patches. These findings highlight the importance of evaluating repository-level code agents beyond final patch correctness by examining the quality of their reasoning processes.

cs.SE↗

arXiv2026

Layer-Number-Controlled Symmetry Breaking and Surface-State Transport in Rhombohedral Graphene Multilayers

Rhombohedral multilayer graphene hosts layer-polarized flat bands, providing an intriguing platform for correlated and topological electronic states; however, the role of layer number in governing symmetry breaking and surface screening remains elusive. Here we prepare rhombohedral graphene multilayers and systematically conduct electrical transport measurements. We uncover an unconventional layer dependence of phase transitions: the critical displacement field (D$_{c}$) for the layer-antiferromagnetic (LAF)-to-semimetal transitions remains constant across tetralayer to hexalayer graphene, whereas the D$_{c}$ for semimetal-to-layer-polarized-insulator (LPI) transition increases with layer number, defying unscreened Coulomb interaction models. In hexalayer graphene, surface-state-dominated transport emerges, with Landau levels (LLs) and resistive peaks selectively controlled by adjacent gates, a signature of strong interlayer screening absent in thinner stacks. High magnetic fields reveal valley-layer-locked LLs and dissipative states possibly from interlayer backscattering, highlighting the presence of decoupled surface states. Our findings establish layer number as a key tuning knob for engineering correlated and topological phases in rhombohedral graphene multilayers.

cond-mat.mes-hall↗

arXiv2026

POEM: Phase-Aware $\mathrm{SO}(2)$ Feature Rotation for Time Series Forecasting Under Periodicity Drift

Deep learning has advanced time series forecasting, but periodicity drift, in which cycle timing and phase vary over time, remains a challenging problem. Existing methods predominantly model these sequences on fixed time grids, suffering from a limited ability to accommodate phase-related variation. To address this limitation, we propose \textbf{POEM}, a phase-aware forecasting framework based on latent feature rotation using the special orthogonal group in two dimensions, denoted by $\mathrm{SO}(2)$. POEM aims to reduce the phase-related variability by learning a phase-correction coordinate and applying an invertible $\mathrm{SO}(2)$-based rotation to paired latent features. To extrapolate this correction coordinate, Directional Phase Increment Attention (DPIA) retrieves historical phase increments from similar temporal contexts and integrates them into future phase corrections. Experiments demonstrate that POEM achieves competitive performance, while qualitative visualizations suggest that the learned phase-aware transformation makes latent trajectories more regular.

cs.LG↗

arXiv2026

Mitigating Package Hallucinations in Large Language Models via Model Editing

Large language models (LLMs) have demonstrated strong capabilities in software engineering tasks, such as code generation, library recommendation, and dependency configuration. However, recent studies show that LLMs may suffer from package hallucination, where they generate non-existent or invalid package names. These hallucinations can be exploited in software supply chain attacks, as attackers may register malicious packages under hallucinated names. Therefore, mitigating package hallucination is important for improving the reliability and security of LLM-assisted software development. In this paper, we introduce BOUND, a lightweight localized model editing framework for mitigating package hallucinations in LLMs. BOUND formulates package hallucination mitigation as a package-validity boundary editing problem, where the boundary refers to the model's ability to distinguish valid packages from hallucinated package names under a given task context. It first locates modules related to package hallucination through a risk-aware localization strategy, and then edits these modules with lightweight LoRA adapters using a boundary-aware objective that reinforces valid packages, suppresses hallucinated packages, and preserves locality behavior. Experimental results show that BOUND effectively reduces package hallucinations while preserving valid package recommendations. In the package recommendation task, BOUND reduces package-level hallucination rate (Package-HR) by 79.9% on edit prompts and by 65.4% on unseen prompts. The learned package-validity boundary further generalizes to other package-related tasks, reducing Package-HR by 12.8% in code generation and by 34.0% in pip install recommendation. These results show that BOUND refines the package-validity boundary of LLMs and improves the reliability of package-related outputs.

cs.SE↗

arXiv2026

Towards CSI-Native Foundation Models: A Channel-Adaptive Roadmap for 6G

Wireless foundation models offer a path toward reusable channel state information (CSI) intelligence for sixth-generation (6G) systems. However, existing generic-backbone adaptation and CSI pretraining methods often treat CSI as task tensors rather than propagation-conditioned channel responses, thereby failing to capture the intrinsic time-frequency-spatial geometry of wireless environments. This paper presents a channel-adaptive roadmap toward CSI-native foundation models, proposing a unified framework that aligns pretraining, positional modeling, and attention control with three channel requirements: scale-aware heterogeneous exposure, physical time-frequency-antenna coordinates, and correlation-bounded token interaction. Extensive experiments demonstrate the superiority of the proposed framework across three dimensions: zero-shot generalization, reducing NMSE by more than 4 dB across spatial-temporal-frequency tasks; scale extrapolation, yielding up to a 5.4 dB gain under 8 times unseen antenna scaling; and inference efficiency, accelerating mobility-aware processing by up to 18.8%. A system-level evaluation with Sionna SYS further shows that the proposed framework uses only 7.01% of dense-pilot overhead, reaches -18.64 dB average NMSE, and improves average net spectral efficiency by 36.6% over dense LMMSE and 15.5% over WiFo, indicating that CSI-native representation learning can support pilot-efficient radio access.

cs.LG↗

arXiv2026

Dynamic TMoE: A Drift-Aware Dynamic Mixture of Experts Framework for Non-Stationary Time Series Forecasting

Non-stationary time series forecasting is challenged by evolving distribution shifts that static models struggle to capture. While Mixture-of-Experts (MoE) architectures offer a promising paradigm for decoupling complex drift patterns, existing approaches are limited by fixed expert pools and memoryless routing, hampering their ability to adapt to abrupt regime shifts. To address this, we propose Dynamic TMoE, a framework that unifies architectural evolution with temporal continuity during learning phase. By detecting distribution shifts via Maximum Mean Discrepancy (MMD), we dynamically instantiate heterogeneous experts and prune redundant ones to optimize capacity. Additionally, a temporal memory router leverages recurrent states and an anomaly repository to ensure stable, context-aware expert selection without requiring test-time updates. Experiments on nine benchmarks demonstrate state-of-the-art performance, reducing MSE by 10.4% and MAE by 7.8%. Code is available at https://github.com/andone-07/Dynamic-TMoE.

cs.LG↗

arXiv2026

Heterogeneous Memory Design Exploration for AI Accelerators with a Gain Cell Memory Compiler

As memory increasingly dominates system cost and energy, heterogeneous on-chip memory systems that combine technologies with complementary characteristics are becoming essential. Gain Cell RAM (GCRAM) offers higher density, lower power, and tunable retention, expanding the design space beyond conventional SRAM. To this end, we create an OpenGCRAM compiler supporting both SRAM and GCRAM. It generates macro-level designs and layouts for commercial CMOS processes and characterizes area, delay, and power across user-defined configurations. The tool enables systematic identification of optimal heterogeneous memory configurations for AI tasks under specified performance metrics.

cs.AR↗

arXiv2026

HeterCSI: Channel-Adaptive Heterogeneous CSI Pretraining Framework for Generalized Wireless Foundation Models

Wireless foundation models promise transformative capabilities for channel state information (CSI) processing across diverse 6G network applications, yet face fundamental challenges due to the inherent dual heterogeneity of CSI across both scale and scenario dimensions. However, current pretraining approaches either constrain inputs to fixed dimensions or isolate training by scale, limiting the generalization and scalability of wireless foundation models. In this paper, we propose HeterCSI, a channel-adaptive pretraining framework that reconciles training efficiency with robust cross-scenario generalization via a new understanding of gradient dynamics in heterogeneous CSI pretraining. Our key insight reveals that CSI scale heterogeneity primarily causes destructive gradient interference, while scenario diversity actually promotes constructive gradient alignment when properly managed. Specifically, we formulate heterogeneous CSI batch construction as a partitioning optimization problem that minimizes zero-padding overhead while preserving scenario diversity. To solve this, we develop a scale-aware adaptive batching strategy that aligns CSI samples of similar scales, and design a double-masking mechanism to isolate valid signals from padding artifacts. Extensive experiments on 12 datasets demonstrate that HeterCSI establishes a generalized foundation model without scenario-specific finetuning, achieving superior average performance over full-shot baselines. Compared to the state-of-the-art zero-shot benchmark WiFo, it reduces NMSE by 7.19 dB, 4.08 dB, and 5.27 dB for CSI reconstruction, time-domain, and frequency-domain prediction, respectively. The proposed HeterCSI framework also reduces training latency by 53% compared to existing approaches while improving generalization performance by 1.53 dB on average.

cs.LG↗

arXiv2025

CREME: Robustness Enhancement of Code LLMs via Layer-Aware Model Editing

Large language models (LLMs) have demonstrated impressive capabilities in code generation, where the natural language prompt plays a crucial role in conveying user intent to the model. However, prior studies have shown that LLMs are highly sensitive to prompt perturbations. Minor modifications in wording, syntax, or formatting can significantly reduce the functional correctness of generated code. As perturbations frequently occur in real-world scenarios, improving the robustness of LLMs to prompt perturbations is essential for ensuring reliable performance in practical code generation. In this paper, we introduce CREME (Code Robustness Enhancement via Model Editing), a novel approach that enhances LLM robustness through targeted parameter updates. CREME first identifies robustness-sensitive layers by comparing hidden states between an original prompt and its perturbed variant. Then, it performs lightweight parameter editing at the identified layer to reduce performance degradation. We evaluate CREME on two widely used code generation benchmarks (HumanEval and MBPP) along with their perturbed counterparts. Experimental results show that CREME improves Pass@1 accuracy by 63% on perturbed prompts while maintaining stable performance on clean inputs, with accuracy deviations within 1%. Further analysis reveals that robustness-sensitive layers are primarily concentrated in the middle and deeper layers of the network, and their locations vary across different model architectures. These insights provide a valuable foundation for developing future robustness-oriented editing strategies.

cs.SE↗

arXiv2025

Gate Dielectric Engineering with an Ultrathin Silicon-oxide Interfacial Dipole Layer for Low-Leakage Oxide-Semiconductor Memories

We demonstrate a gate dielectric engineering approach leveraging an ultrathin, atomic layer deposited (ALD) silicon oxide interfacial layer (SiL) between the amorphous oxide semiconductor (AOS) channel and the high-k gate dielectric. SiL positively shifts the threshold voltage (V$_T$) of AOS transistors, providing at least four distinct $V_T$ levels with a maximum increase of 500 mV. It achieves stable $V_T$ control without significantly degrading critical device parameters such as mobility, on-state current, all while keeping the process temperature below 225 $^{\circ}$C and requiring no additional heat treatment to activate the dipole. Positive-bias temperature instability tests at 85 $^{\circ}$C indicate a significant reduction in negative $V_{T}$ shifts for SiL-integrated devices, highlighting enhanced reliability. Incorporating this SiL gate stack into two-transistor gain-cell (GC) memory maintains a more stable storage node voltage ($V_{SN}$) (reduces $V_{SN}$ drop by 67\%), by limiting unwanted charge losses. SiL-engineered GCs also reach retention times up to 10,000 s at room temperature and reduce standby leakage current by three orders of magnitude relative to baseline device, substantially lowering refresh energy consumption.

cond-mat.mtrl-sci↗

arXiv2025

The Future of Memory: Limits and Opportunities

Memory latency, bandwidth, capacity, and energy increasingly limit performance. In this paper, we reconsider proposed system architectures that consist of huge (many-terabyte to petabyte scale) memories shared among large numbers of CPUs. We argue two practical engineering challenges, scaling and signaling, limit such designs. We propose the opposite approach. Rather than create large, shared, homogenous memories, systems explicitly break memory up into smaller slices more tightly coupled with compute elements. Leveraging advances in 2.5D/3D integration, this compute-memory node provisions private local memory, enabling accesses of node-exclusive data through micrometer-scale distances, and dramatically reduced access cost. In-package memory elements support shared state within a processor, providing far better bandwidth and energy-efficiency than DRAM, which is used as main memory for large working sets and cold data. Hardware making memory capacities and distances explicit allows software to efficiently compose this hierarchy, managing data placement and movement.

cs.AR↗