Search arXiv⌕ Search

arXiv subjects

Jiajun Liu

Publications and source records attributed to Jiajun Liu.

At least 19 recordsLinked to original sources
arXiv2026

Qwen-Planner-Agent: A Closed-Loop AI-for-AI Framework for Real-World Mobile Planner Agents

The rapid progression of large language models is extending AI from passive content generation into the active workflows of engineering and scientific discovery. This shift raises a compelling question: can AI be both the object of development and an active participant in building next-generation AI systems? We explore this question by building Qwen-Planner-Agent within a closed-loop AI-for-AI framework for scalable development and iterative improvement. Mobile planning offers a demanding test of this approach: complex, long-horizon tasks challenge agent reliability, while costly real-device interaction limits development scalability. The framework connects data production, model training, and deployment through a shared action-feedback-verification contract. (i) AI for Data builds a human-gated agentic data flywheel in which specialized agents construct tasks, collect interaction trajectories, curate and balance training data, and use training feedback to guide subsequent data generation. (ii) AI for Training combines a supervised planning cold start with hybrid-environment online agentic reinforcement learning, where we introduce Competence-Aware Reward-and-Advantage Engineering (CARE) to reduce reasoning and tool-use costs while preserving task performance. (iii) AI drives model--harness co-evolution through an execution-evidence-driven loop that orchestrates memory, skills, and tools at runtime and feeds structured action feedback and preserved failure traces back into coordinated model and harness adaptation. Qwen-Planner-Agent achieves the best overall performance among all evaluated models and systems on MobilePA-Bench, improving over its base model across tool use, memory, skills, and sub-agent coordination. Further evaluations of our model show improvements across non-mobile agentic benchmarks while largely preserving general capabilities.

cs.AI↗
arXiv2026

Embodied Agents Take Control: Minimal-Interface Zero-Shot Agents Rival Industrial-Scale Policies in Vision-and-Language Navigation

Autonomous embodied agents must sustain a long decision-making loop that involves perceiving, acting, verifying, and self-correcting over many steps. Current systems sustain this loop through task-specific workflows or embodied policies. However, these fixed workflows and policies offer limited flexibility across environments and often lack effective recovery strategies when execution goes wrong. We find that a general-purpose agent can instead sustain the loop on its own. We term this organization agentic embodied control: the reasoning model directly steers every action, keeping reasoning and control aligned. Using zero-shot navigation as a controlled testbed, we equip three coding-agent harnesses with only a monocular RGB camera and discrete actions. At default effort, replicated opus-5 runs average $70.7\pm3.5$% success, while fable-5 reaches 78% at maximum effort. When a trained waypoint tool is offered alongside primitives, the hybrid fable-5 agent reaches $76.7\pm0.6$% at default effort, using half the environment steps and under a quarter of the wall time. Across the ablations, model choice dominates performance variation. Observed harness differences are modest, and forced waypoints help weaker models but can hinder stronger ones. Although longer horizons, latency, and context growth remain barriers to sustained autonomy, these results show that a general-purpose model can already achieve competitive embodied control without a navigation policy.

cs.RO↗
arXiv2026

Dex-X: Learning Visual-Tactile Dexterous Manipulation From Human Videos with Simulated Interaction

Human videos are an abundant source of dexterous manipulation behaviors, but they lack tactile information that is crucial for contact-rich interaction. This raises a fundamental question: can robots learn deployable visual-tactile dexterous manipulation policies from human video demonstrations without robot-side data collection? We present DEX-X, a framework for learning visual-tactile dexterous manipulation from human videos through simulation. Our key insight is that simulation can serve as a tactile completion engine. Given monocular human demonstrations, DEX-X reconstructs hand-object interactions in simulation, where physically grounded contact dynamics provide tactile supervision unavailable in the original videos. Leveraging this recovered tactile information, we train visual-tactile dexterous manipulation policies and distill them into deployable policies operating on point-cloud observations and tactile sensing. We demonstrate zero-shot sim-to-real transfer on a dexterous hand-arm platform across diverse grasping and contact-rich tool-use tasks. The teacher policy achieves 65.9% average success across six task categories in simulation, while the distilled visual-tactile policy achieves 93% success on real-world cube picking and 53% on the challenging table-cleaning task. Zero-shot generalization to unseen object geometries is also observed on object-picking tasks. Our results suggest that simulated interaction is a key bridge between human videos and deployable dexterous manipulation policies, providing the missing physical supervision needed for scalable robot skill learning from Internet-scale human video data.

cs.RO↗
arXiv2026

MobilePA-Bench: Benchmarking Mobile Planner Agents on Complex Real-World Tasks

As on-device LLM agents evolve into personal copilots, the mobile operating system has become a key testbed for this paradigm, making rigorous capability evaluation essential. Yet existing benchmarks fall into two camps, each with a critical blind spot: GUI-centric benchmarks test surface-level screen manipulation while overlooking background tool use and long-horizon planning, whereas static function-calling benchmarks rely on offline API matching that is detached from real runtime constraints. To close this gap, we present \textbf{MobilePA-Bench}, an interactive, stateful, and tool-centric benchmark for evaluating the tool-calling and planning abilities of mobile planning agents. MobilePA-Bench runs on an executable sandbox that maintains live application databases and returns structured feedback, spanning $13$ functional domains and $212$ realistic mobile tools. Beyond basic tool use, it evaluates a central planning agent along three advanced dimensions: \emph{(1)~Sub-agent Collaboration}---decomposing a complex task and delegating specialized work to capable sub-agents; \emph{(2)~Memory Usage}---recalling stored memories, user profiles, and past preferences to resolve implicit requests; and \emph{(3)~Skill Usage}---invoking pre-packaged composite skills instead of planning every step from scratch. Extensive experiments show that current frontier LLMs remain unreliable in mobile settings: performance drops sharply under strict tool ordering, permission limits, and unexpected runtime errors. By pairing an interactive function-calling sandbox with evidence-based verification, MobilePA-Bench serves as both a practical diagnostic benchmark and an interactive foundation for agentic reinforcement learning---accelerating the development of dependable mobile agents.

cs.AI↗
arXiv2026

SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD

Full-parameter post-training of trillion-parameter-scale MoE models introduces substantial system-level challenges for large-scale distributed training, including severe memory pressure, non-overlapped communication overhead, and inefficient kernel execution. While most large-scale LLM training systems are built around GPU-based clusters, this report presents an end-to-end optimization practice on the Ascend NPU SuperPOD. Using the DeepSeek-V4 model family as the target workload, we develop a hierarchical optimization framework spanning model-level parallelism, computation-communication orchestration, and low-level kernel execution. The resulting system achieves 34.22% Model FLOPs Utilization (MFU) with a 2.93x improvement over the open-source baseline recipe while maintaining training stability. Building on this optimized infrastructure, we further establish a CPT and SFT workflow for complex Operations Research (OR) tasks. We refer to the integrated framework as SLAI T-Rex. Using DeepSeek-V4-Flash, we develop OR-oriented CPT and SFT data pipelines that combine collected domain resources with solver-verified synthetic optimization documents. The resulting dataset contains 10K high-quality SFT samples spanning four task categories and three problem representations. The specialized model achieves the highest average zero-shot Pass@1 score among the evaluated models, reaching 71.81% and outperforming GPT-5.4-Mini and the base DeepSeek-V4-Flash model by 3.98 and 11.27 percentage points, respectively. Overall, this work demonstrates a full-stack pathway from efficient trillion-parameter model post-training on Ascend infra to domain-specialized Flash models for solver-grounded mathematical modeling, advancing frontier-model systems for complex reasoning.

cs.CL↗
arXiv2026

Time-Frequency Consistency Learning for Robust Speech Deepfake Detection

Recently, speech deepfake detection (SDD) has achieved significant progress. However, its robustness evaluation remains largely confined to controlled additive noise scenarios, lacking systematic investigation of the complex distortions introduced by acoustic front-end (AFE) processing pipelines in real-world deployments. In this work, we simulate a unified AFE pipeline comprising acoustic echo cancellation, noise suppression, automatic gain control, and voice activity detection (VAD), and conduct a comprehensive evaluation of current state-of-the-art models. The results show that the nonlinear and time-frequency coupled distortions introduced by AFE significantly degrade detection performance. To address this issue, we propose a Time-Frequency Consistency Learning (TFCL) framework, which aims to learn invariant spoofing representations that remain stable before and after AFE processing. We observe that AFE not only introduces temporal misalignment (e.g., segment-level shifts caused by VAD), but also weakens or distorts critical frequency-domain cues. To this end, TFCL employs an attention-driven soft alignment mechanism to capture cross-temporal dependencies, along with frequency-domain structural consistency constraints to enforce feature invariance. As a result, the model is able to maintain stable representations under both temporal perturbations and spectral distortions. Extensive experimental results demonstrate that the proposed method effectively mitigates the performance degradation caused by AFE processing, significantly improving the robustness of SDD in real-world scenarios. The code is available at https://github.com/JunXue-tech/TFCL.

cs.SD↗
arXiv2026

The conditional higher moment risk measure: second-order asymptotics with FGM contagion

This paper investigates second-order asymptotic expansions for the conditional higher moment (CoHM) coherent risk measure under a Farlie-Gumbel-Morgenstern (FGM) dependence structure, capturing a weak contagion between a primary loss risk and a reference risk. Assuming that the primary risk belongs to the Fréchet, Weibull, or Gumbel maximum domain of attraction, we systematically derive second-order asymptotic expansions using extreme value theory and second-order regular variation theory. Compared with existing first-order results, our refined approximations capture higher-order tail behavior and dependence effects more accurately. Numerical simulations confirm that the second-order asymptotics substantially reduce approximation errors, especially at extreme confidence levels. Empirical applications to insurance claim data further illustrate the practical superiority of the second-order approach.

q-fin.RM↗
arXiv2026

A Conformal Selection Framework for Individual Treatment Beneficiaries with Auxiliary External Data

Identifying patients who are likely to benefit from a treatment is central to precision medicine and can guide follow-up trials, enrichment designs, and individualized decisions. Although randomized controlled trials (RCTs) provide evidence on efficacy, they are usually powered to estimate average treatment effects rather than patient-level benefit. Meanwhile, artificial intelligence and machine learning methods offer flexible tools for estimating heterogeneous treatment effects, especially when augmented by real-world data (RWD). However, in practice, these estimated effects are often translated into decisions through simple ranking or thresholding rules, which can ignore uncertainty and multiplicity when many patients are evaluated simultaneously. Motivated by this, we propose a model-agnostic conformal inference framework for uncertainty-aware beneficiary selection. The framework reformulates CATE-based treatment-benefit selection as a multiple-testing problem. For each candidate, we test whether the conditional treatment benefit exceeds a clinically meaningful threshold and construct a conformal p-value using RCT-based calibration. These p-values are then adjusted by the Benjamini-Hochberg procedure to control the false discovery rate (FDR) among selected beneficiaries. To improve efficiency when RCT sample sizes are limited, external data, such as RWD, can be used to train flexible treatment effect models, while conformal calibration remains anchored in the RCT data. It can be paired with conventional machine learning algorithms and emerging tabular foundation models. Simulations show that the framework maintains FDR control, with power depending on the base learner and external-data comparability. A case study in early-stage non-small-cell lung cancer illustrates how the method identifies candidate profiles with evidence of benefit from limited resection to reduce overtreatment.

stat.ME↗
arXiv2026

An Estimand-Focused Approach for AUC Generalization and Cross-Study Benchmarking

The area under the ROC curve (AUC) is the standard measure of a biomarker's discriminatory accuracy; however, AUC is rarely treated as a population-specific estimand. When validation cohorts differ from the intended target population in case mix, Naïve AUC estimates can mislead both generalization and cross-study comparison. We develop an estimand-focused framework that anchors biomarker AUC inference to a prespecified target population, aligning with the ICH E9(R1) estimand perspective adapted to discrimination rather than treatment effect. The framework supports two scientific goals: generalizing a study-specific AUC to a clinically relevant target population, and benchmarking AUCs across studies on a common population footing. Methodologically, we extend calibration weighting to the U-statistic formulation of AUC, allowing valid estimation even when the target population is characterized only by summary-level covariate information. This setting is common in biomarker validation, where individual-level target data are often unavailable and existing transportability methods may not be applicable. When patient-level real-world data are accessible, the proposed augmented variants provide double robustness and improved efficiency. We establish asymptotic properties and study their performances through comprehensive simulations. Furthermore, we demonstrate the proposed framework on the POWER trials, evaluating baseline stair-climb power (SCP) as a prognostic marker for 6-month survival in advanced non-small-cell lung cancer (NSCLC). Unlike prior work on transporting model-based predictive accuracy, our framework targets the biomarker-level estimand directly and addresses cross-study comparability - an issue not resolved by current methods.

stat.ME↗
arXiv2026

Sparsified Kolmogorov-Arnold Networks for Interpretable Quantum State Tomography

Machine-learning approaches to quantum state tomography can achieve high reconstruction fidelity, but the physical structure used by the trained model often remains implicit. Here we ask whether a sparsified Kolmogorov-Arnold Network (KAN) can be used not only as a regressor, but also as an inspectable reconstruction rule whose internal organization can be checked against known Pauli structure. We study a controlled three-qubit GHZ-family benchmark in which all 63 non-identity Pauli expectation values are used to reconstruct three GHZ-subspace variables: the population imbalance $z$, the real off-diagonal component $c$, and the imaginary off-diagonal component $s$. Under finite-shot sampling and depolarizing noise, external ablation identifies the extended 12-channel GHZ-relevant Pauli set from the 63 measurements, with exact top-12 recovery across the tested shot counts and depolarizing-noise strengths. These support patterns remain stable across multi-seed random-initialization and noise-level analyses, and collapse under random-label controls. The dominant pruned input-hidden-output pathways organize Z-type population observables and X/Y off-diagonal observables in a pattern consistent with the analytic GHZ Pauli grouping, and sparse formula recovery recovers the canonical signed Pauli relations. The contribution of the KAN is therefore pathway-level structural interpretability within a neural reconstruction model, rather than superior sparse regression. Together with negative controls, these probes provide a consistency chain for auditing learned reconstruction rules against known physical structure.

quant-ph↗
arXiv2026

Exploring the Scale and Diversity of Speech Anti-spoofing Datasets: Experiments and Analysis

The scale of speech anti-spoofing datasets has grown exponentially over the past decade, driven by the assumption that larger data leads to better performance. However, it remains unclear whether indiscriminate scaling commensurately improves model generalization. This study challenges the "scale-first" paradigm by decoupling the impacts of training data scale versus diversity. Through experiments on representative datasets, we report two key findings: (1) Larger is not always better. Expanding data scale excessively under fixed generation methods yields negligible returns and may even degrade cross-domain generalization due to overfitting.(2) Diversity outweighs scale. A smaller composite training set featuring diverse attacks significantly outperforms larger-scale datasets with limited diversity in cross-dataset evaluations. We conclude that future dataset construction should prioritize the diversity of generation methods over scale to effectively enhance model generalization.

cs.SD↗
arXiv2026

Ask When It Pays: Cost-Aware Open-Ended Interaction for Instance Goal Navigation

Instance Goal Navigation (IGN) requires an embodied agent to find a specific object instance among distractors from an under-specified natural-language description. Such ambiguity often cannot be resolved from perception and language alone, making interaction with an oracle a natural mechanism for disambiguation. Prior interactive methods allow oracle queries but treat lightweight clarification and route-level guidance alike, letting agents boost success rate through repeated high-information questions rather than by resolving the underlying ambiguity efficiently. We recast interactive IGN as a cost-sensitive uncertainty-reduction problem, where the agent should ask the question whose answer provides the largest reduction in navigation uncertainty relative to its penalty. To this end, we apply an information-gain analysis on existing navigation corpora to identify which cues reduce navigation uncertainty, yielding a compact set of question types and data-derived weights. However, existing interactive navigation benchmarks do not model the cost of different question types or evaluate how efficiently agents use interaction, making them unsuitable for studying cost-sensitive interaction. Based on this taxonomy, we construct a benchmark for diagnosing interaction behavior and efficiency, together with a Weighted Success Rate metric that penalizes each query by its derived cost. We further propose a zero-shot MLLM navigator that selectively queries at each decision step only when the expected uncertainty reduction justifies the interaction cost.

cs.CV↗
arXiv2026

TTT-VLA: Test-Time Latent Prompt Optimization for Vision-Language-Action Models

Vision-Language-Action (VLA) models trained on large-scale data have made remarkable progress, but they remain vulnerable to distribution shifts at deployment time. Recent VLA models suggest that prompts can serve as an efficient interface for steering policy behavior, but existing prompt-based steering typically relies on external guidance. This raises a natural question: can test-time training (TTT) for VLA be achieved by optimizing a prompt, so that the steering interface itself can be learned and adapted from interaction? We address this question with TTT-VLA, a test-time training framework based on Latent Prompt Optimization (LPO). During training, the latent prompt is learned with an additional proxy task, providing an extra learned conditioning signal for policy learning. At test time, TTT is performed by collecting interaction data from the current environment and optimizing only the latent prompt on those data using the proxy task's self-supervised signal, without modifying the policy itself. Experiments on SimplerEnv demonstrate that the proposed method consistently improves task success rates in both single- and multi-embodiment settings. Further analysis shows that the gains arise primarily from correcting a small number of critical decisions rather than globally altering policy behavior. These results suggest that LPO provides an effective and practical pathway for deployment-time improvement of foundation manipulation policies.

cs.RO↗
arXiv2026

ReCoVR: Closing the Loop in Interactive Composed Video Retrieval

Composed video retrieval (CoVR) searches for target videos using a reference video and a modification text, but existing methods are restricted to a single interaction round and cannot support the progressive nature of real-world visual search. To bridge this gap, we first formalize interactive composed video retrieval, a multi-turn extension of CoVR, where users progressively refine their search intent through natural-language feedback across turns. Adapting existing interactive retrieval methods to this setting reveals two structural weaknesses: reliance on a single retrieval channel and an open-loop retrieval design that consumes user feedback but does not diagnose whether its own retrieval trajectory is drifting or stagnating. To address these limitations, we propose ReCoVR (Reflexive Composed Video Retrieval), a dual-pathway architecture built on reflexive perception, where the system treats its retrieval history as diagnostic evidence alongside user feedback. Specifically, an Intent Pathway routes heterogeneous feedback to complementary retrieval channels, while a Reflection Pathway performs trajectory-level reflection to monitor result evolution and correct retrieval errors across turns. Experiments on multiple benchmarks show that ReCoVR consistently outperforms interactive baselines, notably achieving 74.30% R@1 after just one interactive round on the WebVid-CoVR-Test dataset.

cs.IR↗
arXiv2026

Knowledge Poisoning Attacks on Medical Multi-Modal Retrieval-Augmented Generation

Retrieval-augmented generation (RAG) is a widely adopted paradigm for enhancing LLMs in medical applications by incorporating expert multimodal knowledge during generation. However, the underlying retrieval databases may naturally contain, or be intentionally injected with, adversarial knowledge, which can perturb model outputs and undermine system reliability. To investigate this risk, prior studies have explored knowledge poisoning attacks in medical RAG systems. Nevertheless, most of them rely on the strong assumption that adversaries possess prior knowledge of user queries, which is unrealistic in deployments and substantially limits their practical applicability. In this paper, we propose M\textsuperscript{3}Att, a knowledge-poisoning framework designed for medical multimodal RAG systems, assuming only limited distribution knowledge of the underlying database. Our core idea is to inject covert misinformation into textual data while using paired visual data as a query-agnostic trigger to promote retrieval. We first propose a unified framework that introduces imperceptible perturbations to visual inputs to manipulate retrieval probabilities. Besides, due to the prior medical knowledge in LLMs, naively poisoned medical content with explicit factual errors can be corrected during generation. Thus, we leverage the inherent ambiguity of medical diagnosis and design a covert misinformation injection strategy that degrades diagnostic accuracy while evading model self-correction. Experiments on five LLMs and datasets demonstrate that M\textsuperscript{3}Att consistently produces clinically plausible yet incorrect generations. Codes: https://github.com/ypr17/M3Att.

cs.CR↗
arXiv2026

Scalable Quantum Error Mitigation with Physically Informed Graph Neural Networks

Quantum error mitigation (QEM) provides a practical route for estimating reliable observables on noisy intermediate-scale quantum (NISQ) devices. Traditional QEM strategies, including zero-noise extrapolation (ZNE) and Clifford data regression (CDR), rely on noise scaling or global regression, and their performance is constrained by the exponential growth of the system degrees of freedom. We construct a graph-enhanced mitigation (GEM) framework, which incorporates physical information into the model representation. In this work, quantum circuits are encoded as attributed graphs. Hardware-level physical information is mapped to node and edge features: local noise parameters such as calibration parameters $T_1$, $T_2$, and readout errors are encoded at nodes, while coupling-related information such as two-qubit gate errors is encoded as edge features. Graph neural networks are used to model how errors propagate along the physical coupling structure and build up into non-local correlations. This allows the model to capture local interactions and part of the resulting non-local correlations across qubits. A dual-branch affine correction is applied to maintain consistency with physical constraints. Experiments on 10-qubit and 16-qubit random circuits executed on superconducting quantum processors show that GEM provides a level of accuracy comparable to CDR at small scales, while yielding lower mean absolute error and improved stability in zero-shot transfer to larger systems. Results of the traditional QEM strategy indicate that global regression methods remain effective in low-dimensional settings but become less reliable as system degrees of freedom grow. In contrast, GEM makes use of local physical structures to show better scalability and generalization, while preserving the overall error propagation patterns. This work provides a practical scalable approach to QEM for NISQ devices.

quant-ph↗
arXiv2026

Projection of purification performance for the RELICS experiment

The RELICS (REactor neutrino LIquid xenon Coherent elastic Scattering) experiment employs a dual-phase liquid xenon time projection chamber to search for Coherent Elastic Neutrino-Nucleus Scattering (CE$ν$NS) induced by reactor neutrinos. To detect these sub-keV nuclear recoils and minimize signal attenuation, it is critical to maintain a sufficiently low impurity concentration in the detector. This work presents a comprehensive purity evolution model developed to describe impurity migration inside the detector. Utilizing measured material outgassing rates as input parameters, the model incorporates non-uniform transport mechanisms of the impurities, including circulation, vaporization, and condensation. The model is validated using data from a dedicated prototype detector. Based on this validated model, projections for the purification performance of the upcoming RELICS-10 and RELICS-50 detectors are provided.

physics.ins-det↗
arXiv2026

Development of a dual-phase xenon time projection chamber prototype for the RELICS experiment

The RELICS (REactor neutrino LIquid xenon Coherent elastic Scattering) experiment aims to detect coherent elastic neutrino-nucleus scattering from reactor antineutrinos using a dual-phase xenon time projection chamber. To validate the detector concept and ensure technical reliability for the full-scale experiment, a dedicated prototype was designed, constructed, and operated. This work presents an overview of the design, construction, and operational performance of the prototype, with emphasis on its major subsystems, including the TPC, cryogenic and xenon purification systems, slow control, and data acquisition. During operation, the detector demonstrated the capability to achieve a sub-keV energy threshold required for the RELICS physics program, as reflected by a measured single electron gain of 34.30~$\pm$~0.01~(stat.)~PE/e$^-$ and the successful detection of 0.27~keV L-shell decay events from $^{37}$Ar. In addition, essential data analysis techniques and simulation frameworks were developed and validated, establishing the methodological foundation for future RELICS operations. The successful construction and operation of this prototype confirm the feasibility of the core technologies and provide a crucial experimental basis for the final RELICS detector.

physics.ins-det↗