Searcharxiv⌕ Search

arXiv subjects

Tianyu Jia

Publications and source records attributed to Tianyu Jia.

At least 19 recordsLinked to original sources
arXiv2026

Predictive Structure to Thermal Conductivity Modeling Framework for BEOL Interconnect Stacks in Advanced Technology Nodes Enabled by Extensive Layer Resolved Thermal Measurements

The increasing structural complexity of BEOL interconnect stacks in advanced integrated circuits demands a structure-aware thermal conductivity (\k{appa}) modeling framework. However, generalizable models derived from layer-resolved thermal measurements that quantitatively capture the dependence of \k{appa} on interconnect structures remain lacking, limiting predictive thermal analysis. Here, we establish an experimentally derived, structure-aware \k{appa} modeling framework enabled by time-domain thermoreflectance measurements with ~100 nm depth resolution. Statistical analysis of a compiled dataset comprising over 40 experimentally measured layer-resolved \k{appa} values across diverse BEOL layers reveals a generalizable empirical structure-to-\k{appa} relationship, enabling predictive modeling based on interconnect structure. An intra-layer three-dimensional \k{appa} distribution model based on effective medium theory and realistic layouts further resolves spatial \k{appa} variations within practical interconnect layers. Together, these models establish an experimentally derived, structure-aware \k{appa} modeling framework for predictive and generalizable thermal analysis of advanced interconnect stacks and 3D ICs.

cond-mat.mes-hall↗
arXiv2026

LearNAT: Learning NL2SQL with AST-guided Task Decomposition for Large Language Models

Natural Language to SQL (NL2SQL) aims to translate natural language queries into executable SQL statements, offering non-expert users intuitive access to databases. While recent approaches leveraging large-scale private LLMs such as GPT-4 have achieved state-of-the-art results, they face two critical challenges: the lack of openness and reproducibility, and the prohibitive computational cost of test-time scaling. To address these issues, we explore improving the model-level performance of small-scale public LLMs in NL2SQL under resource-constrained settings. Our exploratory experiments reveal the potential of task decomposition for enhancing NL2SQL performance, but also highlight the difficulty of enabling LLMs to decompose queries effectively. Motivated by these findings, we propose LearNAT, a novel framework designed to enhance decomposition capabilities of LLM. LearNAT introduces (1) a Decomposition Synthesis Procedure, which leverages AST-guided search with pruning strategies to generate verifiable and efficient decompositions, and (2) Margin-Aware Reinforcement Learning, which provides fine-grained preference optimization for multi-step reasoning beyond standard DPO. Extensive experiments on benchmark datasets demonstrate that LearNAT significantly improves the performance of small-scale LLMs, achieving results comparable to GPT-4 with only a 7B parameter model. These results validate the effectiveness of verifiable decomposition and fine-grained preference learning in advancing NL2SQL towards openness, transparency, and efficiency. Our code is publicly available at https://github.com/MrBlankness/LearNAT.

cs.CL↗
arXiv2026

The Weakest Link Tells It All: Outcome-Supervised Process Reward Modeling via Learnable Credit Assignment

Process reward models (PRMs) enhance the reasoning capabilities of large language models (LLMs) by providing fine-grained feedback, yet training PRMs typically requires expensive stepwise annotations. Outcome-supervised PRMs offer a scalable alternative by learning from final-answer correctness alone, but this introduces a fundamental *credit assignment* challenge, i.e., attributing outcomes to responsible reasoning steps. Existing approaches rely on either uniform or causal assignment, both of which fail to anchor credit in step correctness and thus hinder process error identification. In this work, we propose Outcome-Supervised Process Reward Modeling via **L**earnable **C**redit **A**ssignment (**LCA**), an outcome-supervised PRM framework that jointly learns credit assignment and reward modeling under the principle of *Weakest Link Assignment: a reasoning chain is as strong as its weakest link*. To address mutual dependence between credit assignment and reward modeling, we formalize outcome-supervised PRM as a Multiple Instance Learning (MIL) problem and introduce Softmax-Weighted-Sum (SWS) pooling, an MIL pooling technique tailored for strong dependence and redundancy among reasoning states. We prove Bayes consistency of our algorithm under mild assumptions. Extensive experiments demonstrate that **LCA** consistently outperforms state-of-the-art outcome-supervised PRMs across multiple tasks and backbones. Code is available at https://anonymous.4open.science/r/LCA.

cs.LG↗
arXiv2026

AccelCIM: Systematic Dataflow Exploration for SRAM Compute-in-Memory Accelerator

SRAM-based compute-in-memory (CIM) offers high computational density and energy efficiency for deep neural network (DNN) accelerators, but its limited capacity causes on/off-chip data movement overhead for large DNN models. Existing CIM accelerator studies typically assume that DNN models fit entirely on-chip, leaving efficient dataflow design largely untapped. This paper introduces AccelCIM, a systematic dataflow exploration framework for SRAM CIM accelerator, which addresses two key limitations of prior work. (1) It formulates a systematic dataflow design space spanning CIM macro configurations and macro-array organizations. (2) It introduces rigorous design evaluation using cycle-accurate architectural simulation and post-layout PPA analysis. We conduct an extensive design space exploration and apply AccelCIM to representative LLM applications, providing practical insights for the principled design of CIM accelerators.

cs.AR↗
arXiv2026

ChipLight: Cross-Layer Optimization of Chiplet Design with Optical Interconnects for LLM Training

In large-scale distributed LLM training, communication between devices becomes the key performance bottleneck. Chiplet technology can integrate multiple dies into a package to scale-up node performance with higher bandwidth. Meanwhile, optical interconnect (OI) technology offers long-reach, high-bandwidth links, making it well suited for scale-out networks. The combination of these two technologies has the potential to overcome communication bottlenecks within and across packages. In this work, we present ChipLight, a cross-layer multi-objective design and optimization method for training clusters leveraging chiplet and OI. We first abstract an architecture model for such complex clusters, co-optimizing chiplet architecture, training parallel strategy, and OI network topology. Based on such models, we tailor the design space exploration flow by combining both black-box and white-box methodologies. Evaluated by our experimental results, ChipLight achieves significantly improved training efficiency and provides valuable design insights for the development of future training clusters.

cs.AR↗
arXiv2026

Stackelberg Self-Annotation: A Robust Approach to Data-Efficient LLM Alignment

Aligning large language models (LLMs) with human preferences typically demands vast amounts of meticulously curated data, which is both expensive and prone to labeling noise. We propose Stackelberg Game Preference Optimization (SGPO), a robust alignment framework that models alignment as a two-player Stackelberg game between a policy (leader) and a worst-case preference distribution (follower). The proposed SGPO guarantees $\mathcal{O}(ε)$-bounded regret within an $ε$-Wasserstein ball, offering formal robustness to (self-)annotation noise. We instantiate SGPO with Stackelberg Self-Annotated Preference Optimization (SSAPO), which uses minimal human-labeled "seed" preferences and iteratively self-annotates new prompts. In each iteration, SSAPO applies a distributionally robust reweighting of synthetic annotations, ensuring that noisy or biased self-labels do not derail training. Remarkably, using only 2K seed preferences -- about 1/30 of standard human labels -- SSAPO achieves strong win rates against GPT-4 across multiple benchmarks within three iterations. These results highlight that a principled Stackelberg formulation yields data-efficient alignment for LLMs, significantly reducing reliance on costly human annotations.

cs.LG↗
arXiv2026

CREATE: Cross-Layer Resilience Characterization and Optimization for Efficient yet Reliable Embodied AI Systems

Embodied Artificial Intelligence (AI) has recently attracted significant attention as it bridges AI with the physical world. Modern embodied AI systems often combine a Large Language Model (LLM)-based planner for high-level task planning and a reinforcement learning (RL)-based controller for low-level action generation, enabling embodied agents to tackle complex tasks in real-world environments. However, deploying embodied agents remains challenging due to their high computation requirements, especially for battery-powered local devices. Although techniques like lowering operating voltage can improve energy efficiency, they can introduce bit errors and result in task failures. In this work, we propose CREATE, a general design principle that leverages heterogeneous resilience at different layers for synergistic energy-reliability co-optimization. For the first time, we conduct a comprehensive error injection study on modern embodied AI systems and observe an inherent but heterogeneous fault tolerance. Building upon these insights, we develop an anomaly detection and clearance mechanism at the circuit level to eliminate outlier errors. At the model level, we propose a weight-rotation-enhanced planning algorithm to improve the fault tolerance of the LLM-based planner. Furthermore, we introduce an application-level technique, autonomy-adaptive voltage scaling, to dynamically adjust the operating voltage of the controllers. The voltage scaling circuit is co-designed to enable online voltage adjustment. Extensive experiments demonstrate that without compromising task quality, CREATE achieves 40.6% computational energy savings on average over nominal-voltage baselines and 35.0% over prior-art techniques. This further leads to 29.5% to 37.3% chip-level energy savings and approximately a 15% to 30% improvement in battery life.

cs.AR↗
arXiv2026

Purely gravitational dark matter production in warm inflation

We consider an appealing scenario for the production of purely gravitational dark matter in the background of warm inflation, a mechanism that maintains stable thermal bath during inflation. Through systematic investigation of various gravitational production channels, we reveal distinctive features compared to the standard inflation scenario. Notably, the inflaton annihilation channel in warm inflation exhibits markedly different thermodynamics from the standard inflation paradigm, leading to a suppression on the production of sub-inflaton-mass dark matter. For the production channel of inflationary vacuum fluctuations, we find an abundance-mass correlation of $ρ_χ\propto m_χ^{1/2}(m_χ^{5/2})$ for the sub-Hubble-mass dark matter with minimal(conformal) coupling. Our results also indicate that a minimum temperature threshold of $10^{-6}M_P$ is necessary for warm inflation, which allows adequate dark matter production. With observational constraints, our results provide stringent limits on the mass range of purely gravitational dark matter with sufficient density: $10^{-8}-10^{-2}M_P$ for minimal coupling and $10^{-14}-10^{-2}M_P$ for conformal coupling.

astro-ph.CO↗
arXiv2025

ATMPlace: Analytical Thermo-Mechanical-Aware Placement Framework for 2.5D-IC

Rising demand in AI and automotive applications is accelerating 2.5D IC adoption, with multiple chiplets tightly placed to enable high-speed interconnects and heterogeneous integration. As chiplet counts grow, traditional placement tools, limited by poor scalability and reliance on slow simulations, must evolve beyond wirelength minimization to address thermal and mechanical reliability, critical challenges in heterogeneous integration. In this paper, we present ATMPlace, the first analytical placer for 2.5D ICs that jointly optimizes wirelength, peak temperature, and operational warpage using physics-based compact models. It generates Pareto optimal placements for systems with dozens of chiplets. Experimental results demonstrate superior performance: 146 percent and 52 percent geometric mean wirelength improvement over TAP 2.5D and TACPlace, respectively, with 3 to 13 percent lower temperature and 5 to 27 percent less warpage, all achieved approximately 10 times faster. The proposed framework is general and can be extended to enable fast, scalable, and reliable design exploration for next-generation 2.5D systems.

math.NA↗
arXiv2025

Tasa: Thermal-aware 3D-Stacked Architecture Design with Bandwidth Sharing for LLM Inference

The autoregressive decoding in LLMs is the major inference bottleneck due to the memory-intensive operations and limited hardware bandwidth. 3D-stacked architecture is a promising solution with significantly improved memory bandwidth, which vertically stacked multi DRAM dies on top of logic die. However, our experiments also show the 3D-stacked architecture faces severer thermal issues compared to 2D architecture, in terms of thermal temperature, gradient and scalability. To better exploit the potential of 3D-stacked architecture, we present Tasa, a heterogeneous architecture with cross-stack thermal optimizations to balance the temperature distribution and maximize the performance under the thermal constraints. High-performance core is designed for compute-intensive operations, while high-efficiency core is used for memory-intensive operators, e.g. attention layers. Furthermore, we propose a bandwidth sharing scheduling to improve the bandwidth utilization in such heterogeneous architecture. Extensive thermal experiments show that our Tasa architecture demonstrates greater scalability compared with the homogeneous 3D-stacked architecture, i.e. up to 5.55 $\tccentigrade$, 9.37 $\tccentigrade$, and 7.91 $\tccentigrade$ peak temperature reduction for 48, 60, and 72 core configurations. Our experimental for Llama-65B and GPT-3 66B inferences also demonstrate 2.85x and 2.21x speedup are obtained over the GPU baselines and state-of-the-art heterogeneous PIM-based LLM accelerator

cs.AR↗
arXiv2025

Intuitive control of supernumerary robotic limbs through a tactile-encoded neural interface

Brain-computer interfaces (BCIs) promise to extend human movement capabilities by enabling direct neural control of supernumerary effectors, yet integrating augmented commands with multiple degrees of freedom without disrupting natural movement remains a key challenge. Here, we propose a tactile-encoded BCI that leverages sensory afferents through a novel tactile-evoked P300 paradigm, allowing intuitive and reliable decoding of supernumerary motor intentions even when superimposed with voluntary actions. The interface was evaluated in a multi-day experiment comprising of a single motor recognition task to validate baseline BCI performance and a dual task paradigm to assess the potential influence between the BCI and natural human movement. The brain interface achieved real-time and reliable decoding of four supernumerary degrees of freedom, with significant performance improvements after only three days of training. Importantly, after training, performance did not differ significantly between the single- and dual-BCI task conditions, and natural movement remained unimpaired during concurrent supernumerary control. Lastly, the interface was deployed in a movement augmentation task, demonstrating its ability to command two supernumerary robotic arms for functional assistance during bimanual tasks. These results establish a new neural interface paradigm for movement augmentation through stimulation of sensory afferents, expanding motor degrees of freedom without impairing natural movement.

cs.RO↗
arXiv2025

LP-Spec: Leveraging LPDDR PIM for Efficient LLM Mobile Speculative Inference with Architecture-Dataflow Co-Optimization

LLM inference on mobile devices faces extraneous challenges due to limited memory bandwidth and computational resources. To address these issues, speculative inference and processing-in-memory (PIM) techniques have been explored at the algorithmic and hardware levels. However, speculative inference results in more compute-intensive GEMM operations, creating new design trade-offs for existing GEMV-accelerated PIM architectures. Furthermore, there exists a significant amount of redundant draft tokens in tree-based speculative inference, necessitating efficient token management schemes to minimize energy consumption. In this work, we present LP-Spec, an architecture-dataflow co-design leveraging hybrid LPDDR5 performance-enhanced PIM architecture with draft token pruning and dynamic workload scheduling to accelerate LLM speculative inference. A near-data memory controller is proposed to enable data reallocation between DRAM and PIM banks. Furthermore, a data allocation unit based on the hardware-aware draft token pruner is developed to minimize energy consumption and fully exploit parallel execution opportunities. Compared to end-to-end LLM inference on other mobile solutions such as mobile NPUs or GEMV-accelerated PIMs, our LP-Spec achieves 13.21x, 7.56x, and 99.87x improvements in performance, energy efficiency, and energy-delay-product (EDP). Compared with prior AttAcc PIM and RTX 3090 GPU, LP-Spec can obtain 12.83x and 415.31x EDP reduction benefits.

cs.AR↗
arXiv2025

Patch-wise Structural Loss for Time Series Forecasting

Time-series forecasting has gained significant attention in machine learning due to its crucial role in various domains. However, most existing forecasting models rely heavily on point-wise loss functions like Mean Square Error, which treat each time step independently and neglect the structural dependencies inherent in time series data, making it challenging to capture complex temporal patterns accurately. To address these challenges, we propose a novel Patch-wise Structural (PS) loss, designed to enhance structural alignment by comparing time series at the patch level. Through leveraging local statistical properties, such as correlation, variance, and mean, PS loss captures nuanced structural discrepancies overlooked by traditional point-wise losses. Furthermore, it integrates seamlessly with point-wise loss, simultaneously addressing local structural inconsistencies and individual time-step errors. PS loss establishes a novel benchmark for accurately modeling complex time series data and provides a new perspective on time series loss function design. Extensive experiments demonstrate that PS loss significantly improves the performance of state-of-the-art models across diverse real-world datasets.

cs.LG↗
arXiv2025

Structured and sparse partial least squares coherence for multivariate cortico-muscular analysis

Multivariate cortico-muscular analysis has recently emerged as a promising approach for evaluating the corticospinal neural pathway. However, current multivariate approaches encounter challenges such as high dimensionality and limited sample sizes, thus restricting their further applications. In this paper, we propose a structured and sparse partial least squares coherence algorithm (ssPLSC) to extract shared latent space representations related to cortico-muscular interactions. Our approach leverages an embedded optimization framework by integrating a partial least squares (PLS)-based objective function, a sparsity constraint and a connectivity-based structured constraint, addressing the generalizability, interpretability and spatial structure. To solve the optimization problem, we develop an efficient alternating iterative algorithm within a unified framework and prove its convergence experimentally. Extensive experimental results from one synthetic and several real-world datasets have demonstrated that ssPLSC can achieve competitive or better performance over some representative multivariate cortico-muscular fusion methods, particularly in scenarios characterized by limited sample sizes and high noise levels. This study provides a novel multivariate fusion method for cortico-muscular analysis, offering a transformative tool for the evaluation of corticospinal pathway integrity in neurological disorders.

stat.AP↗
arXiv2025

EdgeMM: Multi-Core CPU with Heterogeneous AI-Extension and Activation-aware Weight Pruning for Multimodal LLMs at Edge

Emerging multimodal LLMs (MLLMs) exhibit strong cross-modality perception and reasoning capabilities and hold great potential for various applications at edge. However, MLLMs typically consist of a compute-intensive modality encoder and a memory-bound LLM decoder, leading to distinct bottlenecks for hardware designs. In this work, we present a multi-core CPU solution with heterogeneous AI extensions, which are based on either the compute-centric systolic array or memory-centric digital compute-in-memory (CIM) co-processors. In addition, dynamic activation-aware weight pruning and bandwidth management are developed to enhance bandwidth efficiency and core utilization, improving overall performance. We implemented our solution using commercial 22nm technology. For representative MLLMs, our evaluations show EdgeMM can achieve 2.84x performance speedup compared to laptop 3060 GPU.

cs.AR↗
arXiv2025

Nonlinear dynamics of oscillons and transients during preheating after single field inflation

In the single-field model, the preheating process occurs through self-resonance of inflaton field. We study the nonlinear structures generated during preheating in the $α$-attractor models and monodromy models. The potentials have a power law form $\propto\left|ϕ\right|^{2n}$ near the origin and a flat region away from bottom, which are consistent with current cosmological observations. The Floquet analysis shows that potential parameters in monodromy model have a significant influence on the region of resonance bands. Besides we investigate the formation of nonlinear structures, the equation of state and the energy transfer through the (3+1) dimensional lattice simulation. We find that all the models exhibit similar nonlinear dynamics. Furthermore, our analysis reveals that the behavior of the inflation model at small field values, which is governed by the parameter $n$, exerts a more substantial influence on the nonlinear dynamics. The behavior at large field values, controlled by the parameter $q$, has a negligible impact.

astro-ph.CO↗
arXiv2025

Lightweight Channel-wise Dynamic Fusion Model: Non-stationary Time Series Forecasting via Entropy Analysis

Non-stationarity is an intrinsic property of real-world time series and plays a crucial role in time series forecasting. Previous studies primarily adopt instance normalization to attenuate the non-stationarity of original series for better predictability. However, instance normalization that directly removes the inherent non-stationarity can lead to three issues: (1) disrupting global temporal dependencies, (2) ignoring channel-specific differences, and (3) producing over-smoothed predictions. To address these issues, we theoretically demonstrate that variance can be a valid and interpretable proxy for quantifying non-stationarity of time series. Based on the analysis, we propose a novel lightweight \textit{C}hannel-wise \textit{D}ynamic \textit{F}usion \textit{M}odel (\textit{CDFM}), which selectively and dynamically recovers intrinsic non-stationarity of the original series, while keeping the predictability of normalized series. First, we design a Dual-Predictor Module, which involves two branches: a Time Stationary Predictor for capturing stable patterns and a Time Non-stationary Predictor for modeling global dynamics patterns. Second, we propose a Fusion Weight Learner to dynamically characterize the intrinsic non-stationary information across different samples based on variance. Finally, we introduce a Channel Selector to selectively recover non-stationary information from specific channels by evaluating their non-stationarity, similarity, and distribution consistency, enabling the model to capture relevant dynamic features and avoid overfitting. Comprehensive experiments on seven time series datasets demonstrate the superiority and generalization capabilities of CDFM.

cs.LG↗
arXiv2025

Leveraging Compute-in-Memory for Efficient Generative Model Inference in TPUs

With the rapid advent of generative models, efficiently deploying these models on specialized hardware has become critical. Tensor Processing Units (TPUs) are designed to accelerate AI workloads, but their high power consumption necessitates innovations for improving efficiency. Compute-in-memory (CIM) has emerged as a promising paradigm with superior area and energy efficiency. In this work, we present a TPU architecture that integrates digital CIM to replace conventional digital systolic arrays in matrix multiply units (MXUs). We first establish a CIM-based TPU architecture model and simulator to evaluate the benefits of CIM for diverse generative model inference. Building upon the observed design insights, we further explore various CIM-based TPU architectural design choices. Up to 44.2% and 33.8% performance improvement for large language model and diffusion transformer inference, and 27.3x reduction in MXU energy consumption can be achieved with different design choices, compared to the baseline TPUv4i architecture.

cs.AR↗