arXiv subjects

Tian Lan

Publications and source records attributed to Tian Lan.

At least 19 recordsLinked to original sources

arXiv2026

Too Hard, Too Easy, Just Right: Performance-Driven Challenge Adaptation in Virtual Reality Firefighter Training

Effective firefighter training requires practice under dynamic and stressful emergency conditions, yet real-world live-fire exercises are costly, risky, and difficult to scale. Virtual Reality (VR) offers a safe and controllable alternative, but many existing VR firefighter training systems rely on fixed scenarios or provide limited difficulty adaptation, making it difficult to maintain an appropriate level of challenge for skill development as trainees' performance changes over time. To address this limitation, we present ProACT, a performance-driven adaptive training system that regulates scenario difficulty through closed-loop interactions between trainee performance and simulated fire dynamics. Specifically, ProACT combines a global module that continuously adjusts scene-level fire spread dynamics in real time based on trainee performance with a local module that periodically updates object-level flammability across the environment. Our evaluation shows that ProACT improves both objective training performance and subjective user experience compared with existing adaptive and non-adaptive baselines. We further discuss how adaptive immersive environments can extend beyond firefighting to broader training domains.

cs.AI↗

arXiv2026

Diff-Feat: Diffusion-Based Cross-Modal Feature Extraction for Multi-Label Classification

Multi-label classification has broad applications and depends on powerful representations capable of capturing multiple interactions. We introduce Diff-Feat, a simple yet effective framework that extracts intermediate features from pre-trained diffusion-Transformer models for both images and text, and fuses them for downstream tasks. We observe distinct patterns in optimal feature selection: for vision tasks, the most discriminative intermediate feature along the diffusion process occurs at the middle step and is located at the middle block in Transformer. In contrast, for language tasks, the best feature occurs at the noise-free step and is located at the deepest block. Notably, a consistent phenomenon across diverse datasets: a fixed "Middle Layer" yields the best performance on various downstream classification tasks for images (under DiT-XL/2-256$\times$256 and various U-ViT architectures). We further analyze the underlying mechanism and its scaling behavior across model scales. To efficiently select representations, we devise a heuristic local-search algorithm that pinpoints the locally optimal "image-text"$\times$"block-timestep" pair among a few candidates, avoiding an exhaustive grid search. A simple fusion-linear projection followed by addition-of the selected representations achieves 46.1% mAP on Visual Genome 500 (with 500 categories) and 80.2% mAP on Pix2Cap-COCO (with 80 categories), consistently outperforming strong baseline methods. Code is available at https://github.com/lt-0123/Diff-Feat.

cs.CV↗

arXiv2026

PlaceReasoner-Beta: Reasoning-Driven Macro Placement and Benchmarking

Automated macro placement remains a fundamental challenge in VLSI physical design. Despite decades of research, existing approaches predominantly optimize hand-crafted proxy objectives, such as estimated wirelength, and typically produce placements through one-shot numerical optimization, limiting their ability to incorporate visual layout context, codified design expertise, and downstream physical-design feedback in a unified loop. We present PlaceReasoner-Beta, a verifier-guided multi-agent framework that reformulates macro placement as a closed-loop reasoning problem rather than black-box optimization. A vision-language model (VLM) planner generates candidate placements from the floorplan image, macro specifications, and connectivity structure; a geometric verifier enforces physical legality and expert placement principles; a physical verifier refines candidates using early implementation feedback; and a post-route optimizer further improves promising layouts using final PPA. To enable reproducible evaluation, we introduce PlaceReasoner-Bench, a fully open end-to-end benchmark built from open RTL designs, EDA tools, and technology libraries. It comprises 8 designs at two aspect ratios, yielding 16 tasks with fixed floorplans and I/O assignments, so methods differ only in macro positions and orientations and are evaluated using routed PPA and DRC rather than pre-route proxies. Across the benchmark, PlaceReasoner-Beta achieves the best timing among DRC-clean methods on all square tasks, reducing post-route TNS by 61.2% at 1:1 and 53.0% at 2:1 relative to the classical baseline field. It also shortens routed wirelength on most designs despite never explicitly optimizing it, demonstrating that reasoning over spatial structure under physical-design feedback can improve end-to-end layout quality beyond proxy-objective optimization.

cs.AI↗

arXiv2026

CAESAR: Mixed Reality Search-and-Rescue Platform for Cognitive Attack Evaluation

Mixed Reality (MR) enables immersive training by tightly coupling virtual content with the physical environment and supporting embodied interaction. In safety-critical collaborative tasks such as search-and-rescue, however, users rely heavily on virtual elements for perception, coordination, and decision-making, making them vulnerable to cognitive attacks that subtly distort visual feedback, virtual content, or system synchronization. While prior MR security research has examined individual attack mechanisms, evaluations are typically conducted in isolated or single-user settings, and there is a lack of standardized collaborative testbeds for systematic adversarial experimentation. In this paper, we develop CAESAR, an end-to-end multi-user MR framework for systematic evaluation of cognitive attacks in collaborative search-and-rescue training. CAESAR features a modular architecture that integrates four classes of cognitive attacks within a synchronized collaborative scenario and incorporates a multimodal data collection pipeline that captures behavioral, performance, and subjective measures under adversarial conditions. We conduct a controlled user study with an in-depth analysis of the latency-based attack, demonstrating measurable effects on task performance and user behavior while characterizing users' subjective experience under adversarial conditions.

cs.HC↗

arXiv2026

MOSS: Multimodal Origami Swamp Robot System

Robotic systems for environmental monitoring and ecological surveying in wetlands and swamps must operate across diverse terrestrial and aquatic conditions. Conventional fixed-geometry wheels cannot adapt their contact geometry to these varying conditions. This work presents MOSS (Multimodal Origami Swamp Robot System), a compact amphibious robotic platform with 3D-printed origami wheels. The compact crank mechanism switches the wheel between contracted and expanded configurations, changing its width, body diameter, and fin orientation simultaneously. The contracted configuration is better suited to efficient rolling on firmer ground, whereas the expanded configuration provides greater contact and terrain engagement on softer surfaces. MOSS was evaluated on soil, sand, mud with varying water content, and water, in addition to in-motion morphing and terrestrial-aquatic transition tests. The more energy-efficient wheel configuration depended on terrain. The contracted configuration achieved a lower cost of transport on firmer soil and mud, whereas the expanded configuration performed better on wetter mud, sand, and water, reducing the cost of transport by up to 29%. Expanded wheels also reduced slip in wet mud and reduced both slip and hub sinkage in sand. MOSS successfully completed configuration changes in $1.95 \pm 0.11$ s and demonstrated continuous locomotion across soil-mud-water transitions. Outdoor trials further demonstrated operation across natural deformable terrain and water. These results show that adaptive wheel geometry provides terrain-dependent locomotion benefits and enables a single compact platform to traverse substantially different terrestrial and aquatic conditions.

cs.RO↗

arXiv2026

Optimizing Prompt Sequences using Monte Carlo Tree Search for LLM-Based Optimization

Mathematical optimization is a core tool in engineering, operations research, and data science, but modeling real problems remains difficult and time-consuming. In practice, an expert must work with clients to translate informal domain descriptions into precise objectives, constraints, and solver-ready code; small mistakes in this process can lead to infeasible, nonconvex, or suboptimal formulations, and the iterative process is often slow. Large language models (LLMs) could automate this translation from natural language to optimization code, but standard prompting often fails on realistic tasks: it struggles with long, structured task descriptions, mishandles constraints, and produces infeasible, non-convex, or non-executable programs. We introduce Monte Carlo Tree Search for Reasoning Decomposition Plans (MCTS-RDP), which performs pre-hoc test-time scaling by searching over decomposition plans and their resulting programs before committing to the final output. MCTS-RDP treats plan construction as a sequential decision problem: given a problem description, it decomposes the task descriptions into logical components and generates candidate plan steps for each component. Monte Carlo Tree Search (MCTS) then explores and refines Reasoning Decomposition Plans (RDPs) prior to code synthesis, balancing exploration and exploitation across alternative plan sequences. Each candidate plan is rolled out into executable code, executed, and scored with an execution-grounded evaluator measuring executability, constraint satisfaction, and solution quality; this feedback is backpropagated through the search tree to guide subsequent search. Across wireless optimization tasks and the NL4OPT and NLP4LP benchmarks, MCTS-RDP consistently outperforms strong baselines, improving execution success, reward, stability, and optimality attainment.

cs.LG↗

arXiv2026

MELD: Recursive Self-Improvement via Hierarchical Coordination for Multi-Agent Scientific Discovery

Multi-agent scientific discovery with large language models raises a coupled learning problem: research assignments determine both which experiments are pursued and the experiences from which agents adapt. We introduce MELD, a unified hierarchical test-time training framework that jointly adapts a team-level Coordinator policy and agent-level individual Scientist policies within a single discovery process. The Coordinator generates a joint portfolio of search branches and research directives over a shared derivation tree, while Scientists develop hypotheses and execution plans implemented by a shared frozen Executor. Both levels receive alternating GRPO-style parameter updates under a shared smooth best-of-budget objective, allowing team organization and individual research behavior to co-evolve in a recursive self-improvement loop. Our analysis establishes objective alignment for the hierarchy, characterizes conditions for coordination gains over independent parallel discovery and the first-order benefit of private adaptation, and derives a finite-budget regret bound for adapted models. On 22 MLGym and MLE-bench tasks, MELD is best on 20 of them and achieves the highest aggregate normalized scores on MLGym, using a 4B backbone for coordination and research. Most baselines use proprietary Scientists, and all systems share one frozen Executor.

cs.AI↗

arXiv2026

Principal component error in high-dimensional factor models

In a statistical factor model, principal components (or eigenvectors) of a sample covariance matrix serve as estimates of principal directions, the true drivers of co-movement of a collection of observed variables. We write the often substantial error in these estimates as a sum of two interpretable terms, which we show have almost sure asymptotic limits as the number of variables grows with sample size bounded. This scenario is commonplace in financial economics, genomics, machine learning and signal processing. Out-of-subspace error measures the distance from an estimate to the subspace spanned by population factor exposures. It can be expressed in terms of data, providing an estimable floor for asymptotic error. In-subspace error is driven by a rotation error within that subspace, which arises from the fixed sample size of the latent factor returns and cannot be estimated from data alone. We illustrate our error analysis with a three-factor simulation of the US public equity market, showing the dependence of the magnitude of the error and its components on dimension and sample size. In that simulation, out-of-subspace error dominates. Researchers who rely on principal component analysis to estimate factor models can use our results to quantify errors in model-based predictions and attributions.

math.ST↗

arXiv2026

Uniform estimates for elliptic systems with critical Sobolev coefficients

The $W^{1,p}$ estimate for uniformly elliptic operators of divergence form with VMO coefficients is classical. In particular, this provides $W^{1,p}$ estimates for elliptic operators with coefficients in $W^{1,n}$. However, this theory does not yield constants in the estimates controlled by the $W^{1,n}$ norm of the coefficients, since the total $W^{1,n}$ norm does not control the rate of decay for the mean oscillation. In this work, we establish uniform estimates for divergence-form elliptic systems whose coefficients lie in the critical Sobolev--Lorentz spaces $W^{1,(n,s)}$ with $1\le s<\infty$, have an $L^\infty$-bounded spatially symmetric part, and satisfy an integral ellipticity condition. The constant in the estimate is controlled by the corresponding norms of coefficients and the ellipticity constant.

math.AP↗

arXiv2026

APRiL: An Open-ended Benchmark for Chinese Social Awareness in Large Language Models

Large language models are increasingly used in social interactions, where appropriate responses require recognizing implicit cues, inferring mental states, identifying conflicts, and considering potential consequences. However, existing benchmarks often focus on specific social abilities, constrained task formats, or particular interaction settings, offering limited coverage of social awareness across diverse contexts. To address this gap, we introduce APRiL, an open-ended benchmark for Chinese social awareness with instance-specific rubrics. APRiL contains 2,196 Social Awareness Evaluation Instances across six categories and two complementary tasks: Social Scenario Analysis evaluates models' understanding of social cues, mental states, and conflict structures, while Socially Aware Response assesses whether they can translate this understanding into appropriate, natural responses. Experiments on 12 models from six families reveal a consistent understanding-to-response gap. Higher-quality scenario analyses improve responses, while weak analyses may reduce response quality. APRiL provides a systematic resource for advancing socially aware language models in Chinese contexts. Our data and code are available at https://anonymous.4open.science/r/APRiL-05DE/.

cs.CL↗

arXiv2026

VETime: Vision Enhanced Zero-Shot Time Series Anomaly Detection

Time-series anomaly detection (TSAD) requires identifying both immediate Point Anomalies and long-range Context Anomalies. However, existing foundation models face a fundamental trade-off: 1D temporal models provide fine-grained pointwise localization but lack a global contextual perspective, while 2D vision-based models capture global patterns but suffer from information bottlenecks due to a lack of temporal alignment and coarse-grained pointwise detection. To resolve this dilemma, we propose VETime, the first TSAD framework that unifies temporal and visual modalities through fine-grained visual-temporal alignment and dynamic fusion. VETime introduces a Reversible Image Conversion and a Patch-Level Temporal Alignment module to establish a shared visual-temporal timeline, preserving discriminative details while maintaining temporal sensitivity. Furthermore, we design an Anomaly Window Contrastive Learning mechanism and a Task-Adaptive Multi-Modal Fusion to adaptively integrate the complementary perceptual strengths of both modalities. Extensive experiments demonstrate that VETime significantly outperforms state-of-the-art models in zero-shot scenarios, achieving superior localization precision with lower computational overhead than current vision-based approaches. Code available at: https://github.com/yyyangcoder/VETime .

cs.CV↗

arXiv2026

PDAGENT-BENCH: Characterizing, Grounding, and Architecting LLM/VLM Agents for VLSI Physical Design

Large Language Models and vision-language models have shown remarkable success in the front-end design of Very Large-Scale Integrated Circuits, yet their capabilities for VLSI physical design remain significantly underexplored. The primary cause is the lack of standardized benchmarks for evaluating agentic physical design workflows that require high-dimensional, multi-stage optimization under strict design constraints, coordinated interaction with diverse Electronic Design Automation tools, and iterative refinement. This work introduces PDAGENT-BENCH, a comprehensive and multi-dimensional benchmark for evaluating LLM/VLM-based agents across the physical design stack. PDAGENT-BENCH integrates both task-level assessment and workflow-level execution. The benchmark suite contains 353 curated problems that combine conceptual questions with real-world industrial artifacts, with expert-validated references and executable solutions. In addition, the benchmark provides a unified, human-aligned agentic physical design workflow framework that enables closed-loop evaluation of holistic physical design in realistic EDA environments. Experiments on 11 state-of-the-art models reveal that while modern LLMs/VLMs perform competitively on conceptual tasks, they remain substantially limited in tool-centric execution (e.g., 42.2% on Innovus script generation) and long-horizon, multi-stage reasoning. Our studies further show that human-skill-enhanced agentic workflows significantly improve end-to-end physical design performance. PDAGENT-BENCH establishes a standardized, reproducible, and realistic evaluation framework for advancing LLM/VLM-driven holistic physical design automation. To ensure full reproducibility, we release \textsc{PDAgent-Bench} at https://github.com/fenggev56/PDAGENT-BENCH.

cs.AR↗

arXiv2026

Algebraic Tensor Network Renormalization and Holographic duality

Emergent generalized symmetries and the holographic principle play very important roles in understanding quantum phase transitions. In recent years, tensor-network renormalization with generalized symmetry and the corresponding fixed-point tensor formulation have been proposed as a discrete spacetime framework for reformulating conformal field theory. Nevertheless, a complete holographic formulation within this framework is still lacking, and the current formulation does not directly accommodate sector partition functions that are modular covariant in general. In this work, we formulate $\mathcal{R}$-TNR, which builds the categorical constraints of a fusion category $\mathcal{R}$ into local tensor spaces and preserves these constraints under Loop-TNR. For suitable choices of $\mathcal{R}$, the initial tensors constructed here flow under $\mathcal{R}$-Loop-TNR to stable gapless fixed points. In the Ising, tricritical Ising, and three-state Potts conformal field theory examples, a broad range of initial parameters leads to the same gapless fixed point in each example. The method resolves conformal towers by anyon sector and yields scaling dimensions, conformal spins, and magnitudes of sector-resolved structure constants with extremely high accuracy compared to previous methods. These results realize a topological bootstrap based on stable symmetry-topological-order in which finite categorical and geometric input, together with an initial tensor that need not be fine-tuned, suffices to recover the infrared conformal field theory and its sector-resolved conformal data.

cond-mat.stat-mech↗

arXiv2026

Regularity of critical immersions for scale-invariant curvature energies in even dimensions

For immersions of a closed $2h$-dimensional manifold $Σ$ into $\mathbb R^m$, we consider curvature energies that are invariant under ambient isometries and dilations, with the principal term $\int_Σ \big|\nabla^{(h-1)} \vec{\mathrm{I\!I}}\big|_g^2\,d\text{vol}_g$ and lower-order submanifold invariants. Following the four-dimensional approach developed in joint work with Bernard, Martino, and Rivière, we prove that every weak critical immersion in the natural Sobolev class $W^{h+1,2}$, whose induced metric and its inverse have $L^\infty$ coefficients, is real-analytic in harmonic coordinates. The proof combines geometric conservation laws, structural identities, and elliptic estimates with critical Sobolev coefficients to obtain Morrey decay and bootstrap to full regularity.

math.AP↗

arXiv2026

Anlu: Enabling In-Context Time Series Anomaly Detection in Foundation Models via Counterfactual Supervision

Whether a time-series pattern is anomalous often depends on the operating regime of the monitored process. A missing event can signal a fault in one regime and be routine in another, and the query alone may not reveal which regime applies. We study in-context learning (ICL) for time series anomaly detection (TSAD) through reference-conditioned detection, where a reference record provides evidence about expected behavior and model parameters remain fixed at inference. Supplying the reference is not enough: when training anomalies are recognizable from the query alone, the detector can fit its targets while ignoring the reference. We therefore introduce counterfactual supervision, which pairs one query with two references that support different normal rules and labels the query under each. At positions where the two labels disagree, no detector that ignores the reference can fit both targets. Anlu learns from this supervision by adding a reference memory and zero-initialized gated adapters to a frozen time-series foundation model (TSFM) pretrained for anomaly detection. On the 350 TSB-AD-U evaluation sequences, Anlu raises the mean VUS-PR of the frozen TSFM from 0.542 to 0.607. Replacing the reference with zeros lowers Anlu's score to 0.499.

cs.LG↗

arXiv2026

TIPS: Topological Ill-Posedness Probing and Steering in Large Language Models

Ill-posed questions, including those involving ambiguity, under-specification, or conflicting statements, may admit no valid answer or multiple plausible answers, posing a significant challenge for large language models (LLMs) despite their otherwise strong performance. Existing work often treats LLM reasoning as a black box and focuses on input-output analysis. Can a compact, unified topological representation of internal model states capture diverse sources of ill-posedness and steer reasoning toward responses appropriate to each source? To this end, we study the internal state from a topological perspective by treating the contextual hidden states of its prompt tokens at a single transformer layer as a point cloud, capturing the input problem's relational structure as encoded by the model. Our analysis shows that zero- and one-dimensional persistent homology---tracking how connected components merge as the distance threshold increases ($H_0$) and the emergence and filling of one-dimensional loops in the filtration ($H_1$), respectively---yield seven compact descriptors of ill-posedness that also serve as control signals for steering model reasoning. Thus, we develop a topology-conditioned steering mechanism. It retrieves topologically similar positive and negative examples for each query to form a local activation-space contrast. The resulting query-specific direction modifies internal states during prefill and decoding, guiding the model's responses toward suitable abstention or clarification tailored to the underlying source of ill-posedness. Empirical evaluations on two datasets and open-weight models spanning four families demonstrate the effectiveness of this compact representation for ill-posedness classification, while topology-conditioning significantly improved steering performance.

cs.AI↗

arXiv2026

Global Optimization on Graph-Structured Data via Gaussian Processes with Spectral Representations

Bayesian optimization (BO) is a powerful framework for optimizing expensive black-box objectives, yet extending it to graph-structured domains remains challenging due to the discrete and combinatorial nature of graphs. Existing approaches often rely either on full graph topology, which is impractical for large or partially observed graphs, or on incremental local exploration, which can lead to slow convergence. We introduce a scalable framework for global optimization over graphs that builds Gaussian process (GP) surrogates from low-rank spectral representations inferred from sparse structural observations. The method fits a low-rank graph surrogate to the observed connections and uses its spectrum to define a kernel over all nodes, so that function evaluations and structural observations jointly refine a global model, enabling efficient global search and principled uncertainty estimation even with limited data. We further show how a limited budget of structural observations can be allocated according to its expected value for the optimization. We also provide theoretical analysis establishing conditions under which the surrogate geometry is identified under different sampling regimes, and bounding the optimization error incurred under an estimated geometry. Experiments on synthetic and real-world datasets demonstrate that our approach achieves lower regret and improved optimization performance compared to prior methods.

cs.LG↗

arXiv2026

Generalist Representation, Specialist Detection: TS-Router for Time-Series Anomaly Detection

Time-series anomaly detection (TSAD) is difficult to generalize across datasets because heterogeneous temporal dynamics imply different notions of normality and favor different detection criteria. While time-series foundation models provide transferable representations, coupling them with a fixed anomaly-scoring mechanism can overlook this variation. This motivates a different perspective on foundation-model-based TSAD: using foundation models to coordinate specialized anomaly criteria rather than directly imposing a universal one. Based on this view, we propose \textbf{TS-Router}, a generalist-representation, specialist-detection framework that estimates the relative competence of heterogeneous anomaly detectors from pretrained temporal representations and selects suitable specialists for each target series. To avoid relying on specialist-performance labels from real tasks, we derive soft competence supervision from specialists' relative performance on labeled simulated tasks. At deployment, routing requires no target anomaly labels, and only the selected specialists are fitted unsupervisedly on the target series. We bound Top-$k$ set-competence regret under representation coverage and conditional competence stability. Across 16 real-world benchmarks and four complementary evaluation metrics, TS-Router achieves the best overall average rank. Controlled ablations with multiple frozen TSFM encoders further support the use of pretrained representations for competence estimation and adaptive specialist selection. The code is available at https://anonymous.4open.science/r/TS-Router-D8FF.

cs.LG↗