Searcharxiv⌕ Search

arXiv subjects

Yu Guo

Publications and source records attributed to Yu Guo.

At least 19 recordsLinked to original sources

Agent-Centric Observation Adaptation for Robust Visual Control under Dynamic Perturbations

Visual control maps visual streams to actions in reinforcement learning and vision-language-action (VLA) policies. Changing weather, noise, blur, and compression can obscure different task cues within one episode. Existing control benchmarks mainly test scene distractions, while generic restorers optimize whole-image fidelity. We introduce the Visual Degraded Control Suite (VDCS) to evaluate seven degradations under Markov switching. We propose Agent-Centric Observations with Mixture-of-Experts (ACO-MoE), which routes each frame to a repair expert, restores RGB content, and composes task-relevant foregrounds. On eight VDCS tasks, corruption lowers DreamerV3's mean return from 769 to 562. ACO-MoE raises it to 730, recovering 81.2% of the lost return. It also improves returns on DMC-GB and RoboSuite and retains 70% to 97% of clean-input mean return under six unseen perturbation families. We train ACO-MoE on robot demonstrations and deploy it with a 3B-parameter VLA policy on a Piper arm for cube placement in blue and white boxes and cup placement on a coaster. On the blue-box and cup tasks, motion blur and physical low light with added noise reduce success from 100% to at most 10%. ACO-MoE raises success to 75% to 95% over 20 trials per task and condition.

cs.RO↗

Inertial Video World Models

The physical world has inertia: it evolves from what already exists rather than being recreated at every instant. However, many existing video world models lack an explicit mechanism for inertial state evolution, leaving the continuity of motion and structure implicit in video prediction. During autoregressive rollouts, repeated reconstruction can accumulate errors and destabilize the predicted dynamics. To address this limitation, we introduce InertialWorld, an Inertial Video World Model that makes inertia a structural prior for autoregressive prediction: existing scene evidence is carried across time rather than reconstructed at every step. Central to this formulation is Inertial Memory, which turns memory from passive predictive context into an evolving video state with two complementary views: a dense Eulerian grid preserves scene-aligned evidence, while persistent Lagrangian carriers bind local content to learned canonical coordinates. At each step, the model transports the carrier content according to action-conditioned motion, predicts the complete next video state from the transported carriers and dense grid, and uses a learned gate to combine transported and newly predicted content into the memory for the next step. Extensive experiments across multiple domains, including robot manipulation, 3D game simulation, and open-world navigation, demonstrate that InertialWorld achieves competitive autoregressive prediction performance against strong video world model baselines. Codes link: https://github.com/YuanGao-YG/InertialWorld

cs.CV↗

PMR: MoE Routing with Shared Prototypes and Congestion Pricing

Conventional linear routers in sparse mixture-of-experts models select experts directly from token-expert matching scores, coupling input matching with expert allocation. We propose Prototype-Mediated Routing (PMR), a two-stage router with shared prototypes. PMR first computes each token's distribution over prototypes, then uses it to mix prototype-conditioned expert distributions according to the law of total probability. This formulation models input matching and expert allocation separately, explicitly represents many-to-many associations between prototypes and experts, and provides separate interfaces for context adaptation and load regulation. Within this framework, short-term states summarize preceding tokens to adapt the prototypes, while congestion prices updated from dispatch feedback achieve loss-free load balancing by adjusting the conditional expert distributions. We use a two-pass blockwise approximation to accelerate short-term state processing, achieving over 20 times the training throughput of the eager tokenwise PMR implementation while maintaining high agreement in routing decisions. Across three pretraining configurations with 0.43B, 1.18B, and 0.48B total parameters, PMR improves mean zero-shot accuracy on seven benchmarks over linear routing baselines with matched expert architectures by 1.54, 2.65, and 1.45 percentage points (pp), respectively. The corresponding gains on ARC-Easy are 6.71, 5.80, and 3.92 pp. In parameter-efficient fine-tuning of 7B-8B backbones, PMR's improvements over MixLoRA vary across backbones, with gains of up to 1.76 pp in mean accuracy across four tasks.

cs.AI↗

Rethinking Stepwise Model Routing: A Cost-Efficient Table Reasoning Perspective

Large Reasoning Models (LRMs) achieve strong performance on table reasoning tasks but incur substantial inference cost due to long reasoning traces. Stepwise model routing mitigates this issue by dynamically assigning reasoning steps to smaller or larger models. However, stepwise model routing for table reasoning remains underexplored. Through empirical analysis, we find that reasoning steps involving tables contain two types of tokens with distinct uncertainty distributions: table tokens grounded in table structure, such as cell values and headers, and text tokens representing surrounding natural-language reasoning. The uncertainty of both token types is correlated with the risk that the model makes an error in the next reasoning step. However, existing methods fail to model them separately, leading to suboptimal routing decisions. To address this, we propose EcoTab, a table-aware stepwise routing framework for efficient table reasoning. At each reasoning step, EcoTab separately estimates the uncertainties of table tokens and text tokens, maps them to next-step failure risks for the small model, and combines the two risks for routing. Experiments on multiple table reasoning benchmarks show that EcoTab consistently outperforms strong baselines and achieves a better balance between accuracy and efficiency.

cs.CL↗

Rubric-Guided Process Reward for Stepwise Model Routing

Stepwise model routing improves the efficiency of Large Reasoning Models (LRMs) by assigning each reasoning step to a suitable model. Recent methods formulate routing as a sequential decision process and train the router with reinforcement learning. However, although they model routing as a process, they still supervise the router with outcome rewards. Such rewards only reflect final answer correctness and fail to evaluate intermediate routing decisions, which can weaken performance and generalization. To address this gap, we propose RoRo, a rubric-guided process reward framework for stepwise model routing. RoRo first collects diverse routing trajectories and constructs preference pairs based on outcome, cost, and process quality. It then trains a Rubricor to generate a query-specific evaluation rubric and a Judge to score routing trajectories under this rubric through alternating optimization. The resulting process rewards are combined with outcome rewards to optimize the routing policy via GRPO. Experiments on five reasoning benchmarks under both same-family and cross-family settings show that RoRo consistently outperforms strong baselines and achieves better accuracy and cost trade-offs.

cs.AI↗

Measure of set coherence

Set coherence is a quantum resource for a set of states that characterizes whether a collection of quantum states contains coherence, namely whether these states fail to be simultaneously incoherent in a common basis. In this work, we show that the free operations of set coherence are common commutativity-preserving quantum channels, from which we then establish an axiomatic framework for set-coherence measure (SCM). Within this framework, we investigate the robustness based measures deduced by taking, respectively, the average, maximum, and sum of the single-state robustnesses over a common reference basis, where the first two were previously introduced in the literature and we check here whether they satisfy our SCM axioms. Consequently, we show that the averaged robustness is an SCM but it is neither unified nor complete, the maximal robustness is a unified SCM but is neither complete nor strongly superadditive, whereas the sum robustness is a complete SCM. Furthermore, we propose several classes of SCMs based on pairwise commutators. We show that the Schatten $p$-norm based SCMs are complete and strongly superadditive. We also find that the universal fourth-order Bargmann gap measure proposed in the literature is a complete and strongly superadditive quantifier of set coherence. Finally, we discuss the operational meaning of the SCM in the context of optimal approximate quantum broadcasting, and show that the amount of set coherence can closely reflect the performance of this task.

quant-ph↗

Bayesian inference of the magnetic component of quark-gluon plasma

The chromo-magnetic monopoles (CMM), emergent topological excitations of non-Abelian gauge fields carrying chromo-magnetic charge, have long been postulated to play an important role in the vacuum confinement of quantum chromodynamics (QCD), the deconfinement transition at temperature $T_c\approx 160\rm MeV$, as well as the strongly coupled nature of quark-gluon plasma (QGP). While such CMMs have been found to provide solutions for challenging puzzles from heavy-ion collision measurements, they were typically introduced as model assumptions in the past. Here we show how their very existence can be determined and their abundance extracted in a data-driven way for the first time. Using the \textsc{cujet3} framework for calculations of jet energy loss and analyzing a comprehensive experimental data set for nuclear modification factor ($R_{\mathrm{AA}}$) and elliptic flow ($v_2$) of high-transverse-momentum hadrons, the fraction of CMMs in the QGP is obtained by Bayesian inference and is found to be substantial in the $1\sim 2 T_c$ region. The posterior CMM fraction is further validated by excellent agreement with additional data and is also shown to predict QGP transport properties quantitatively consistent with the state-of-the-art knowledge.

hep-ph↗

Dual-Hypergraph Indexing: Bridging Knowledge Islands for Multi-Hop Reasoning in Retrieval-Augmented Generation

While hypergraph-based Retrieval-Augmented Generation (RAG) effectively captures higher-order multi-entity correlations, existing paradigms treat extracted hyperedges as isolated factual assertions. This structural fragmentation engenders rigid "knowledge islands" that bottleneck multi-hop causal inference, temporal tracking, and narrative synthesis. To systematically address these challenges, we introduce Dual-Hypergraph Indexing (DHI), a hierarchical representation framework that elevates discrete facts into structured analytical insights. DHI couples a foundational entity-relation factual hypergraph ($H_K$) with an elevated deep-insight hypergraph ($H_D$) via a dual-pathway aggregation algorithm. Specifically, DHI employs: (1) importance-driven hub aggregation via 5-metric topological profiling and adaptive thresholding to capture spatial semantic clusters; and (2) temporal chunk-chain progressive aggregation via sliding-window greedy exploration to track chronological evolutions. Across five benchmarks, DHI achieves state-of-the-art performance, boosting logical coherence by +1.53 on the multidisciplinary Mix benchmark and scoring 85.78\% on complex medical pathology reasoning tasks. DHI provides a robust architecture for next-generation multi-hop RAG.

cs.IR↗

Asymmetric Dynamic Routing: Balancing Reasoning Depth and Computational Efficiency in Hypergraph RAG

While graph-based and hypergraph-based Retrieval-Augmented Generation (RAG) significantly mitigate hallucinations in Large Language Models (LLMs), existing structure-based RAG systems typically adopt static traversal strategies regardless of the query complexity. We identify this ``static retrieval fallacy'' as a primary source of computational redundancy for simple queries and cognitive context gaps for complex reasoning tasks. To balance reasoning quality and inference efficiency, we propose Asymmetric Dynamic Routing (ADR), an intent-conditioned retrieval framework operating over hierarchical knowledge graphs. ADR employs a lightweight structured classifier to dynamically dispatch queries among three asymmetric topological traversal operators: localized fact anchoring, bottom-up adjacency diffusion, and top-down insight grounding, which collectively enable bidirectional information flow across hierarchical knowledge layers. Extensive empirical evaluations across five domain-specific corpora demonstrate that ADR maintains strong reasoning performance while reducing prompt token consumption by up to 48.7\% and end-to-end query latency by 45.3\%, yielding a favorable quality--efficiency trade-off for query-adaptive Hypergraph RAG.

cs.IR↗

FRUC: Feedforward Dynamic Scene Reconstruction from Uncalibrated Collaborative Driving Views

We present FRUC, a feedforward 3D Gaussian Splatting framework for dynamic scene reconstruction from uncalibrated collaborative driving views. Existing multi-agent reconstruction frameworks are often hindered by rigid prerequisites, demanding precise spatial calibration and slow per-scene optimization. In this paper, we rethink this task by conceptualizing a distributed multi-vehicle network as a spatio-temporally unstructured ego-centric multi-camera system, where the core challenge lies in enhancing ego-centric occluded geometry through collaboration without degrading the ego's accurately observed visible geometry, while preserving reconstruction efficiency. For efficient reconstruction, FRUC is built upon a visual grounded geometric Transformer backbone to enable one-shot, calibration-free inference from a flexible number of multi-vehicle views. To achieve non-destructive geometric supplementation under uncalibrated cross-agent misalignment, FRUC first introduces an ego-centric causal occlusion field that explicitly derives occlusion evolution as latent priors by modeling agent-wise spatio-temporal correlations. Guided by these occlusion priors, it further formulates cross-agent integration as a deterministic residual denoising process via zero-initialized injection, turning challenging cross-agent fusion into bounded residual learning for robust collaborative blind-spot completion. Through extensive evaluations on real-world V2X-Real and UrbanIng-V2X datasets, FRUC is shown to be a new state-of-the-art for the scene reconstruction of dynamic collaborative driving environments, significantly outperforming existing methods in both rendering quality and efficiency. Code is available at https://github.com/yihangtao/FRUC.git.

cs.CV↗

Heavy quark energy loss in the Bayesian-inference-improved CUJET model

We compute jet quenching observables--the nuclear modification factor $R_{AA}$ and elliptic anisotropy coefficient $v_2$--for open-heavy flavor mesons with high transverse momentum in an improved CUJET model, in which the strong coupling parameter as well as the temperature-dependent fraction of chromo-magneto monopoles are extracted through Bayesian inference using high-$p_T$ light-hadron $R_{\mathrm{AA}}$ and $v_2$ measurements from RHIC and the LHC. We show in this work that such a Bayesian-inference-improved CUJET model provides a good description of available high-$p_T$ charm- and bottom-hadron observables based on the same posterior parameters. The corresponding heavy-quark spatial diffusion parameter, $2πT D_s$, is also computed, which is broadly consistent with current phenomenological and lattice-QCD constraints. These results support a universal description for the medium modification of high-$p_T$ light- and heavy-flavor energy loss and indicate an important role played by the chromo-magnetic degrees of freedom near the confinement transition region of quantum chromodynamics (QCD).

hep-ph↗

Image-based investigation of the zebrafish developmental process using in vivo dynamic and multi-contrast optical coherence tomography

We demonstrate in vivo dynamic optical coherence tomography (DOCT) imaging of zebrafish development from 2 weeks to 12 months post-fertilization, integrated with polarization-sensitive OCT (PS-OCT), OCT angiography (OCTA), and histological validation. Two DOCT algorithms were utilized: logarithmic intensity variance and late OCT correlation decay speed, which characterize the occupancy of dynamic scatterers and their motion speeds, respectively. Our results show that skin stripes exhibit high DOCT signals and it varies among the pigment-cell types. Furthermore, the combination of DOCT and PS-OCT captures the maturation of these stripes. In addition, DOCT and OCTA successfully visualized the developmental progression of blood and lymphatic vessels, as well as spinal tissues.

physics.optics↗

SAGE: A Statistical Acceptance Gate for Self-Evolving Agents

Large Language Model (LLM)-based agents increasingly self-evolve by editing a persistent skill document that encodes their workflow, tool-use rules, and decision logic. This loop has two steps, an optimizer that proposes a candidate edit and a gate that accepts or rejects it. Prior work has concentrated on the optimizer, while the gate still follows a naive rule that keeps any edit which improves an aggregate validation score. We show that this rule fails in two ways. First, it admits permanent regressions, since an edit can raise the average while breaking items the skill already solves. Second, it is vulnerable to the Optimizer's Curse, since the best observed score on a finite and noisy validation set is upward biased. To solve the above two limitations, we propose a statistical acceptance gate for self-evolving agents (SAGE). Compared with previous work, SAGE has two contributions. First, SAGE proposes a per-item paired comparison that evaluates the current skill and the edited skill on identical validation items, which exposes regressions that an aggregate score hides and penalizes them asymmetrically. Second, SAGE also employs a one-sided paired test that commits an edit only when its wins are statistically reliable against its losses, and it abstains otherwise. SAGE is a conservative refinement of the standard gate that recovers the baseline exactly at a boundary setting. It commits only a subset of the baseline's edits, filtering out those whose gains are unreliable or purchased by breaking already-solved items. Across five benchmarks and four backbone LLMs under an equal-budget protocol, SAGE lowers the regression rate in 19 of 20 settings and matches the baseline in the remaining one, for example from 36.5% to 0% on LiveMath and from 42.8% to 0% on OfficeQA with DeepSeek-V4. SAGE also attains the highest final score in all 20 settings, raising LiveMath from 34.15 to 48.78.

cs.AI↗

Quantum-state texture measure and texture transformation

Quantum-state texture quantifies structural irregularities of a state in a selected basis and has emerged as a valuable resource for gate characterization in universal circuits. Consequently, a class of contractive distance based texture measures and a framework for constructing convex-roof extended texture measure have been proposed. Here, we extend this theory in several directions. We examine the convertibility of a pure state to any state under texture-free operations, and fully solve the deterministic transformation problem between any two states in the qubit case. We then propose two classes of texture measures: convex-function-based measures and texture cost defined via minimal cost of texture contained in pure states. Moreover, we show that every texture monotone gives rise to a non-negative function which admits the same properties as the function used in the convex-roof extension---thus providing a converse to that construction. Our work thereby offers a relatively comprehensive resource-theoretic formulation of quantum-state texture.

quant-ph↗

Proximity-CLIP: Text-Guided Semantic Proximity Learning for Zero-Shot Anomaly Detection

Vision-language models offer a promising approach for zero-shot anomaly detection (ZSAD). However, due to object-centric bias, normal and anomalous text prototypes exhibit a high semantic overlap. While enforcing strict orthogonality between them improves discriminability, mapping highly contiguous visual inputs onto drastically orthogonal prototypes introduces a geometric dilemma, disrupting the pre-trained structural continuity. To address this problem, we propose Proximity-CLIP, a framework that visually calibrates the semantic margin to guide visual adaptation. First, we introduce a visually-calibrated semantic proximity learning mechanism that uses a bounded dynamic regularization to learn an appropriate semantic margin, ensuring discriminative separation while preserving structural alignment. Second, we design an Anomaly Query Module (AQM) driven by these text priors. Using the calibrated anomalous prototype as a semantic query, the AQM actively retrieves localized defect cues from contextual visual patches, mitigating the dilution of subtle anomalies during global pooling. Extensive experiments demonstrate that Proximity-CLIP outperforms current state-of-the-art methods across multiple ZSAD benchmarks with minimal architectural modifications.

cs.CV↗

CF-VLA: Efficient Coarse-to-Fine Action Generation for Vision-Language-Action Policies

Flow-based vision-language-action (VLA) policies offer strong expressivity for action generation, but suffer from a fundamental inefficiency: multi-step inference is required to recover action structure from uninformative Gaussian noise, leading to a poor efficiency-quality trade-off under real-time constraints. We address this issue by rethinking the role of the starting point in generative action modeling. Instead of shortening the sampling trajectory, we propose CF-VLA, a coarse-to-fine two-stage formulation that restructures action generation into a coarse initialization step that constructs an action-aware starting point, followed by a single-step local refinement that corrects residual errors. Concretely, the coarse stage learns a conditional posterior over endpoint velocity to transform Gaussian noise into a structured initialization, while the fine stage performs a fixed-time refinement from this initialization. To stabilize training, we introduce a stepwise strategy that first learns a controlled coarse predictor and then performs joint optimization. Experiments on CALVIN and LIBERO show that our method establishes a strong efficiency-performance frontier under low-NFE (Number of Function Evaluations) regimes: it consistently outperforms existing NFE=2 methods, matches or surpasses the NFE=10 $π_{0.5}$ baseline on several metrics, reduces action sampling latency by 75.4%, and achieves the best average real-robot success rate of 83.0%, outperforming MIP by 19.5 points and $π_{0.5}$ by 4.0 points. These results suggest that structured, coarse-to-fine generation enables both strong performance and efficient inference. Our code is available at https://github.com/EmbodiedAI-RoboTron/CF-VLA.

cs.CV↗

TempoGround: State-Aware Streaming Visual Grounding with Vision-Language Models

Visual grounding maps language referents to spatial targets and is central to open-vocabulary perception with vision-language models. Existing methods have made substantial progress on single-frame and video-based visual grounding, yet under streaming inputs they still suffer from identity drift, cross-frame inconsistency, and fragile localization under partial occlusion. To address these issues, we present TempoGround, a VLM-native framework that detects cross-frame object correspondence and explicitly models object presence states, thereby enabling accurate and consistent visual grounding under streaming inputs. The key is a curriculum prediction mechanism guided by state-aware cross-frame correspondence: TempoGround resolves 2D instance association, predicts whether each object newly enters, continues in, or leaves the view, decodes the 2D box, and then lifts it to a camera-frame 3D box. As token-level supervision alone cannot capture the geometric objectives of streaming grounding, we further introduce Streaming Grounding Reinforcement (SGR), which optimizes TempoGround with verifiable Grounding, Identity, and Consistency rewards, jointly reinforcing persistent localization and temporally consistent predictions. We carefully design a three-stage training strategy and train TempoGround on large-scale data. We evaluate visual grounding under causally streaming inputs on multiple challenging benchmarks: TempoGround improves F1_2D@0.5 and F1_2D@0.95 by 4.4 and 0.5 on average, and F1_3D@0.25 and AP_3D by 6.2 and 7.5, respectively. These results demonstrate that TempoGround provides a practical foundation for visual grounding under streaming inputs.

cs.CV↗

Measure of set imaginarity

Recent studies have shown that Bargmann invariants provide effective detectors of set imaginarity. In this paper, we investigate set imaginarity as a quantum resource in qubit systems. By exploiting the structure of Bargmann invariants, we show that the free operations for qubit set imaginarity consist precisely of common unital operations and common planarized operations. Based on this characterization, we introduce an axiomatic framework for set-imaginarity measures (SIMs). In particular, we propose two refined notions, namely unified SIMs and complete SIMs, which allow a more fine-grained quantification of set imaginarity. To make these notions concrete, we construct two qubit SIMs from the Bargmann invariants of three-state subsets. We prove that one of them is a unified SIM, while the other satisfies the stronger requirements of a complete SIM. Furthermore, we revisit the robustness of set imaginarity previously introduced in the literature. We show that, although this robustness is a valid SIM for qubit systems, it is neither a unified SIM nor a complete SIM. To overcome this limitation, we propose an improved robustness-type measure and rigorously prove that it defines a complete qubit SIM.

quant-ph↗