arXiv subjects

Yi Du

Publications and source records attributed to Yi Du.

At least 19 recordsLinked to original sources

arXiv2026

Family Seiberg-Witten equation on Kahler surface and $π_i(\Symp)$ on multiple-point blow ups of Calabi-Yau surfaces

Let $ω$ be a Kahler form on $M$, which is a torus $T^4$, a $K3$ surface or an Enriques surface, let $M\#n\overline{\mathbb{CP}^2}$ be $n-$point Kahler blowup of $M$. Suppose that $κ=[ω]$ satisfies certain irrationality condition. Applying techniques related to deformation of complex objects, we extend the guage-theoretic invariant on closed Kahler suraces developed by Kronheimer\cite{Kronheimer1998} and Smirnov\cite{Smirnov2022}\cite{Smirnov2023}. As a result, we show that even dimensional higher homotopy groups of $\Symp(M\#n\overline{\mathbb{CP}^2},ω)$ are infinitely generated.

math.GT↗

arXiv2026

L2-contraction and Asymptotic Stability of Monotone Shock Waves for Cubic and Quartic KdV-Burgers Equations

We investigate the nonlinear stability of monotone traveling shock waves arising in the generalized Korteweg-de Vries-Burgers equation with a power-law flux, focusing on the cubic and quartic cases p=3 and p=4, respectively, in the presence of both dissipative and dispersive effects. Since shocks connecting distinct end states destroy the integrable structure of the equation, we introduce a slowly varying reference connection to establish global well-posedness for the Cauchy problem in an affine Sobolev space. A key ingredient in the stability analysis is to establish a new two-sided comparison estimate for the derivative of a monotone shock-wave profile. After a monotone coordinate transformation, this estimate allows the viscous dissipation to be compared with the natural vanishing weight arising in a shifted Poincare inequality. Together with an $L^\infty$ bootstrap controlling the higher-order relative-flux terms, this yields a global $L^2$-contraction property modulo a time-dependent shift for sufficiently small $H^1$ perturbations in the admissible parameter regimes. We obtain exponential decay of the shock profile and show asymptotic convergence to the shifted shock in $L^q(\R)$, $2<q\leq \infty$, with vanishing shift velocity.

math.AP↗

arXiv2026

ProSearch: Benchmarking Multi-Constraint Protocol Retrieval in Experimental Science

In AI for Science scenarios, experimental design and method reproduction require retrieving protocols that satisfy multiple interdependent constraints, yet existing benchmarks lack systematic evaluation of this capability. To address this gap, we introduce ProSearch, a benchmark for multi-constraint protocol retrieval in experimental science. ProSearch has three key features: (1) realism, with queries grounded in real experimental records and expert-defined templates; (2) complexity, with 60 experimental fields and their dependencies; and (3) diagnosticity, with 3-5 explicit constraints per query, including numerical and negation constraints. We construct ProSearch through LLM-based extraction, data-driven constraint selection, controlled query generation, and evidence-based relevance validation. The resulting benchmark contains 1,515 queries across six experimental science domains. We conduct extensive experiments on ProSearch with 19 representative retrievers and complementary analyses. The best-performing retriever achieves an average nDCG@10 of only 0.290, which improves to 0.474 with reranking. Structured diagnostic analysis shows that explicit constraint matching provides a strong ranking signal, while manual error analysis identifies joint constraint binding as a key bottleneck. ProSearch provides a testbed for systematically evaluating scientific retrieval systems.

cs.IR↗

arXiv2026

DRIVE: Fine-tuning via Data Contribution- and Diversity-aware Weighting with Prior Regularization

Single-stage fine-tuning is pervasive, yet catastrophic forgetting persists when the source pre-training data is inaccessible. Data weighting affords a natural lever to modulate sample influence. Yet existing weighting methods lack a principled mechanism to explicitly arbitrate knowledge retention against new-task adaptation, and provide no unified differentiable objective to accommodate auxiliary signals such as sample diversity and sample characteristics. This is because current schemes privilege one goal over the other, and no explicit gradient-compatible encoding exists for such signals. To address this, we propose DRIVE, a principled and efficient weight assignment framework. It derives per-sample weights via a unified differentiable objective where three coupled signals co-evolve: Shapley additivity splits contribution into anti-forgetting and adaptation components, which are reshaped by a covariance-based diversity term and stabilized by a sample-characteristic prior. Far from a mere combination, this yields an end-to-end gradient-based scheme balancing stability against plasticity. Theoretically, we prove that DRIVE enjoys superior stability-plasticity guarantees over existing weighting schemes and enhances parameter estimation robustness. Empirically, it attains superior overall performance, faster convergence, and stronger robustness. Code at https://github.com/luciabcd/DRIVE.

cs.LG↗

arXiv2026

NADS: Navigator-Guided Data Selection for Mitigating Catastrophic Forgetting in Fine-Tuning

Catastrophic forgetting is a central obstacle in single-stage fine-tuning of large language models (LLMs) on a new task, especially when the original pretraining data are unavailable. Synthetic data offers a practical mitigation path, but candidate pools are often noisy and redundant, making effective selection essential. Existing methods often rely on indirect proxy signals whose alignment with actual training benefit can be limited, while utility--diversity selection over large candidate pools remains computationally demanding. To address these challenges, we propose NAvigator-guided Data Selection (NADS), a fine-tuning framework for LLMs. NADS first follows the target fine-tuning recipe without preservation regularization to obtain a navigator model, whose deviation from the pretrained model exposes recipe-induced capability drift. It then uses the predictive divergence between the navigator and the pretrained model to score forgetting-aware utility for each synthetic candidate, and constructs a constraint set through an efficient utility--diversity selection strategy. Finally, NADS fine-tunes on the new-task data while distilling pretrained behavior on the selected constraints to preserve general capabilities. Experiments show that NADS delivers a stronger balance between new-task performance and general capability preservation, while reducing utility--diversity selection overhead. Code at: https://github.com/zjh-vinky/NADS.

cs.LG↗

arXiv2026

Point-Level Refinable Semantic Mapping for Spatial Understanding

Open-vocabulary 3D semantic maps have emerged as a key spatial memory representation for long-horizon robot autonomy, enabling robots to reason about objects, their spatial relationships, and past observations across time. However, existing object-centric semantic maps are primarily designed for static environments and rely on an append-only point integration strategy, where newly observed points are accumulated but previously integrated points are rarely reevaluated. Consequently, perception errors accumulate over time, and outdated points persist after objects move or disappear. Although recent dynamic mapping methods update object-level representations in changing scenes, they do not explicitly revise noisy or obsolete points within each object. We propose an online refinable semantic mapping framework that continuously refines object representations through probabilistic point-level confidence updating. Each object point maintains a confidence score that is updated online using multi-view temporal evidence: consistent observations reinforce reliable points, while inconsistent segmentation masks, depth measurements, and visibility cues suppress spurious and outdated points. Combined with object tracking and re-identification to preserve instance identities under motion and relocation, our framework enables object-centric semantic maps to be continuously refined rather than monotonically accumulated. Experiments demonstrate improved mapping accuracy over the previous method by 3.46% on Replica and 2.55% on ScanNet, show robustness in dynamic environments and under perception noise, and provide an effective spatial memory for object navigation and embodied question-answering with success rates of 82.4% and 76.7%.

cs.RO↗

arXiv2026

AnyNav: Visual Neuro-Symbolic Friction Learning for Off-road Navigation

Off-road navigation is critical for a wide range of field robotics applications from planetary exploration to disaster response. However, it remains a longstanding challenge due to unstructured environments and the inherently complex terrain-vehicle interactions. Traditional physics-based methods struggle to accurately capture the nonlinear dynamics underlying these interactions, while purely data-driven approaches often overfit to specific motion patterns, vehicle geometries, or platforms, limiting their generalization in diverse, real-world scenarios. To address these limitations, we introduce AnyNav, a vision-based friction estimation and navigation framework grounded in neuro-symbolic principles. Our approach integrates neural networks for visual perception with symbolic physical models for reasoning about terrain-vehicle dynamics. To enable self-supervised learning in real-world settings, we adopt the imperative learning paradigm, employing bilevel optimization to train the friction network through physics-based optimization. This explicit incorporation of physical reasoning substantially enhances generalization across terrains, vehicle types, and operational conditions. Leveraging the predicted friction coefficients, we further develop a physics-informed navigation system capable of generating physically feasible, time-efficient paths together with corresponding speed profiles. We demonstrate that AnyNav seamlessly transfers from simulation to real-world robotic platforms, exhibiting strong robustness across different four-wheeled vehicles and diverse off-road environments.

cs.RO↗

arXiv2026

HierDTA: Hierarchical Structure-Aware Representation Learning for Drug--Target Affinity Prediction

Drug--target affinity (DTA) prediction plays an important role in computer-aided drug discovery, where reliable computational models can help prioritize compound--protein pairs before experimental validation. A key challenge in DTA prediction is to effectively mine multi-level molecular information from drug structures while accurately modeling the binding-site environment of target proteins. To address these issues, we introduce HierDTA, a hierarchical structure-aware framework for DTA prediction. \textcolor{black}{HierDTA represents each drug as an atom--motif heterogeneous graph and introduces an augmentation module that propagates functional-motif semantics to the corresponding atoms. For proteins, kinase-domain structures are converted into residue interaction graphs. The resulting atom and residue representations are integrated using cross-attention to capture pair-specific relationships, while molecule fingerprints and pretrained ESM-2 embeddings provide complementary global information. Compared with state-of-the-art methods, HierDTA achieves superior performance on the Davis dataset across all metrics and attains the highest $r_m^2$ of 0.791 on KIBA. It also exhibits strong generalization under cold-start settings, particularly on KIBA, where MSE is reduced by 21.7\% in the unseen-protein scenario. Ablation study and interpretability analysis supported the contributions of the hierarchical augmentation and interaction components.} Case studies further showed that HierDTA provided early enrichment of candidate bioactive compounds. These results indicate that HierDTA is an effective framework for affinity prediction and target-specific virtual screening.

cs.LG↗

arXiv2026

HAVANA: A Hierarchical Video-Action Model for Vision-Language Navigation

Vision-Language Navigation (VLN) requires long-horizon instruction-consistent planning and short-horizon local planning in current geometry. Video-action models (VAMs) offer a promising way to learn predictive representations for navigation. However, existing VAMs jointly predict visual futures and local trajectories in a single process, even though they operate at different paces: future prediction remains stable across multiple updates while local trajectories must react to fresh observations. To bridge this gap, we introduce HAVANA, a dual-system VAM that decouples predictive reasoning from reactive local planning. A slow system produces a compact language-conditioned latent plan from observation history, while a fast system combines the latest plan with the current observation to generate local trajectories. Their asynchronous execution enables rapid adaptation without repeated future prediction, and no future video is generated during inference. Experiments on OmniNavBench demonstrate state-of-the-art navigation performance over prior baselines, while real-world deployments further show robust generalization to unseen environments across different robot embodiments. Our code and checkpoints will be available at https://github.com/Anastasiawd/HAVANA.

cs.CV↗

arXiv2026

Token Utility Is Selection-Conditioned: Coupled Selection of Prompt Context and Response Supervision for Instruction Tuning

Token selection for instruction tuning is coupled: supervision utility depends on retained context, and vice versa. Existing methods largely optimize prompt compression or response supervision separately. Combining selectors can value supervision using removed context, or context using excluded supervision, creating selection-state mismatch. BRIDGE (Budgeted Response-Prompt Interaction via Directional Gradient-guided Efficient Token Selection) models selection-conditioned token utility through a shared validation-directed surrogate objective. BRIDGE routes response utility to supporting prompt tokens through attention, then alternates budgeted selection to revalue each side under the other's retained subset. Updates reuse precomputed interactions without repeated model scoring, while structure-aware projection promotes coherent supervision spans. Across three model families and three task domains, BRIDGE performs best overall among evaluated selection methods. Its MATH advantage over independent selection grows with compression, reaching 1.66 percentage points on Llama-3.2-3B.

cs.LG↗

arXiv2026

One Distribution, Two Channels: Coverage-Calibrated KV Retention and Compensation

Long-context decoding is increasingly constrained by key--value (KV) cache memory and bandwidth. Existing fixed-budget compression methods typically separate retention from compensation, while a retention ranking specifies neither discarded attention mass nor the direction of induced output error. We start from an exact factorization: eviction error equals evicted attention mass times the directional gap between the evicted centroid and retained output, highlighting the importance of set-level coverage in retention and mass-preserving memory writing. We introduce CORE COverage Calibration and Evicted-Mass REdistribution for KV Cache, which distills an offline allocation combining query utility and log-determinant coverage into a lightweight cache-aware indexer. At inference, one calibrated distribution drives both channels: its Top-$B$ ordering retains complementary KV states, while its excluded allocation mass and conditional weights parameterize latent-memory writes without a separate write-weight predictor or online log-determinant evaluation. Our analysis provides a four-term pre-compensation error certificate, characterizes non-additive coverage interactions, and establishes mass-independent write stability with hierarchical bounds through recurrent and query-adaptive normalization. Across three backbones, CORE exceeds the strongest RULER baseline by up to 3.78 points at 90% compression; LongBench and repeated-eviction evaluations further demonstrate strong effectiveness and decoding efficiency.

cs.AR↗

arXiv2026

UniPose9D: Universal Category-Agnostic Object Pose Estimation

Object pose estimation is a fundamental problem in 3D vision. Although recent state-of-the-art approaches achieve strong performance, generalization to novel categories and unseen scenes remains challenging. We propose UniPose9D, a unified model for category-agnostic 9D object pose estimation: given an instance mask/ROI and either an RGB-D observation or an RGB image with predicted depth, the model estimates rotation, translation, and metric size without category labels, CAD models, mean-shape priors, or reference views. Specifically, UniPose9D samples point pairs from the observed object geometry and uses DINOv2 and PointNet features to predict NOCS coordinates for each pair. To improve accuracy, we introduce a point-pair-based RANSAC N-hop Kabsch-Umeyama algorithm with an adaptive threshold. We further employ flow matching to address symmetric ambiguities and construct a large-scale training set by curating and aligning pose annotations from existing public datasets. Experiments across eight datasets show that a single unified model achieves competitive performance on standard benchmarks while generalizing to unseen objects, unseen categories, and in-the-wild scenarios. Our code and model are available at https://github.com/qq456cvb/UniPose9D.

cs.CV↗

arXiv2026

OCH3R: Object-Centric Holistic 3D Reconstruction

Object-centric scene understanding is a fundamental challenge in computer vision. Existing approaches often rely on multi-stage pipelines that first apply pre-trained segmentors to extract individual objects, followed by per-object 3D reconstruction. Such methods are computationally expensive, fragile to segmentation errors, and scale poorly with scene complexity. We introduce OCH3R, a unified framework for Object-Centric Holistic 3D Reconstruction from a single RGB image. OCH3R performs one forward pass to simultaneously predict all object instances with their 6D poses and detailed 3D reconstructions. The key idea is a transformer architecture that predicts per-pixel attributes, including CLIP-based category embeddings, metric depth, normalized object coordinates (NOCS), and a fixed number of 3D Gaussians representing each object. To supervise these Gaussian reconstructions, we transform them into canonical space using the predicted 6D poses and align them with pre-rendered canonical ground truth, avoiding costly per-image Gaussian label generation. On standard indoor benchmarks, OCH3R achieves state-of-the-art performance across monocular depth estimation, open-vocabulary semantic segmentation, and RGB-only category-level 6D pose estimation, while producing high-fidelity, editable per-object reconstructions. Crucially, inference is fully feed-forward and scales independently of the number of objects, offering orders-of-magnitude speedups over conventional multi-stage pipelines in cluttered scenes.

cs.CV↗

arXiv2026

When a Relaxed PEP Is Exact: The Sharp Queried-Gradient Rate of Nesterov's Fast Gradient Method

We determine the exact worst-case value, at every horizon $N\geq7$, of the smallest queried gradient norm generated by Nesterov's fast gradient method on smooth convex functions. Let $t_0=1$ and $t_{k+1}=(1+\sqrt{1+4t_k^2})/2$, and let $x_0,\ldots,x_N$ denote the points at which the method evaluates gradients. For every such $N$ and every dimension $d\geq N-4$, we prove \[ \sup_{\substack{f\in\F_{0,L}(\R^d),\ x_\star\in\arg\min f \norm{x_0-x_\star}\leq R}} \min_{0\leq k\leq N}\norm{\nabla f(x_k)}^2 =\frac{L^2R^2}{\sum_{k=0}^N t_k^2}. \] The relaxed-PEP upper bound is due to Kim and Fessler, who also reported tight numerical solutions of the exact-interpolation PEP at selected horizons. What remained missing was an analytic matching family valid uniformly over the horizon. For every $N\geq7$, we construct such a family using an FGM-specific spherical polytope $K_N$ and the standard projection-envelope function \[ f_N(x)=\max_{g\in K_N}\left\{\ip{x}{g}-\frac12\norm{g}^2\right\}, \qquad \nabla f_N(x)=\Proj_{K_N}(x). \] Every queried gradient has the same norm, and the vertices of $K_N$ are generated from a three-dimensional seed by a one-dimensional spherical cone lift. The lift preserves all projection inequalities and raises the adversary dimension by one at each horizon. The projection/Moreau-envelope template itself is classical; the new ingredients are the FGM-specific algebraic seed, the proof that it attains the relaxed bound, and the common-latitude lift that propagates this exactness to every $N\geq7$. We state precise hypotheses for that propagation and do not claim that every rank-one relaxed PEP admits such a seed.

math.OC↗

arXiv2026

Comprehend, Divide, and Conquer: Feature Subspace Exploration via Multi-Agent Hierarchical Reinforcement Learning

Feature selection aims to preprocess the target dataset, find an optimal and most streamlined feature subset, and enhance the downstream machine learning task. Among filter, wrapper, and embedded-based approaches, the reinforcement learning (RL)-based subspace exploration strategy provides a novel objective optimization-directed perspective and promising performance. Nevertheless, even with improved performance, current reinforcement learning approaches face challenges similar to conventional methods when dealing with complex datasets. These challenges stem from the inefficient paradigm of using one agent per feature and the inherent complexities present in the datasets. This observation motivates us to investigate and address the above issue and propose a novel approach, namely HRLFS. Our methodology initially employs a Large Language Model (LLM)-based hybrid state extractor to capture each feature's mathematical and semantic characteristics. Based on this information, features are clustered, facilitating the construction of hierarchical agents for each cluster and sub-cluster. Extensive experiments demonstrate the efficiency, scalability, and robustness of our approach. Compared to contemporary or the one-feature-one-agent RL-based approaches, HRLFS improves the downstream ML performance with iterative feature subspace exploration while accelerating total run time by reducing the number of agents involved.

cs.AI↗

arXiv2026

VL-Nav: Neuro-Symbolic Reasoning-based Vision-Language Navigation

Navigating unseen, large-scale environments based on complex and abstract human instructions remains a formidable challenge for autonomous mobile robots. Addressing this requires robots to infer implicit semantics and efficiently explore large-scale task spaces. However, existing methods, ranging from end-to-end learning to foundation model-based modular architectures, often lack the capability to decompose complex tasks or employ efficient exploration strategies, leading to robot aimless wandering or target recognition failures. To address these limitations, we propose VL-Nav, a neuro-symbolic (NeSy) vision-language navigation system. The proposed system intertwines neural reasoning with symbolic guidance through two core components: (1) a NeSy task planner that leverages a symbolic 3D scene graph and image memory system to enhance the vision language models' (VLMs) neural reasoning capabilities for task decomposition and replanning; and (2) a NeSy exploration system that couples neural semantic cues with the symbolic heuristic function to efficiently gather the task-related information while minimizing unnecessary repeat travel during exploration. Validated on the DARPA TIAMAT Challenge navigation tasks, our system achieved an 83.4% success rate (SR) in indoor environments and 75% in outdoor scenarios. VL-Nav achieved an 86.3% SR in real-world experiments, including a challenging 483-meter run. Finally, we validate the system with complex instructions in a 3D multi-floor scenario.

cs.RO↗

arXiv2026

Vision-Language Memory for Spatial Reasoning

Spatial reasoning is a critical capability for intelligent robots, yet current vision-language models (VLMs) still fall short of human-level performance in video-based spatial reasoning. This gap mainly stems from two challenges: a semantic-geometric misalignment that prevents consistent 3D understanding, and the absence of persistent memory to retain 3D representation and understanding across frames. To address these limitations, we present VLM$^2$, a Vision-Language Model with persistent Memory for spatial reasoning with a view-consistent, 3D-aware representation purely from 2D videos. Specifically, we incorporate a dual-memory module consisting of a working memory that operates as a sliding window to focus on immediate context, and an episodic memory that consolidates and stores critical information across frames. This design enables bounded and efficient spatial reasoning under a fixed computational cost. Extensive experiments on multiple benchmarks show that VLM$^2$ achieves state-of-the-art performance among video-based models, significantly advancing the frontier of visual-spatial intelligence.

cs.CV↗

arXiv2026

Two-Dimensional Altermagnetism in Epitaxial CrSb Ultrathin Films

Altermagnets constitute an emerging class of collinear magnets that exhibit zero net magnetization yet host spin-split electronic bands arising from non-relativistic spin-space-group symmetries. Realization of altermagnetism in the two-dimensional (2D) limit remains an outstanding challenge because dimensional reduction suppresses kZ dispersion and destabilizes the symmetry operations essential for spin compensation. Here, we investigate ultrathin CrSb films grown epitaxially on Bi2Te3 substrate and uncover the evolution of altermagnetism in the 2D limit. Scanning tunneling microscopy (STM), quasiparticle interference (QPI), angle-resolved photoemission spectroscopy (ARPES), and density functional theory (DFT) calculations show that interfacial symmetry breaking in the one-unit-cell (1 UC) limit gives rise to localized electronic states and uncompensated magnetic moments. These interfacial effects become weakened from 7/4 UC, accompanied by the recovery of a bulk-like coordination environment and the emergence of altermagnetic electronic characteristics. Our results show that the essential altermagnetic electronic structure of CrSb survives at a thickness of only ~1.05 nm, demonstrating the robustness of altermagnetism in the 2D limit and opening opportunities for integrating stray-field-free spin order into low dimensional spintronic architectures.

cond-mat.mtrl-sci↗