Computer Science (arXiv)

Browser agents perform well on short, clean demonstrations, but real deployment is fundamentally different: agents must sustain dozens of decisions on live websites while recovering from mistakes and navigating complex UIs. We argue that closing this gap requires alignment at every level of the pipeline, including execution, supervision, optimization, and evaluation, rather than scale alone. We present Wuying-Browser-Agent, a unified framework that addresses each of these levels. A structured browser harness provides stable execution primitives and decision-oriented context management. Reflection and UI-specialized Curriculum SFT (RUIC-SFT) explicitly trains on recovery trajectories and complex-UI interactions. Divergence-Aware Online GRPO (DAO-GRPO) improves long-horizon credit assignment through potential-based reward shaping and divergence-aware step weighting. Finally, we introduce BrowserBench, a bilingual real-web benchmark of 350 tasks averaging 37.9 steps, because most existing benchmarks are too short to expose long-horizon failure modes. Wuying-Browser-Agent-27B achieves 80.6\% on WebVoyager, 66.7\% on Online-Mind2Web, and 65.1\% on BrowserBench, establishing a new open-source state of the art on browser-use benchmarks. The same pipeline also transfers beyond browser use, demonstrating strong general agentic ability and reaching an average score of 73.8 on Tau2-Bench, Claw-Eval, and BFCL-v4.

Vision-language navigation agents are often evaluated on their ability to follow route-like instructions toward a fixed goal. Yet, real navigation instructions often depend on observed states of the environment: if a condition holds, then follow one path, otherwise take another. Such instructions require an agent to evaluate scene evidence, select the correct logical branch, and execute the corresponding navigation behavior. Existing evaluations provide limited control over conditional branch execution, making it difficult to determine whether agents fail because of perception, grounding, navigation, or logical decision-making. We introduce CondVLN, a scene-graph-grounded benchmark for diagnosing conditional branching in vision-language navigation. CondVLN programmatically generates instructions whose branch conditions are grounded in verifiable 3D scene-graph predicates, with controlled variation in branch depth, dependency chain length, spatial composition, evidence observability, and instruction horizon. CondVLN contains over 11,500 generated conditional instructions across AI2-THOR, Matterport3D, Gibson, and ReplicaCAD, and evaluates agents using standard VLN metrics and branch-specific diagnostics: Branch Selection Accuracy and Conditional Success Rate. Evaluating four state-of-the-art VLN agents (VLN-Zero, NaVid, NaVILA, and Open-Nav) shows that conditional branching exposes failures that are not captured by standard success rate or path length alone: agents can navigate plausibly while committing to a branch inconsistent with the observed scene condition. We also present a lightweight neurosymbolic branch-selection model that separates condition grounding from navigation execution, improving performance by 2x. CondVLN provides a reusable testbed for measuring whether embodied agents can not only follow instructions, but follow the right instruction under the right condition.

Large language model-based recommender systems (LLM-RSs) have demonstrated remarkable capabilities, but are computationally unsustainable for many real-world applications. Compact LLMs offer a practical alternative, yet their reduced capacity often requires reasoning or knowledge distillation methods that increase latency or depend on larger models. Combined with autoregressive generation, these approaches face severe scalability bottlenecks. In contrast, discriminative LLM-RSs enable efficient full-corpus ranking through embedding similarity, but compact backbones remain limited in expressiveness and structural adaptivity. We propose the Fusion of Layer-wise Exits for Sequential Recommendation (FLEXRec), a discriminative framework that enhances compact LLMs while retaining scalable full-corpus ranking. FLEXRec inserts prediction heads (i.e., exits) at multiple transformer layers and adaptively fuses their score distributions. An adaptive continuous router (AC-Router) dynamically selects both the number and identity of exits for each user sequence, while a novel target-k hinge loss regulates routing sparsity. Experiments on three real-world datasets with Qwen 3 1.7B and Llama 3.2 3B show that FLEXRec achieves state-of-the-art accuracy among compact-backbone methods while remaining highly efficient. Code: https://github.com/xurong-liang/FLEXRec

Animatable human avatars are routinely reconstructed from multi-view video under a silent assumption: that every pixel of a frame observes the same instant of the body's motion. Rolling-shutter (RS) sensors expose image rows sequentially, so within one frame the head and the feet of a moving person are separated by tens of milliseconds of articulated motion, and every scanline sees a different pose. Feeding such video to a state-of-the-art avatar bakes the distortion into the canonical representation, where it survives as shear and wobble under novel views and novel poses. Worse, every camera in a rig follows its own readout schedule, so the multi-view consistency that drives the reconstruction is violated even when the geometry is correct. We present RS-Avatar, which reconstructs a sharp, undistorted, animatable 3D Gaussian avatar directly from RS video. The formulation is minimal: a motion-aware avatar already renders the body at several sub-frame instants, and where a blur model averages those renderings, a rolling-shutter model composites them scanline by scanline. Changing that operator is the only modification required. On RS-ZJU, a benchmark we build from ZJU-MoCap, this improves novel-view synthesis over training as if the frames were instantaneous, on every subject. A motion-aware blur model built on the same sub-frame machinery does not transfer, and in fact falls below the shutter-oblivious baseline: the machinery is reusable, the operator is not.

Reinforcement Learning (RL) has been promising in single-turn LLM fine-tuning. However, long-horizon agentic reasoning introduces increasingly branching interactions and sparse rewards, exposing several limitations of RL: its heavyweight backpropagation-based training stack makes it impractical to fine-tune larger LLMs, and longer-horizon trajectories make credit assignment in RL substantially harder. This paper argues that evolution strategies (ES) can be a better choice for fine-tuning long-horizon LLM agents. Compared with agentic RL, ES offers three key advantages: 1) Model Scalability: ES enables full-parameter optimization with only minimal, inference-level GPU memory, making it possible to fine-tune large LLMs. 2) Flexibility: its lightweight, black-box feedback interface makes ES fine-tuning easy to compose with prompt-space evolution (e.g., skill optimization & test-time compute); and 3) Long-Horizon Scalability: ES performs trajectory-level parameter attribution without decomposing rewards across horizons, yielding better scalability than Agentic RL as the horizon length grows. Based on this insight, we propose Agentic ESOpt, a full-parameter agentic fine-tuning framework tailored to flexible parameter--context co-evolution. At each step, Agentic ESOpt samples perturbations around the current LLM parameters, evaluates the resulting agents with rewards, and applies an online reward-weighted update. To improve the exploration--adaptation trade-off, Agentic ESOpt further introduces a cosine decay schedule of the perturbation scale $σ$. On WebArena-Lite, full-parameter optimization of Qwen-3.5-27B improves the No Skill baseline by 6.69%. In test-time automatic heuristic design, Agentic ESOpt performs online prompt--parameter co-evolution, improving its matched baseline in 28 of 36 settings.

While adversarial prompt tuning can enhance robustness of vision-language models efficiently, we find that existing methods aggravate robust generalization overfitting on seen classes, leading to a rapid degradation in performance against adversarial examples of unseen classes as training progresses. We empirically identify that this degradation stems from the tendency of the model to learn pseudo-robust features (i.e., non-generalizable shortcuts). To mitigate this, we propose ADAPT (Adversarial Disentangled Prompt Tuning), a robust prompt tuning framework following the philosophy of ``Learning What Not to Learn''. Specifically, ADAPT uses a dual-prompt mechanism with a target prompt and a pool of decoy prompts. During training, the decoy prompts are guided to entrap diverse pseudo-robust features, while the target prompt is constrained to be orthogonal to the decoys in the embedding space to learn robust features. By disentangling the robust features from the pseudo-robust features, ADAPT effectively prevents robust generalization overfitting. We further provide an analysis showing that the orthogonal loss bounds the effect of shifts in pseudo-robust features on unseen classes, yielding a testing error guarantee. Empirically, extensive experiments demonstrate that ADAPT substantially improves the robustness of the target prompt on unseen classes. The code is available at https://github.com/cheny02/ADAPT-ACMMM2026.

We develop a column-wise chi-squared geometry for discrete memoryless channels (DMCs) yielding tight, logarithm-free bounds on mutual information, channel dispersion, and finite-blocklength coding rates without evaluating logarithms of the channel matrix. The key parameter is~\(η\)---the worst-case relative deviation of a transition probability from its output marginal, which is small precisely when the channel is close to the fully noisy channel $t_{ij}=s_j$. We prove three main results: (1) a third-order ratio expansion showing \(I(X;Y)/χ^2(X;Y)\to 1/2\) as \(η\to 0\) with an \(O(η)\) skewness correction; (2) a two-sided dispersion equivalence bounding \(V(X;Y)\) above and below by \(χ^2(X;Y)\) with explicit constants \(c_{\pm}(η)\to 1\); and (3) a certified robust design rate \(R_{\mathrm{cert}}(n,\varepsilon)\) with total certification gap \(O(η)+O(η/\sqrt{n})+O(\log n/n)\). The certified bounds on \(I\) and \(V\) require only addition, multiplication, division, and square roots; the final rate also uses \(Q^{-1}(\varepsilon)\).

Formal verification is a crucial technique for ensuring the functional correctness of hardware designs. In the context of property checking, a key challenge is how to efficiently prove a user-specified property in the face of increasingly complex RTL designs. To address this challenge, abstraction techniques are often employed to reduce system complexity and accelerate the verification process. However, prior RTL abstraction methods either require significant manual effort or rely on rule-based techniques that lack flexibility. This paper introduces NeuroAbs, a neuro-symbolic framework for RTL abstraction. NeuroAbs first uses LLM-assisted RTL analysis to identify signals suitable for abstraction. It then combines LLM-based abstraction with an AST-based symbolic RTL representation to better align the generated abstraction with the intended transformation. The soundness of each abstraction is checked using satisfiability modulo theories (SMT) solving. If the abstraction is too coarse for a successful proof, NeuroAbs applies counterexample-guided abstraction refinement (CEGAR) to iteratively refine the model. Experimental results show that NeuroAbs significantly improves the efficiency of hardware property checking across a range of verification tasks.

Post-training with supervised chain-of-thought fine-tuning and reinforcement learning from verifiable rewards has substantially improved the mathematical reasoning capabilities of large language models (LLMs). However, their application to signal processing problems remains relatively under-explored. This report investigates reinforcement fine-tuning strategies for adapting Qwen2.5-3B-Base to graduate-level signal mathematical problems from WirelessMATHBench-XL, a comprehensive benchmark for mathematical reasoning in this domain. We examine two training paradigms: (i) direct reinforcement learning (RL) on WirelessMATHBench-XL with verifiable rewards; and (ii) supervised fine-tuning (SFT) on a distilled wireless-domain chain-of-thought corpus, followed by the same domain-specific RL stage. Across both paradigms, we benchmark Group Relative Policy Optimization (GRPO), Group Sequence Policy Optimization (GSPO), and Geometric-Mean Policy Optimization (GMPO). We aim to assess whether domain-aware CoT SFT serves as an effective initialization for subsequent RL, and whether GSPO or GMPO offer advantages in stability or accuracy over GRPO for signal reasoning tasks. Our best model achieves an overall accuracy of 39.12\%, representing a more than threefold improvement over the untrained Base model (12.37\%).

Time Series Foundation Models (TSFMs) have recently emerged as a highly promising paradigm for cross-domain zero-shot forecasting. However, existing evaluation protocols predominantly rely on static benchmarks with fixed historical test windows. While these benchmarks provide a valuable baseline snapshot, they evaluate an average performance on a fixed history, failing to capture how models behave in continuously evolving real-world environments characterized by seasonal variations, distribution shifts, and unexpected events. To bridge this gap, we introduce LiveHouse-TS, the first open-world living benchmark infrastructure for TSFMs. By evaluating models prequentially on real future data in open-world environments, LiveHouse-TS shifts time series benchmarking from snapshot accuracy to continuous temporal validity. Rather than acting as a one-off leaderboard, our infrastructure serves as a continuous time series infrastructure designed to explore vital, long-term scientific questions: Can model rankings be maintained over the long term? Which models remain genuinely robust under distribution shifts? Extensive streaming evaluations across 11 domains with 17 datasets demonstrate that static rankings undergo a dramatic reshuffling under a live protocol.

3D Gaussian Splatting has made Gaussian primitives a highly efficient representation for real-time novel view synthesis, but its rasterisation-based formulation relies on screen-space approximations that limit accurate view-dependent ordering and the integration of secondary ray effects such as reflections, refractions, and shadows. Gaussian ray tracing addresses these limitations by evaluating explicit ray-primitive intersections, yet it remains costly to train. We observe that the main bottleneck is not ray traversal alone, but the pixel-centric backward propagation, where many threads concurrently accumulate gradients into the same primitive parameters, causing severe atomic contention and thread serialisation. We present 3DGART, a practical training framework for ray-traced Gaussian rendering. Our key idea is to reorganise backward propagation around primitives rather than pixels. Using conservative perspective-correct screen-space bounds, we build a compact intermediate buffer and a tile-primitive mapping that allows each thread to accumulate the contribution of one primitive over its covered pixels within a tile. This transforms gradient computation from a contention-heavy scatter operation into a structured gather-like process. On Mip-NeRF 360, 3DGART achieves an $\approx 3-3.5\times$ raw training speedup over per-pixel baseline and $\approx4 \times$ over 3DGRT on Mip-NeRF 360 while improving quality. More importantly, 3DGART makes fully ray-traced Gaussian training practical, reaching runtimes competitive with rasterisation-based pipelines while preserving benefits of ray tracing.

Identifying individuals in historical photographs is a critical task across fields such as history, journalism, genealogy, and archival research. While AI-based facial recognition can efficiently generate candidate matches, it often produces ambiguous results that require deeper analysis and contextual interpretation. Existing platforms lack robust support for collaborative deliberation, especially in uncertain or high-stakes cases. We present SleuthTalk, a private collaborative workspace integrated into Civil War Photo Sleuth, designed to scaffold structured comparison, discussion, and group decision-making. SleuthTalk enables users to curate custom shortlists, annotate facial features, and build consensus through structured feedback. In a mixed-methods evaluation with experienced historical photo researchers, SleuthTalk enhanced self-reported confidence, surfaced diverse perspectives, and supported transparent, reflective identifications.

We study the trade-off between firm-side fairness and coalition stability in many-to-one matching markets with transferable payments. For a fixed matching $X$, we characterize the largest supportable core factor by a bottleneck financing problem: $α(X)=1/Φ(X)$, where $Φ(X)=\min_{z\ge0}\max_i R_i(X,z)$. This yields a polynomial-time linear program and local sensitivity formulas for one-worker reallocations. We then develop a maximum-edge round algorithm and a broader class of mutual-top safe choices. Every safe execution is EF1 and, with $t=δ(A)$ denoting the minimum positive-edge quality, guarantees $α(X)\ge\max\{t,1/[m-(m-1)t]\}$ and $SW(X)/OPT\geq t+(1-t)/m$. These bounds give finite-firm lower and upper bounds for the EF1--core minimax frontier, with exact results for two firms and for three firms when $δ\le1/2$; as the number of firms grows, the tight scale-free stability rate is $δ$. We also extend the financing formulation to stronger $EFX^+$ fairness and capacity-constrained markets.

Let $Q$ be a nonzero $s\times t$ $(0,1)$-pattern, and let $m\ge s$ and $n\ge t$. An $m\times n$ matrix is strongly $Q$-forcing if every $1$-entry belongs to an $s\times t$ submatrix equal to $Q$. Let $F^{*}(m,n,Q)$ count these matrices. Put $H=m-s+1$ and $W=n-t+1$. We prove \[ F^{*}(m,n,Q)\ge 2^{HW}. \] Writing $r$ and $c$ for the numbers of nonzero rows and columns of $Q$, equality holds if and only if \[ (H=1\text{ or }r=1)\qquad\text{and}\qquad(W=1\text{ or }c=1). \] Thus the minimum over all nonzero $s\times t$ patterns is $2^{HW}$, attained exactly by singleton patterns when $H,W>1$, and every fixed nonzero pattern has square growth rate $1$. We also refine the count by weight. If $o(Q)$ is the number of $1$-entries of $Q$, then the number of strongly $Q$-forcing matrices at the minimum positive weight $o(Q)$ is $\binom{H+r-1}{r}\binom{W+c-1}{c}$; at every fixed density in $(0,1)$, the logarithmic growth rate is the binary entropy when $m$ and $n$ are comparable. For ordinary forcing, where every $s\times t$ submatrix contains the $1$-entries of $Q$ in their prescribed positions, let $F(m,n,Q)$ be the number of forcing matrices and let $\mathfrak m(m,n,Q)$ be their minimum weight. We prove \[ F(m,n,Q)=2^{mn-\mathfrak m(m,n,Q)} \quad\text{and}\quad 2^{\,mn-\mathfrak m(m,n,Q)+HW} \le F(m,n,Q)F^{*}(m,n,Q) \le 2^{mn}. \] The lower product bound has the same equality cases as the strong-forcing lower bound above, while the upper product bound is attained exactly by singleton patterns. In particular, the product is at least $2$, with equality exactly when $s=m$, $t=n$, and $Q$ is the all-ones pattern.

Existing research on irregular time-series forecasting has primarily focused on model design, while evaluation metrics remain insufficiently studied. Existing benchmarks typically use mean squared error (MSE) as the evaluation metric. We show that, in irregular forecasting, MSE is determined not only by the model prediction but also by the sample-specific timestamp sampling distributions, leading to a biased assessment of the models' continuous-time predictive performance. To address this issue, we propose the Continuous-time Squared Error (CSE), which employs importance weighting to eliminate the influence of the timestamp sampling distributions. We further theoretically prove that CSE's asymptotic estimation error with respect to continuous-time risk is no greater than that of MSE. Finally, we construct a systematic benchmark covering synthetic, semi-synthetic, and eight real-world datasets to validate the effectiveness of CSE and systematically evaluate models' continuous-time predictive performance. Experiments show that CSE can recover continuous-time risk more accurately than MSE, while relying solely on MSE may not fully reflect models' continuous-time predictive performance in real-world scenarios. Our code can be obtained at https://github.com/hnu-vis/ITS-Bench.

Accurate 3D tooth segmentation forms the cornerstone of digital dentistry, yet it remains a formidable challenge due to the inherent intricacy of real-world dentitions, such as crowding, misaligned teeth and high morphological similarity between adjacent teeth. To resolve this, we present B-Spline Embedded Structure Learning, a novel framework that distills the inherent sequential arrangement of teeth into a continuous structural constraint to regularize representation space. Our approach parameterizes the global dental topology by fitting a parametric B-spline trajectory to tooth centers, assigning each point a continuous structural embedding that forces the shared backbone to capture global arch organization. To fully exploit these embedded priors, we introduce a Structure-Aware Dynamic Classifier (SADC) to substitute rigid static templates with adaptive, case-calibrated decision boundaries. SADC regularizes dynamic prototype pooling via a localized Gaussian proximity gate and contextually co-evolves them through an attention block modeling spatial relations and bilateral symmetries across teeth. Extensive evaluations on the 3DTeethSeg22 benchmark demonstrate that our method establishes a new state-of-the-art accuracy with exceptional structural robustness and efficiency in computational overhead, markedly enhancing the model's capacity to handle complex dental configurations.

Group-relative policy optimization has emerged as a key paradigm for training agentic large language models (LLMs) on multi-turn interactive tasks. However, most existing variants fail to distinguish advantages among successful trajectories even when these trajectories differ substantially in their interaction efficiency. For instance, circuitous successes are often assigned the identical outcome reward, causing advantage collapse and severe performance bottlenecks. To this end, we propose Group Planning-aware Policy Optimization (PlanPO), a simple yet effective RL method for learning generalizable planning abilities beyond task-specific high-quality behavior patterns. Specifically, PlanPO introduces coarse-to-fine advantage signals, which capture the relative differences in trajectory-level lengths and turn-level response lengths conditioned on successful trajectories sampled for the same task. Within the group-relative optimization structure, this enables agents to actively learn generalizable and deliberate behaviors spanning interaction planning and textual generation from high-quality rollouts, without degenerating into vanilla length minimization. Experimentally, PlanPO improves over GRPO by 27.2\% on average across the challenging multi-turn benchmarks ALFWorld, WebShop, and SciWorld, outperforming recent powerful baselines while incurring negligible additional training cost.

GPT attention measures token compatibility through dot-product similarity. This mechanism is simple, effective, and memory-efficient. But it does not explicitly model whether strong token features should reinforce or suppress one another. We introduce Q-Interference, a fully classical quantum-inspired attention mechanism for autoregressive language modeling that augments each query and key feature with an amplitude and a learned phase. The resulting attention score is phase-aware which aligned phases contribute constructively while conflicting phases contribute destructively. Although Q-Interference yields a richer interaction rule than similarity alone, a naive implementation of Q-Interference requires a large token-pair-feature interaction tensor, making it memory-intensive and often impractical. To address this limitation, we propose an exact trigonometric factorization that computes the same score using two standard matrix multiplications avoiding materialization of the large intermediate tensor. Q-Interference fits directly into a Transformer block in GPT and leaves the remainder of the model architecture and next-token prediction objective unchanged. Experiments on public benchmark datasets and baseline models show that the proposed reformulation trains stably in a controlled GPT-style setting and provides a consistent memory advantage over naive phase-aware interference attention. These results support the specific contribution of this work: an exact memory-efficient reformulation that makes phase-aware interference attention practical within a standard GPT pipeline.

Compute-optimal scaling laws guide the training of frontier language models yet remain largely unexplored for visual generation. We present a systematic scaling law study for text-to-image diffusion models using Abra, a controlled family of flow-matching transformers trained across three orders of magnitude worth of compute ($10^{19}$ to $10^{22}$ FLOPs), reaching significantly larger compute budgets than previous works. We demonstrate that diffusion models scale just as predictably as language models but require far more data to train optimally: compute optimality occurs at approximately $200$ image tokens per parameter, ten times the Chinchilla compute-optimal prescription for LLMs. We show that unlike language models, diffusion models are robust to overtraining and that practitioners should err on the side of more data rather than a larger model. Finally, we show that this predictability extends beyond training loss to generative quality metrics, optimal CFG settings, representation quality, and even the shape of the training curves, which collapse onto a universal form.

Selecting a small, diverse subset from a large candidate pool often means balancing several incompatible notions of diversity. In trademark curation, for instance, a subset should cover both the language used to describe marks and the visual space of their logos. A single determinantal point process (\DPP) kernel can hide failure in one view, and averaging kernels replaces the multi-view relaxation by an ordinary single-kernel spectral problem. We formulate \emph{fair multi-view determinant selection}: maximize the weakest per-view log determinant of a size-$k$ subset. We smooth this nonsmooth objective and relax it to the Stiefel manifold. The relaxation embeds every discrete subset exactly, but unlike its single-view counterpart it has no closed-form spectral solution in general. Its stationarity condition is a gauge-invariant nonlinear eigenvalue problem with eigenvector-dependent, view-adaptive weights. We derive an adaptive self-consistent-field (\SCF) solver with damping and level shifting, and round the resulting subspace by leverage-score screening followed by fair local refinement. The solver needs only feature-map products for each view. We report conflicting-view synthetic experiments and specify a multimodal USPTO protocol; the real-data multimodal results require aligned logo embeddings and are not claimed in this version.

Reproducible and verifiable builds increase trust in distributed software artifacts by enabling independent parties to detect artifacts produced by compromised build or release pipelines. However, artifact verification requires more than deterministic builds: a verifier must also recover the source state, build environment, dependencies, and build instructions that produced the artifact. Decentralized-build ecosystems make this difficult because artifacts are produced through heterogeneous tools, maintainer-controlled workflows, and fragmented metadata. As a result, it remains unclear how often artifacts in these ecosystems can be independently verified. This paper studies artifact verifiability across four popular decentralized-build package ecosystems. We define an independent verifier model that relies only on registry-derivable metadata and an artifact comparison model with tiered equivalence levels. We implement these models in an Artifact Verification Pipeline and use it to measure artifact verifiability across the target ecosystems. Our results show that, beyond build determinism, verifiability is limited by missing source and build metadata, implicit release transformations, and unconventional build practices. Provenance attestations and embedded VCS metadata improve verification, but they do not provide complete rebuild specifications. These findings identify concrete metadata gaps and ecosystem-level changes needed to make artifact verification practical at package-registry scale.

Irregular time series forecasting is crucial in many domains, such as healthcare and meteorological observation. However, due to the inherent characteristics of irregular time series, including sparse observations and non-uniform sampling, accurately predicting future dynamics remains challenging. In light of these two characteristics, many existing methods aggregate irregular observations into fixed-dimensional estimated response coefficients through predefined basis functions and use these coefficients as sequence representations. Nevertheless, this modeling paradigm still suffers from two key limitations: (i) a potential non-vanishing asymptotic bias caused by ignoring the sampling density of timestamps; and (ii) the limited adaptability of predefined basis functions to diverse temporal patterns. In this study, we propose a Debiased Neural Basis-Function Network (DNBNet) to address these challenges. Its core is a debiased neural basis-function response mechanism, which corrects asymptotic bias through importance sampling while parameterizing basis functions with neural networks to adapt to diverse temporal patterns. In addition, considering the sparsity of irregular data, we design a novel multi-scale decomposition module based on average pooling, together with a mass-aware fusion mechanism, to obtain richer representations. Finally, a dual-branch decoder is employed for forecasting. Extensive experiments on multiple real-world datasets demonstrate the effectiveness of DNBNet and its strong generalizability across diverse irregular time series scenarios. Our code can be obtained at https://github.com/hnu-vis/DNBNet.

Reconstructing dynamic 4D scenes requires jointly estimating correspondence, geometry, object motion, and camera motion. Existing feed-forward methods typically predict dense task-specific maps or independently process source-target pairs, leading to unnecessary computation for sparse queries and limited feature reuse across different frame pairs. We present UniQuery4R, a query-conditioned framework that encodes a multi-frame clip once and selects the source view, target view, and continuous source-image coordinate only at decoding time via source-to-target cross-attention. Each query jointly predicts target correspondence, target-time 3D position, and scene flow, along with source depth, while camera parameters are estimated per view. This design allows the encoded clip to be reused across arbitrary source-target selections and supports both sparse inference and dense reconstruction through batched queries, without learned temporal embeddings tied to a fixed clip length. We further introduce a direction-magnitude parameterization of scene flow with separate supervision for moving and static points. Among the evaluated methods, UniQuery4R achieves the best macro-average results on WorldTrack for both scene-flow estimation and dynamic-point reconstruction.

Existing agentic reasoning systems typically rely on centralized protocols. This design introduces routing bottlenecks and static role allocations that often fail when handling complex multimodal queries. We propose DeAR (Decentralized Agentic Reasoning), a framework that shifts from central control to autonomous peer-to-peer collaboration. DeAR is built on three mechanisms: (1) decentralized capability grounding for query-dependent agent specialization, (2) thought map navigation for targeted peer interactions, and (3) topology update for adaptive error correction. Evaluations across 9 diverse multimodal reasoning and text-based QA benchmarks indicate that DeAR consistently outperforms recent baseline methods, validating that decentralized and adaptive collaboration among agents enhances accuracy in knowledge-intensive reasoning tasks. The source code will be available at https://open_upon_acceptance.

This report presents a two-stage, training-free solution for the MeViS-Text track of the 8th LSVOS Challenge. The task requires a model to localize and segment the object specified by a natural-language expression throughout a video. Such expressions often depend on temporal cues, including actions, interactions, directions, and relative positions. Our first stage uses Gemini-3.1 Pro via API to decompose a video-level event into instance-level targets, select a key frame for each target, and generate a discriminative description aligned with that frame. In the second stage, SAM3-agent produces a pixel-level seed mask on the selected frame, and the SAM3 video tracker propagates the mask bidirectionally through the video. Valid instances are grounded and propagated independently before their frame-wise masks are merged. All local SAM3 processing runs on a single NVIDIA GeForce RTX 4090 without task-specific training or model ensembling. Our method ranked third on the challenge test set, obtaining J&F, J, F, N-acc., T-acc., and Final scores of 0.761, 0.7367, 0.7852, 0.8333, 0.9755, and 0.856593, respectively.