arXiv subjects

Jiaqi Liu

Publications and source records attributed to Jiaqi Liu.

At least 73 records · Page 4Linked to original sources

arXiv2026

Linear-optical test of quantum contextuality with sequential measurements

Quantum contextuality provides a fundamental signature of nonclassical behavior that cannot be explained by noncontextual hidden-variable models. We propose and experimentally implement a linear-optical setup for demonstrating Kochen-Specker contextuality via a violation of the KCBS inequality using single photons. Our scheme employs sequential measurements realized with linear-optical networks and on-off photodetectors. The construction ensures that each co-measured observable is implemented by the same physical operation across different contexts. Our experimental results demonstrate a clear violation of the KCBS inequality and robustness against photon loss. Beyond fundamental investigations, the proposed setup provides a practical tool for probing non-classicality and photon-number statistics of quantum states, which in turn enables the verification of single-photon sources.

quant-ph↗

arXiv2026

Market Efficiency and the Heterogeneous Impact of Financial Liberalization: Evidence from the Shanghai-Hong Kong Stock Connect

This paper examines how the Shanghai-Hong Kong Stock Connect (SHHK Stock Connect) affects the A-H share price premium and whether the policy effect depends on pre-existing market efficiency. Using monthly data for 67 A-H dual-listed firms from 2011 to 2019, we estimate dynamic models using system GMM. We find that the implementation of SHHK Stock Connect is associated with an average 18.4% increase in the A-H price premium, indicating a widening rather than a narrowing of the A-H share price premium. More importantly, this effect is heterogeneous: the policy impact is more pronounced for firms operating in less efficient markets and weaker for those with higher efficiency. We find no significant pricing response at the announcement stage. These findings imply that financial liberalization does not necessarily promote cross-market price convergence; its effects depend on pre-existing market efficiency, suggesting that cross-border liberalization policies may be more effective when accompanied by improvements of market efficiency.

q-fin.GN↗

arXiv2026

ClawArena: Benchmarking AI Agents in Evolving Information Environments

AI agents deployed as persistent assistants must maintain correct beliefs as their information environment evolves. In practice, evidence is scattered across heterogeneous sources that often contradict one another, new information can invalidate earlier conclusions, and user preferences surface through corrections rather than explicit instructions. Existing benchmarks largely assume static, single-authority settings and do not evaluate whether agents can keep up with this complexity. We introduce ClawArena, a benchmark for evaluating AI agents in evolving information environments. Each scenario maintains a complete hidden ground truth while exposing the agent only to noisy, partial, and sometimes contradictory traces across multi-channel sessions, workspace files, and staged updates. Evaluation is organized around three coupled challenges: multi-source conflict reasoning, dynamic belief revision, and implicit personalization, whose interactions yield a 14-category question taxonomy. Two question formats, multi-choice (set-selection) and shell-based executable checks, test both reasoning and workspace grounding. ClawArena comprises 12 multi-turn scenarios spanning 337 evaluation rounds with 45 dynamic updates, evaluated across five agent frameworks and 18 language models from proprietary, community-accessible, and self-hosted sources. Experiments show that model capability accounts for a 29-point score range across models while framework design accounts for up to a 24-point range, that MetaClaw's skill overlay reliably improves score without degrading accuracy, and that belief revision difficulty is determined by update design strategy rather than update volume. Code is available at https://github.com/aiming-lab/ClawArena.

cs.LG↗

arXiv2026

EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agents

Long-term memory is essential for LLM agents that operate across multiple sessions, yet existing memory systems treat retrieval infrastructure as fixed: stored content evolves while scoring functions, fusion strategies, and answer-generation policies remain frozen at deployment. We argue that truly adaptive memory requires co-evolution at two levels: the stored knowledge and the retrieval mechanism that queries it. We present EvolveMem, a self-evolving memory architecture that exposes its full retrieval configuration as a structured action space optimized by an LLM-powered diagnosis module. In each evolution round, the module reads per-question failure logs, identifies root causes, and proposes targeted configuration adjustments; a guarded meta-analyzer applies them with automatic revert-on-regression and explore-on-stagnation safeguards. This closed-loop self-evolution realizes an AutoResearch process: the system autonomously conducts iterative research cycles on its own architecture, replacing manual configuration tuning. Starting from a minimal baseline, the process converges autonomously, discovering effective retrieval strategies including entirely new configuration dimensions not present in the original action space. On LoCoMo, EvolveMem outperforms the strongest baseline by 25.7% relative and achieves a 78.0% relative improvement over the minimal baseline. On MemBench, EvolveMem exceeds the strongest baseline by 18.9% relative. Evolved configurations transfer across benchmarks with positive rather than catastrophic transfer, indicating that the self-evolution process captures universal retrieval principles rather than benchmark-specific heuristics. Code is available at https://github.com/aiming-lab/SimpleMem.

cs.LG↗

arXiv2026

STEPS: A Temporal Smooth Error Propagation Solver on the Manifolds for Test-Time Adaptation in Time Series Forecasting

Test-Time Adaptation (TTA) aims to improve time series forecasting under distribution shifts by using limited observations revealed during inference. However, forecasting TTA must operate in a source-free online setting, where the adaptation signal is short, temporally correlated, and potentially noisy. Existing methods can therefore suffer from weak identifiability, error accumulation, and unstable long-horizon corrections when the revealed prefix is sparse or contaminated. To address these issues, we propose STEPS, a Smooth Temporal Error Propagation Solver for TTA in time-series forecasting. STEPS reformulates forecasting TTA as a Dirichlet Boundary Value Problem on a temporal manifold, where the revealed prefix error serves as the boundary condition for the unknown future error field. Then, STEPS solves a smooth and bounded correction field in prediction space: a Local Solver propagates prefix errors under temporal smoothness, a Global Solver retrieves stable cross-window error memory and Spatiotemporal Manifold Fusion (SMF) integrates both solutions into the final correction. Across six standard benchmarks and four frozen backbones, STEPS achieves an average relative MSE reduction of 26.82% over the zero-shot backbone, exceeding the strongest compared TTA baseline by 12.77%. Additional sparse prefix and contamination tests confirm the robustness of STEPS under limited and noisy prefixes.

cs.LG↗

arXiv2026

"Help Me, But Don't Track Me": Intervention Timing and Privacy Boundaries for Process-Aware AI Tutors

As generative AI (GenAI) tools are increasingly used as informal tutors for mathematics learning, future systems may become more proactive and process-aware in deciding when and how to offer support. Yet such support raises an important design tension: help that is timely may also feel interruptive or overly monitoring. To inform the design of process-aware AI tutors, we surveyed 330 secondary school students in China (Grades 7--11) about their preferred tutoring behaviors, attitudes toward proactive intervention, and acceptable use of learning-process data. We found three design-relevant patterns. First, students preferred autonomy-preserving support, such as hints over direct answers. Second, they favored graduated proactive support over constant interruption, preferring small hints first and stronger assistance only as needed. Third, they drew clear privacy boundaries around learning-process data: students were comfortable with problem-solving steps and mistake patterns, but substantially less comfortable with attention- or behavior-related signals. Together, these findings offer early empirical guidance for designing AI tutors that balance timely support with learner agency, and personalization with perceived privacy boundaries in K-12 contexts.

cs.HC↗

arXiv2026

Gradient blowup of smooth vacuum solutions to 1D compressible Euler equations

We consider the isentropic compressible Euler equations in the half-line which govern the motion of gaseous fluids in contact with stationary vacuum boundary. We construct a large class of solutions that are initially smooth and square-integrable, and which, in finite time, transition to $C^{1-μ}$ regularity for $μ\in [1/2,1)$ near the boundary, leading to the gradient blowup at the boundary. It is based on stability analysis of self-similar waiting time solutions \cite{JLN2025} recently constructed by the authors.

math.AP↗

arXiv2026

ASBench: Image Anomalies Synthesis Benchmark for Anomaly Detection

Anomaly detection plays a pivotal role in manufacturing quality control, yet its application is constrained by limited abnormal samples and high manual annotation costs. While anomaly synthesis offers a promising solution, existing studies predominantly treat anomaly synthesis as an auxiliary component within anomaly detection frameworks, lacking systematic evaluation of anomaly synthesis algorithms. Current research also overlook crucial factors specific to anomaly synthesis, such as decoupling its impact from detection, quantitative analysis of synthetic data and adaptability across different scenarios. To address these limitations, we propose ASBench, the first comprehensive benchmarking framework dedicated to evaluating anomaly synthesis methods. Our framework introduces four critical evaluation dimensions: (i) the generalization performance across different datasets and pipelines (ii) the ratio of synthetic to real data (iii) the correlation between intrinsic metrics of synthesis images and anomaly detection performance metrics , and (iv) strategies for hybrid anomaly synthesis methods. Through extensive experiments, ASBench not only reveals limitations in current anomaly synthesis methods but also provides actionable insights for future research directions in anomaly synthesis

cs.CV↗

arXiv2026

Hidden in Plain Sight: Visual-to-Symbolic Analytical Solution Inference from Field Visualizations

Recovering analytical solutions of physical fields from visual observations is a fundamental yet underexplored capability for AI-assisted scientific reasoning. We study visual-to-symbolic analytical solution inference (ViSA) for two-dimensional linear steady-state fields: given field visualizations (and first-order derivatives) plus minimal auxiliary metadata, the model must output a single executable SymPy expression with fully instantiated numeric constants. We introduce ViSA-R2 and align it with a self-verifying, solution-centric chain-of-thought pipeline that follows a physicist-like pathway: structural pattern recognition solution-family (ansatz) hypothesis parameter derivation consistency verification. We also release ViSA-Bench, a VLM-ready synthetic benchmark covering 30 linear steady-state scenarios with verifiable analytical/symbolic annotations, and evaluate predictions by numerical accuracy, expression-structure similarity, and character-level accuracy. Using an 8B open-weight Qwen3-VL backbone, ViSA-R2 outperforms strong open-source baselines and the evaluated closed-source frontier VLMs under a standardized protocol.

cs.AI↗

arXiv2026

Evaluation as Evolution: Transforming Adversarial Diffusion into Closed-Loop Curricula for Autonomous Vehicles

Autonomous vehicles in interactive traffic environments are often limited by the scarcity of safety-critical tail events in static datasets, which biases learned policies toward average-case behaviors and reduces robustness. Existing evaluation methods attempt to address this through adversarial stress testing, but are predominantly open-loop and post-hoc, making it difficult to incorporate discovered failures back into the training process. We introduce Evaluation as Evolution ($E^2$), a closed-loop framework that transforms adversarial generation from a static validation step into an adaptive evolutionary curriculum. Specifically, $E^2$ formulates adversarial scenario synthesis as transport-regularized sparse control over a learned reverse-time SDE prior. To make this high-dimensional generation tractable, we utilize topology-driven support selection to identify critical interacting agents, and introduce Topological Anchoring to stabilize the process. This approach enables the targeted discovery of failure cases while strictly constraining deviations from realistic data distributions. Empirically, $E^2$ improves collision failure discovery by 9.01% on the nuScenes dataset and up to 21.43% on the nuPlan dataset over the strongest baselines, while maintaining low invalidity and high realism. It further yields substantial robustness gains when the resulting boundary cases are recycled for closed-loop policy fine-tuning.

cs.RO↗

arXiv2026

Towards Safe and Robust Autonomous Vehicle Platooning: A Self-Organizing Cooperative Control Framework

In hybrid traffic environments where human-driven vehicles (HDVs) and autonomous vehicles (AVs) coexist, achieving safe and robust decision-making for AV platooning remains a complex challenge. Existing platooning systems often struggle with dynamic formation management and adaptability, especially under complex and dynamic mixed-traffic conditions. To enhance autonomous vehicle platooning within these hybrid environments, this paper presents TriCoD, a twin-world safety-enhanced Data-Model-Knowledge Triple-Driven Cooperative Decision-making Framework. This framework integrates deep reinforcement learning (DRL) with model-driven approaches, enabling dynamic formation dissolution and reconfiguration through a safety-prioritized twin-world deduction mechanism. The DRL component augments traditional model-driven methods, enhancing both safety and operational efficiency, especially under emergency conditions. Additionally, an adaptive switching mechanism allows the system to seamlessly switch between data-driven and model-driven strategies based on real-time traffic demands, thus optimizing decision-making ability and adaptability. Simulation experiments and hardware-in-the-loop tests demonstrate that the proposed framework significantly improves safety, robustness, and flexibility.

cs.RO↗

arXiv2026

Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memory

AI agents increasingly operate over extended time horizons, yet their ability to retain, organize, and recall multimodal experiences remains a critical bottleneck. Building effective lifelong memory requires navigating a vast design space spanning architecture, retrieval strategies, prompt engineering, and data pipelines; this space is too large and interconnected for manual exploration or traditional AutoML to explore effectively. We deploy an autonomous research pipeline to discover Omni-SimpleMem, a unified multimodal memory framework for lifelong AI agents. Starting from a naïve baseline (F1=0.117 on LoCoMo), the pipeline autonomously executes ${\sim}50$ experiments across two benchmarks, diagnosing failure modes, proposing architectural modifications, and repairing data pipeline bugs, all without human intervention in the inner loop. The resulting system achieves state-of-the-art on both benchmarks, improving F1 by +411% on LoCoMo (0.117$\to$0.598) and +214% on Mem-Gallery (0.254$\to$0.797) relative to the initial configurations. Critically, the most impactful discoveries are not hyperparameter adjustments: bug fixes (+175%), architectural changes (+44%), and prompt engineering (+188% on specific categories) each individually exceed the cumulative contribution of all hyperparameter tuning, demonstrating capabilities fundamentally beyond the reach of traditional AutoML. We provide a taxonomy of six discovery types and identify four properties that make multimodal memory particularly suited for autoresearch, offering guidance for applying autonomous research pipelines to other AI system domains. Code is available at this https://github.com/aiming-lab/SimpleMem.

cs.AI↗

arXiv2026

GeoDiT: A Diffusion-based Vision-Language Model for Geospatial Understanding

Autoregressive models are structurally misaligned with the inherently parallel nature of geospatial understanding, forcing a rigid sequential narrative onto scenes and fundamentally hindering the generation of structured and coherent outputs. We challenge this paradigm by reframing geospatial generation as a parallel refinement process, enabling a holistic, coarse-to-fine synthesis that resolves all semantic elements simultaneously. To operationalize this, we introduce GeoDiT, the first diffusion-based vision-language model tailored for the geospatial domain. Extensive experiments demonstrate that GeoDiT establishes a new state-of-the-art on benchmarks requiring structured, object-centric outputs. It achieves significant gains in image captioning, visual grounding, and multi-object detection, precisely the tasks where autoregressive models falter. Our work validates that aligning the generative process with the data's intrinsic structure is key to unlocking superior performance in complex geospatial analysis.

cs.CV↗

arXiv2026

Ciphertext-Policy ABE for $\mathsf{NC}^1$ Circuits with Constant-Size Ciphertexts from Succinct LWE

We construct a lattice-based ciphertext-policy attribute-based encryption (CP-ABE) scheme for $\mathsf{NC}^1$ access policies with constant-size ciphertexts. Let $λ$ be the security parameter. For an $\mathsf{NC}^1$ circuit of depth $d$ and size $s$ on $\ell$-bit inputs, our scheme has the public-key and ciphertext sizes $O(1)$ (independent of $d$), and secret-key size $O(\ell)$, where the $O(\cdot)$ hides $\operatorname{poly}(λ)$ factors. As an application, we obtain a broadcast encryption scheme for $N$ users with ciphertext size $\operatorname{poly}(λ)$ independent of $\log N$ and key sizes $\operatorname{poly}(λ,\log N)$. Our construction is selectively secure in the standard model under the $\operatorname{poly}(λ)$-succinct LWE assumption introduced by Wee (CRYPTO~2024).

cs.CR↗

arXiv2026

MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wild

Large language model (LLM) agents are increasingly used for complex tasks, yet deployed agents often remain static, failing to adapt as user needs evolve. This creates a tension between the need for continuous service and the necessity of updating capabilities to match shifting task distributions. On platforms like OpenClaw, which handle diverse workloads across 20+ channels, existing methods either store raw trajectories without distilling knowledge, maintain static skill libraries, or require disruptive downtime for retraining. We present MetaClaw, a continual meta-learning framework that jointly evolves a base LLM policy and a library of reusable behavioral skills. MetaClaw employs two complementary mechanisms. Skill-driven fast adaptation analyzes failure trajectories via an LLM evolver to synthesize new skills, enabling immediate improvement with zero downtime. Opportunistic policy optimization performs gradient-based updates via cloud LoRA fine-tuning and Reinforcement Learning with a Process Reward Model (RL-PRM). This is triggered during user-inactive windows by the Opportunistic Meta-Learning Scheduler (OMLS), which monitors system inactivity and calendar data. These mechanisms are mutually reinforcing: a refined policy generates better trajectories for skill synthesis, while richer skills provide higher-quality data for policy optimization. To prevent data contamination, a versioning mechanism separates support and query data. Built on a proxy-based architecture, MetaClaw scales to production-size LLMs without local GPUs. Experiments on MetaClaw-Bench and AutoResearchClaw show that skill-driven adaptation improves accuracy by up to 32% relative. The full pipeline advances Kimi-K2.5 accuracy from 21.4% to 40.6% and increases composite robustness by 18.3%. Code is available at https://github.com/aiming-lab/MetaClaw.

cs.LG↗

arXiv2026

Can RL Improve Generalization of LLM Agents? An Empirical Study

Reinforcement fine-tuning (RFT) has shown promise for training LLM agents to perform multi-turn decision-making based on environment feedback. However, most existing evaluations remain largely in-domain: training and testing are conducted in the same environment or even on the same tasks. In real-world deployment, agents may operate in unseen environments with different background knowledge, observation spaces, and action interfaces. To characterize the generalization profile of RFT under such shifts, we conduct a systematic study along three axes: (1) within-environment generalization across task difficulty, (2) cross-environment transfer to unseen environments, and (3) sequential multi-environment training to quantify transfer and forgetting. Our results show that RFT generalizes well across task difficulty within an environment, but exhibits weaker transfer to unseen environments, which correlates with shifts in both semantic priors and observation/action interfaces. In contrast, sequential training yields promising downstream gains with minimal upstream forgetting, and mixture training across environments improves the overall balance. We further provide detailed analyses and deeper insights, and hope our work helps the community develop and deploy generalizable LLM agents.

cs.AI↗

arXiv2026

Learning with Errors over Group Rings Constructed by Semi-direct Product

The Learning with Errors (\LWE) problem has been widely utilized as a foundation for numerous cryptographic tools over the years. In this study, we focus on an algebraic variant of the \LWE problem called \emph{Group ring} \LWE ($\GRLWE$). We select group rings (or their direct summands) that underlie specific families of finite groups constructed by taking the semi-direct product of two cyclic groups. Unlike the Ring-\LWE problem described in \cite{lyubashevsky2010ideal}, the multiplication operation in the group rings considered here is non-commutative. As an extension of Ring-$\LWE$, it maintains computational hardness and can be potentially applied in many cryptographic scenarios. In this paper, we present two polynomial-time quantum reductions. Firstly, we provide a quantum reduction from the worst-case shortest independent vectors problem (\SIVP) in ideal lattices with polynomial approximate factor to the search version of $\GRLWE$. This reduction requires that the underlying group ring possesses certain mild properties; Secondly, we present another quantum reduction for two types of group rings, where the worst-case \SIVP problem is directly reduced to the (average-case) decision $\GRLWE$ problem. The pseudorandomness of $\GRLWE$ samples guaranteed by this reduction can be consequently leveraged to construct semantically secure public-key cryptosystems.

cs.CR↗

arXiv2026

Speeding Up the Learning of 3D Gaussians with Much Shorter Gaussian Lists

3D Gaussian splatting (3DGS) has become a vital tool for learning a radiance field from multiple posed images. Although 3DGS shows great advantages over NeRF in terms of rendering quality and efficiency, it remains a research challenge to further improve the efficiency of learning 3D Gaussians. To overcome this challenge, we propose novel training strategies and losses to shorten each Gaussian list used to render a pixel, which speeds up the splatting by involving fewer Gaussians along a ray. Specifically, we shrink the size of each Gaussian by resetting their scales regularly, encouraging smaller Gaussians to cover fewer nearby pixels, which shortens the Gaussian lists of pixels. Additionally, we introduce an entropy constraint on the alpha blending procedure to sharpen the weight distribution of Gaussians along each ray, which drives dominant weights larger while making minor weights smaller. As a result, each Gaussian becomes more focused on the pixels where it is dominant, which reduces its impact on nearby pixels, leading to even shorter Gaussian lists. Eventually, we integrate our method into a rendering resolution scheduler which further improves efficiency through progressive resolution increase. We evaluate our method by comparing it with state-of-the-art methods on widely used benchmarks. Our results show significant advantages over others in efficiency without sacrificing rendering quality.

cs.CV↗