Yuan Li

The paper proposes PG-RSSNN, a physics-guided recurrent state-space neural network that improves multi-step prediction stability and accuracy compared to both pure black-box and pure physics models, especially with limited data.

NeuroArmor is a white-box runtime defense that uses prompt-specific safe variants to selectively detect and mitigate jailbreak attacks, significantly reducing attack success rates while maintaining a low false positive rate.

The paper proposes OneReason, a framework that enhances the reasoning capability of generative recommendation models by focusing on improving item perception and structuring user behavior into coherent latent interests.

This paper proposes CUTh-Solver, a GPU-accelerated Preconditioned Conjugate Gradient (PCG)-based sparse solver framework for high-resolution 3D IC thermal simulation, achieving significant speedup over existing methods.

This paper derives closed-form outage probability expressions and proposes a fairness optimization algorithm for a multiuser MISO downlink system that combines rate-splitting multiple access and fluid antenna systems.

PAPERCLAW is a multi-agent system that autonomously curates a domain, generates ideas, and writes venue-compliant papers using large language models and a stoppable hypothesis map.

This paper proposes Trellis, a data foundation that treats experience graphs from long-horizon agentic tasks as first-class, governed, queryable database state.

This paper introduces GEAR, a method for training a vector-quantized tokenizer and an autoregressive generator jointly and end-to-end, resolving the issue of non-differentiable VQ indices.

This paper proposes eBIM, a software-hardware collaborative paradigm for blockchain infrastructure management using RISC-V, and surveys related research and technologies.

This paper explains how Rotary Position Embeddings (RoPE) frequencies in transformer models correspond to the relative-distance structure of training data, and the implications for long-context generalization.

QuReC is a unified framework for all-in-one image restoration using a Degradation-Guided Query Reconstruction Module and a Local-Global Response Calibration Module.

This tutorial-style review focuses on kinematics of inertial measurement units (IMUs) for human motion capture and introduces methods to determine adequate formulations for sensor measurements.

This paper introduces a new approach for high-dimensional one-step generation using a Three-Body Scattering Model (TBSM) with a proper distributional energy.

This paper proposes BlurDriving, a personalized blur system for reducing visual overload in distracted driving using a VR simulator, and evaluates its effectiveness through user studies.

This paper proposes DeltaGate, a lightweight output-layer plugin for sequential recommendation models that preserves backbone embeddings during long user gaps.

This paper proposes a Neural-Collapse-Inspired Prioritization (NCIP) framework for reducing model validation costs in safety-critical domains by using cross-checkpoint prediction variability instead of absolute confidence.

The paper proposes ParticleGS, a visualization-aware framework for compressing large-scale particle simulation data using 3D Gaussian Splatting, achieving higher rendered image quality and faster rendering speed than existing methods.

The paper presents CHASE, an application-driven framework that explores physically feasible Cross-layer Heterogeneous System architectures for executing workloads with diverse requirements.

This paper introduces TWICE, a framework for predicting long-horizon conversion rates in online advertising using two clocks with complementary supervision.

The paper proposes SharpRec, a framework for LLM-based Cross-Domain Sequential Recommendation to address the bottlenecks of cross-domain knowledge conflict and performance saturation in multi-domain fusion.