Jie Zhang

The paper argues that watermarking must be viewed as a monitoring primitive, introducing an observer-based threat model that shows even zero-bit watermarking can enable entity-level attribution through signal aggregation.

The paper introduces Prosecution Decision Prediction (PDP), a new legal AI task that assesses prosecutorial review decisions, showing that current state-of-the-art LLMs perform significantly worse on this task than on standard judgment prediction.

Qwen-VLA introduces a unified embodied foundation model that extends vision-language understanding to continuous action generation, enabling robust, multi-task generalization across diverse robotic tasks and embodiments.

LongTraceRL addresses long-context reasoning challenges by generating highly challenging training data and introducing a fine-grained rubric reward, significantly improving evidence-grounded reasoning in LLMs.

FlowTime proposes a novel Continuous Generative Regression framework using a Flow-based Personalized Prior to accurately model the multimodal and heterogeneous nature of user watch time prediction, significantly outperforming existing state-of-the-art methods.

The paper introduces DEFT, a novel Mixture-of-Experts DRL architecture, to intelligently schedule dynamic cloud workflows with varying deadlines, significantly improving performance over existing single-path schedulers.

The paper introduces Hierarchical Adaptive Budgeter (HAB), a framework that improves LLM reasoning efficiency by adaptively allocating computational resources to match the intrinsic complexity of both problems and individual reasoning steps.

STaR-KV introduces a novel, training-free KV cache compression framework that adaptively re-weights token importance across spatial, temporal, and distributional axes, significantly reducing GPU memory usage for GUI vision-language models while maintaining high accuracy.

The paper introduces and analyzes cross-session stored prompt injection, demonstrating that persistent system state transforms prompt injection from a temporary model-level threat into a long-lived, system-level vulnerability in agentic systems.

The paper introduces AgentCyberRange, an open, multi-range infrastructure for measuring autonomous cyber attack capability in realistic cyber ranges, and evaluates six frontier AI systems.

This paper formulates receiver selection for large-aperture wave receivers as a modal-recoverability problem and proposes a framework with three nested recoverability criteria for individual modal degrees, joint target subspace, and target modes.

This paper proposes SkillOpt-Lite, a minimal viable pipeline for skill optimization in autonomous agents, which accelerates convergence and outperforms full SkillOpt.

This paper introduces Camera-Centric VLA, a new model for Vision-Language-Action policies that predicts camera-centric actions and hand-eye matrix, allowing the policy to figure out camera geometry on its own.

The paper introduces LingBot-World 2.0, an advanced version of a language model with unbounded interaction horizon, rapid response time, diverse interactive elements, and agentic harness integration.

This paper proposes TRM-Raft, a Byzantine-resistant enhancement for Raft consensus that integrates a Blockchain-based Trust and Reputation Model to prevent election forgery and log tampering.

The paper introduces Move First, Commit Later, a selective layer for Global LiDAR-to-BIM initialization that decides whether to commit to a registration based on evidence from multiple submaps, reducing the issue of confident aliasing.

This paper proposes a framework for integrated sensing and communication (ISAC) using mMIMO OFDM and ISAR imaging, developing an adaptive ADMM algorithm for high-resolution image recovery and a joint resource-allocation method.

This paper proposes an end-to-end Markov framework for auditory attention decoding using conditional random fields and an EEG--speech correlation backbone.

This paper proposes SHA-PF, a search hardness-aware LLM-based problem formulation framework for expensive simulation-driven design, which prioritizes rare samples with greater progress potential and requires significantly fewer evaluations to reach design requirements.

The paper introduces VLM-IE3D, a framework that enhances 2D vision-language models with implicit and explicit 3D geometries learned from RGB videos, achieving superior performance on various 3D tasks.