arXiv subjects

Tao Wang

Publications and source records attributed to Tao Wang.

At least 73 records · Page 4Linked to original sources

arXiv2026

Goal-oriented Navigation Instruction Generation with Tour Video Priors

Navigation Instruction Generation (NIG) aims to produce step-by-step natural language instructions for navigation guidance. Existing studies primarily treat NIG as an auxiliary task for vision-andlanguage navigation (VLN), focusing on data augmentation or multi-task learning. However, generating navigation instructions from compact environmental priors requires meticulous spatial reasoning, especially when the target route does not simply follow the demonstrated tour, and remains challenging for current multimodal models. In this work, we introduce VideoNIG, a goal-oriented video-grounded NIG task that generates navigation instructions from ego-centric tour videos, an initial observation, and a textual or visual goal, without relying on intermediate representations such as graphs and maps. We instantiate VideoNIG in a controlled simulator benchmark with 60K tour videos across continuous indoor environments and 37K multimodal prompts with progressive difficulty levels. We further introduce a diagnostic evaluation protocol that combines text similarity, choice-based spatial consistency tests, and downstream navigation execution. To address this task, we propose a two-stage Curriculum Learning framework that decomposes the learning into foundational motion perception and long-horizon navigation reasoning. Specifically, we first employ Action Warmup for spatial action-view alignment, followed by Complexity Progression using trajectories with increasing exploratory difficulty. Extensive experiments show that existing MLLMs struggle with VideoNIG, while our approach significantly improves instruction quality across complementary diagnostic metrics. Finally, integrating VideoNIG-generated instructions with a VLN agent demonstrates the executability of this task formulation for end-to-end navigation.

cs.CV↗

arXiv2026

Ruffled Feathers: Merger-driven galaxy size growth and structural transformation in EAGLE

Galaxy mergers drive both the size growth and the transformation from discs to spheroids, yet the prescriptions used to model these processes in semi-analytic frameworks have not been tested against the realistic merger population in cosmological hydrodynamical simulations. Using $\approx 4{,}500$ mergers identified in the EAGLE simulation, we test an energy-conservation estimator for post-merger galaxy sizes and quantify merger-driven morphological transformation. The predicted remnant half-stellar-mass radius matches the simulated descendant size with a scatter of $\approx 0.12$-$0.15$ dex and no significant systematic dependence on progenitor properties, while a commonly used dissipation correction applied to gas-rich mergers under-predicts the post-merger size by up to $\approx 0.4$ dex in a cosmological context and increases the overall scatter. The per-merger size growth increases monotonically with the stellar mass ratio of the merging pair, from $\lesssim 0.03$ dex for minor mergers to $\approx 0.10$ dex for equal-mass mergers. From the energy-conservation estimator, we analytically derive the size growth efficiency per unit accreted stellar mass, $η\equiv \mathrm{d}\log_{10} r_{\star}/\mathrm{d}\log_{10} M_{\star}$, and show that $η$ reaches $\approx 2$ only in the idealised limit of collisionless minor mergers with zero orbital energy; as $η$ is highly sensitive to the orbital energy at the time of merging, the minor merger channel cannot be established as the driver of the rapid size growth of massive galaxies without better constraints on this quantity. Beyond the size growth, mergers systematically reduce rotational support and increase triaxiality in proportion to mass ratio, but even the most nearly equal-mass mergers do not always fully destroy the disc, in tension with the complete disc destruction assumed in several semi-analytic models.

astro-ph.GA↗

arXiv2026

Benchmarking and Enhancing LLMs for Rule-Intensive Review of National Standard Documents

Large language models (LLMs) increasingly support complex professional tasks, yet their capabilities in rule-intensive document review remain insufficiently evaluated. National standard documents, such as China GB/T standards, offer a representative testbed: they are lengthy, highly structured, and governed by explicit rules for scope, terminology, normative wording, and cross-section consistency. Existing benchmarks focus on domain knowledge and question answering, largely overlooking intrinsic quality review for professional documents. Such reviews rely heavily on human experts, making them costly and difficult to scale. To bridge this gap, we introduce GB/T-Bench, the first benchmark for the structured review of national standard documents. Its GB/T Review Taxonomy is a hierarchical schema covering document structure, scope alignment, normative modality, terminology consistency, and normative references, with 25 diagnosable error types. A controllable counterexample generation mechanism combines deterministic rules and constrained LLM rewriting to process 488 documents into 7,306 traceable review error instances for evaluation. We also develop a diagnosis-oriented evaluation protocol requiring exact matches on error location, review dimension, and error type, plus document-level coverage metrics. We further propose GB/T-Reviewer, a multi-agent framework that converts review knowledge into specialized skills and coordinates global inspection, targeted diagnosis, rule scanning, and result verification. Experiments with 14 mainstream LLMs reveal a substantial human-LLM gap: the strongest model achieves only 0.3280 CMCS versus 0.6640 for experts. GB/T-Reviewer raises the best CMCS to 0.5094, showing the value of structured skill coordination for rule-intensive document review. This work paves the way for trustworthy AI in standardization and other high-stakes document domains.

cs.CL↗

arXiv2026

Answer Presence Drives RAG Rewriting Gains

Retrieval-augmented QA pipelines often route retrieved passages through an LLM \emph{rewriter} before a smaller reader, lifting F1 by tens of points on multi-hop benchmarks; this gain is typically credited to improved evidence quality. We ask whether that lift is causally driven by the gold answer string appearing in the rewritten context rather than by curation per se, using a controlled intervention audit. For each rewritten context we re-run the reader after one of four controlled edits to the compile output: removing the gold answer span, replacing a length-matched random non-answer span (placebo), or injecting the gold into rewrites where it was absent (at the prefix or at a midpoint sentence boundary). Across twelve completed (cell, baseline) intervention runs spanning three reader families (Qwen2.5-7B, Qwen3.5-35B, GLM-4.7), two datasets (HotpotQA, 2WikiMultihopQA), and three compiler arrangements (MA-only, MB-only, MA$+$verify), removing the gold answer drops reader F1 by $28$ to $64$ points beyond the length-matched placebo on paired \texttt{answer-in-compile} strata, and prepending the gold into rewrites that lacked it raises F1 by $+0.7$ to $+9.7$ points in $10$ of $12$ (cell, baseline) combinations. A companion five-sentinel audit shows the conventional single-\texttt{[MASK]} probe is itself sentinel-fragile: on 2Wiki it reports a $+4.12$~F1 ``non-leakage residual'' that flips to $-3.33$ to $-7.81$~F1 under four alternative sentinels and fails an equivalence test for three of those four ($1/4$~pass). We do not propose a new rewriter or mitigation; we release the intervention runner and the sentinel panel so that other rewriter-gain claims can be tested against the same standard.

cs.AI↗

arXiv2026

SPECTRA: On-Device Cognitive Perturbation and Trajectory Analysis for Autonomous Edge-Cloud GUI Grounding

The effectiveness of edge-cloud collaboration for GUI grounding depends on autonomous requesting, where the edge agent selectively offloads complex tasks to the powerful cloud. However, in visually dense scenarios, lightweight edge agents often exhibit overconfident hallucinations, leading to a misalignment between confidence and accuracy that hinders reliable autonomous requesting. To address this, we leverage the observation that an agent's cognitive instability leads to significant latent drift under minute perturbations due to steep decision boundaries. We propose SPECTRA, a lightweight autonomous request framework for edge-cloud GUI grounding, comprising (1) Saliency-Guided Targeted Perturbation and (2) Efficient Cognitive Trajectory Analysis. SPECTRA conducts a visual cognitive stress test by injecting masks into critical visual anchors and quantifies the topological divergence of the agent's high-dimensional cognitive trajectories during the prefill phase, avoiding inefficient output decoding. Experiments demonstrate that SPECTRA performs cloud request assessment without autoregressive decoding. Our GTA1-32B+InfiGUI-G1-3B and GTA1-32B+Holo1.5-3B maintain 93.44% and 95.60% of cloud-only performance with average request rates of 37.58% and 39.24%, respectively.

cs.HC↗

arXiv2026

CodeRescue: Budget-Calibrated Recovery Routing for Coding Agents

Coding agents increasingly operate in executable environments where a failed attempt produces actionable feedback rather than merely an incorrect answer. Existing cost-aware systems typically treat such failures as cascade decisions: try a cheap model first, then escalate hard cases to a stronger and more expensive model. In coding, however, execution feedback can also make further cheap-model recovery worthwhile, raising a budgeted deployment question: when should an agent spend more cheap compute, and when should it escalate? We formulate this post-failure decision as recovery routing over heterogeneous actions and train a supervised router from execution rollouts. To make the same router usable under changing budgets, we add a Conformal Risk Control (CRC) layer that selects a deployment-time cost penalty without retraining and provides marginal expected-cost control under exchangeability. Across held-out failures from five coding benchmarks, cheap recovery and escalation exhibit complementary success patterns. The calibrated frontier improves over fixed actions, prompt-only routers, and a binary cascade baseline; in the main GPT-5.4-nano/GPT-5.4 setting, one CRC-calibrated frontier point exceeds always-escalate solve rate while using 35% of its mean recovery cost. Code is available at https://github.com/Qijia-He/agent-budget-control.

cs.AI↗

arXiv2026

COMPOL: A Unified Neural Operator Framework for Scalable Multi-Physics Simulations

Multiphysics simulations play an essential role in accurately modeling complex interactions across diverse scientific and engineering domains Although neural operators especially the Fourier Neural Operator FNO have significantly improved computational efficiency they often fail to effectively capture intricate correlations inherent in coupled physical processes To address this limitation we introduce COMPOL a novel coupled multiphysics operator learning framework COMPOL extends conventional operator architectures by incorporating sophisticated recurrent and attentionbased aggregation mechanisms effectively modeling interdependencies among interacting physical processes within latent feature spaces Our approach is architectureagnostic and seamlessly integrates into various neural operator frameworks that involve latent space transformations Extensive experiments on diverse benchmarksincluding biological reactiondiffusion systems patternforming chemical reactions multiphase geological flows and thermohydromechanical processes demonstrate that COMPOL consistently achieves superior predictive accuracy compared to stateoftheart methods.

cs.LG↗

arXiv2026

Tracing cosmic star formation history through radio continuum spectral energy distribution and non-thermal emission

As a tracer of massive star formation unaffected by dust, the radio continuum emission provides a unique window into the formation of the first stars and galaxies in the Universe. Recent observations show that the integrated rest-frame mid-radio (~1-10 GHz) luminosity of galaxies serves as one of the most robust tracers of the star formation rate (SFR). These studies further demonstrate that the synchrotron spectral index and the shape of the radio spectral energy distribution (SED) evolves with redshift as a consequence of the cosmic evolution of star formation activity. These findings underscore the importance of deep multi-band radio continuum observations in calibrating the SFR of early galaxies and understanding the astrophysical processes governing their assembly and evolution over cosmic time. This chapter presents recent progress in radio SFR calibrations for star-forming galaxies (SFGs) and reviews radio-continuum studies of the cosmic star formation history (SFH). We highlight the transformative potential of SKA AA4, whose broad frequency coverage and high sensitivity will enable well-constrained radio SEDs for SFGs across a wide redshift range.

astro-ph.GA↗

arXiv2026

When Does Legacy Data Start to Help? Emergent Transfer in Cross-Configuration Robot Learning

Robotic hardware evolves over time, but demonstration data is often tied to a specific sensor and actuator configuration. This raises a practical and underexplored question: when does legacy data begin to benefit an upgraded robot? We study this question on a wheeled humanoid platform across two hardware generations, where both the camera and gripper are changed while the overall morphology remains fixed. Contrary to the common assumption that more cross-configuration data is always helpful, we observe a grokking-like transition: legacy data remains ineffective until the upgraded configuration acquires a minimum level of task competence, after which co-training gains rise sharply before diminishing near saturation. We hypothesize that this task-dependent transition is governed by a transfer threshold and characterize the resulting three-phase pattern. Across real-robot manipulation tasks, we observe all three phases: no measurable benefit at low competence ($10.0\% \rightarrow 10.0\%$), a sharp gain after crossing the threshold ($23.3\% \rightarrow 86.7\%$ on flower insertion), and diminishing returns at high competence ($85.0\% \rightarrow 93.3\%$ on pen insertion). We provide a theoretical account based on gradient alignment and residual policy uncertainty, and derive a phase-aware rule for deciding when to collect more new-hardware data and when to reuse legacy demonstrations. We further validate this three-phase pattern on a mobile dual-arm watering task, with results consistent with our predictions.

cs.RO↗

arXiv2026

Eliminating Propagation Delay: Attention-Based Spatial-Temporal Fusion Graph Convolution Network for Traffic Flow Prediction

Predicting traffic flow is crucial to optimizing transportation systems and improving urban mobility. Many graph convolution-based models have been proposed to extract spatial-temporal features and predict traffic flow. However, most focus on spatial-temporal and semantic correlation in topological relationships. There are two primary problems to address. Firstly, the convolutional structure in the model focuses on utilizing static spatial dependencies and spatial-temporal relationships in topological structures, while neglecting the different information propagation delays between adjacent nodes in the convolution. Secondly, these methods often stack a large number of complex structures, resulting in a substantial increase in computational time during the model training phase, thereby disregarding the model's requirements for timeliness. In this paper, we propose a novel network called the Attention-Based Spatial-Temporal Fusion Graph Convolution Network (A-STFGCN). We design a spatial-temporal fusion block to extract the spatial-temporal feature correlations with propagation delay errors removed and to capture both long-term and short-term temporal characteristics of the data within a multi-head self-attention mechanism based on a mask matrix. Extensive experiments on five real-world datasets demonstrate that our method achieves the best overall performance while having good computation and data utilization efficiency compared with the eight baseline methods.

cs.LG↗

arXiv2026

Vertical Pinching Antenna Systems (V-PAS) Aided UAV Communications

To address the limitation that existing horizontal pinching antenna systems (PAS) are primarily designed for ground users and confined to two-dimensional (2D) coverage, this paper proposes deploying a pinching antenna (PA) along the facade of urban buildings to construct a vertical pinching antenna system (V-PAS) for ultra-low-altitude unmanned aerial vehicle (UAV) communications. The proposed architecture employs a dielectric waveguide continuously deployed along the full height of buildings, extending the coverage capability of PAS from the 2D plane to three-dimensional (3D) airspace, ensuring a stable line-of-sight (LoS) link. We define the concept of pinching multiplicative path loss (PMPL) to characterize the cascaded multiplicative attenuation of the waveguide and free-space path losses. It is found that PMPL is insensitive to vertical distance, rendering V-PAS highly adaptive to building heights. Furthermore, accurate and asymptotic closed-form expressions for outage probability and ergodic rate under lossy waveguide conditions are derived, respectively, and the symmetry and optimality of system performance with respect to the midpoint of the access point (AP) height are discovered and proved. The results show that V-PAS achieves performance advantages over the benchmark without PA in most ultra-low-altitude UAV communication scenarios. Only under extremely high transmission power and large UAV operational area may the outage probability of V-PAS with a lossy waveguide be inferior to that of the benchmark without PA, but the ergodic rate of V-PAS still maintains an advantage. By contrast, V-PAS with a lossless waveguide outperforms the benchmark without PA in UAV communications, representing the theoretical performance upper bound of V-PAS.

eess.SP↗

arXiv2026

Statistical Early Stopping for Reasoning Models

While LLMs have seen substantial improvement in reasoning capabilities, they also sometimes overthink, generating unnecessary reasoning steps, particularly under uncertainty, given ill-posed or ambiguous queries. We introduce statistically principled early stopping methods that monitor uncertainty signals during generation to mitigate this issue. Our first approach is parametric: it models inter-arrival times of uncertainty keywords as a renewal process and applies sequential testing for stopping. Our second approach is nonparametric and provides finite-sample guarantees on the probability of halting too early on well-posed queries. We conduct empirical evaluations on reasoning tasks across several domains and models. Our results indicate that uncertainty-aware early stopping can improve both efficiency and reliability in LLM reasoning, and we observe especially significant gains for math reasoning.

cs.AI↗

arXiv2026

Not Birds of a Feather: Personality-Based Partner Selection in LLM Agents

LLM-based agents increasingly operate in multi-agent ecosystems where a coordinating agent chooses which other agents to work with, and agents are increasingly given personalities through persona prompts. However, whether personality itself influences this endogenous partner choice has not been sufficiently examined: prior work on personality in multi-agent teams has typically fixed team composition exogenously. We present a controlled selection paradigm in which a host agent chooses among six candidate agents that differ only in their Big Five personality descriptions, with capability explicitly equalized (375 trials across five task categories). We find that selection is strongly and systematically personality-dependent. Neutral hosts matched personalities to task types, choosing the open candidate for creative work and the conscientious candidate for most other categories, while the extraverted, agreeable, and balanced candidates were almost never chosen, despite human evidence that agreeableness is among the most performance-relevant traits for teams. Hosts that were themselves assigned personalities selected self-similar partners below chance and chose partners farther from themselves in trait space than random choice would produce. These results suggest that hosts read personality descriptions as signals of task fit rather than as grounds for similarity-based attraction: selection follows task stereotypes and favors complements, the opposite of human homophily. Our findings have direct implications for bias auditing in agent marketplaces and orchestration frameworks.

cs.MA↗

arXiv2026

Joint parameter estimation and multidimensional reconciliation for continuous-variable quantum key distribution

Accurate quantum channel parameter estimation is essential for effective information reconciliation in continuous-variable quantum key distribution (CV-QKD). However, conventional maximum likelihood (ML) estimators rely on a large amount of disclosed data, leading to a significant loss in symbol efficiency. Moreover, the separation between the estimation and reconciliation phases can introduce error propagation. In this paper, we propose a novel joint message-passing scheme that unifies channel parameter estimation and information reconciliation within a Bayesian framework. By leveraging the expectation-maximization (EM) algorithm, the proposed method simultaneously estimates unknown parameters during decoding, eliminating the need for separate ML estimation. Furthermore, we introduce a hybrid multidimensional rotation scheme that removes the requirement for norm feedback, significantly reducing classical channel overhead. To the best of our knowledge, this is the first work to unify multidimensional reconciliation and channel parameter estimation in CV-QKD, providing a practical solution for high-efficiency reconciliation with minimal information disclosure.

quant-ph↗

arXiv2026

S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence

Real-world spatial intelligence requires reasoning over a continuous and evolving 3D world, yet existing VLMs and tool-augmented agents largely remain tied to static, stateless inference from isolated visual observations. We introduce \textbf{\textsc{S-Agent}}, a spatial tool-use agentic paradigm for understanding and reasoning over continuous multi-view images and videos. By formulating spatial reasoning as spatio-temporal evidence accumulation rather than isolated frame-level prediction, \textsc{S-Agent} reshapes spatial perception into scene-centric understanding beyond frame-centric recognition. Specifically, \textsc{S-Agent} casts the VLM as a semantic planner that decides what evidence is needed, while a hierarchy of spatial tools and experts grounds objects in 2D, lifts them into 3D geometric evidence, and aggregates this evidence into high-level spatial knowledge (\textit{e.g.}, counting, measurement, orientation, and relative position). Additionally, a temporal memory mechanism, including Scene Memory for maintaining the evolving scene state and Agent Memory for accumulating reasoning context, enables evidence integration across frames and reasoning steps. Comprehensive experiments on multi-view and video spatial reasoning benchmarks show that \textsc{S-Agent} consistently improves both open-source and closed-source VLMs in a training-free manner. Beyond inference-time augmentation, supervised fine-tuning (SFT) on \textsc{S-Agent}-generated spatial trajectories \textsc{S-300K} yields \textsc{S-Agent-8B}, a compact spatial agent that significantly surpasses similar-scale baselines (e.g., Qwen3-VL-8B) and performs comparably to advanced closed-source models (e.g., GPT-5.4 and Gemini 3).

cs.CV↗

arXiv2026

Upper bound for the first $p$-Steklov eigenvalue in $\mathbb{R}^n$

For $p\in(1,n]$ and any bounded convex domain $Ω\subset\mathbb{R}^n$, we prove the sharp inequality \[ Λ_p(Ω):=\frac{W_p(Ω)}{P(Ω)V(Ω)^{p/n}}\geqω_n^{-p/n}, \qquad W_p(Ω)=\int_{\partialΩ}|x|^p\ dS, \] with equality holding exactly at centered balls. Combining this with the isoperimetric inequality yields the explicit upper bound \[ σ_{1,p}(Ω)\leq \frac{A(n,p)}{r(Ω^*)^{p-1}}, \] where $Ω^*$ is a ball having the same perimeter as $Ω$, and $A(n,p)=1$ for $1<p\leq 2$, $A(n,p)=n^{p/2-1}$ for $2<p\leq n$. When $p=2$, the result recovers the higher-dimensional Weinstock inequality of Bucur et al. [J. Differential Geom. 2021]. We also obtain an explicit upper bound for the first Wentzell eigenvalue of the $p$-Laplacian on convex domains.

math.AP↗

arXiv2026

Spectral Attack on Continuous-Variable Quantum Key Distribution Systems

Continuous-variable quantum key distribution (CVQKD) has attracted extensive attention due to its compatibility and low costs. However, bandwidth mismatch exists to varying degrees between the transmitter and receiver. This may prevent frequency components carrying modulation information from being fully perceived by the legitimate party. In this paper, we identify a practical security loophole caused by bandwidth mismatch and propose a corresponding spectral attack scheme. Different from previous approaches that exploit security loopholes to conceal the excess noise introduced by intercept-resend attacks, this scheme can directly obtain raw-key information without introducing additional disturbances. A proof-of-principle attack on a CVQKD system with filtering operation is constructed to verify the feasibility. Experimental results indicate that Eve can obtain enough information to render the system insecure if this practical security loophole is ignored. Based on the identified security loophole, corresponding defense strategies are proposed. This work helps bridge the gap between theoretical models and practical implementations, providing a reference for defense design in practical quantum communication systems.

quant-ph↗

arXiv2026

Legible Shared Autonomy: Implicit Communication of Robot Belief through Motion

Shared autonomy systems combine user input with autonomous assistance to help users with motor impairments control robot arms to perform everyday manipulation tasks, by inferring user goals and providing appropriate guidance. However, the robot's internal beliefs about user goals cannot be observed by users. Traditional shared autonomy systems provide assistance along efficient shortest paths toward inferred goals, but when multiple objects lie in similar directions, such assistive motion remains ambiguous and fails to reveal the specific goal identified by the robot. This creates two critical problems. First, when the robot correctly infers the goal, users continue controlling because they cannot perceive understanding from ambiguous assistive motion, wasting effort when autonomous completion would suffice. Second, when the robot misunderstands intent, users cannot quickly detect errors until assistive motion diverges significantly, requiring substantial corrective input. We address this by introducing legible motion into shared autonomy, where robot actions must both advance toward the goal and clearly reveal which goal has been inferred, enabling users to understand the robot's beliefs and adjust control accordingly. The robot modulates communication strength through confidence-aware adaptive authority allocation by providing assertive legible assistive actions when confident while increasing user authority when uncertain, transforming shared autonomy into transparent bidirectional collaboration. User studies including simulation, and physical experiments with a six-degree-of-freedom robot arm demonstrate that legible shared autonomy significantly improves users' understanding of robot beliefs and reduces user control effort compared to standard shared autonomy.

cs.RO↗