arXiv subjects

Yi Zhou

Publications and source records attributed to Yi Zhou.

At least 127 records · Page 7Linked to original sources

arXiv2026

CASE -- Condition-Aware Sentence Embeddings for Conditional Semantic Textual Similarity Measurement

The meaning conveyed by a sentence often depends on the context in which it appears. Despite the progress of sentence embedding methods, it remains unclear as how to best modify a sentence embedding conditioned on its context. To address this problem, we propose Condition-Aware Sentence Embeddings (CASE), an efficient and accurate method to create an embedding for a sentence under a given condition. First, CASE creates an embedding for the condition using a Large Language Model (LLM) encoder, where the sentence influences the attention scores computed for the tokens in the condition during pooling. Next, a supervised method is learnt to align the LLM-based text embeddings with the Conditional Semantic Textual Similarity (C-STS) task. We find that subtracting the condition embedding consistently improves the C-STS performance of LLM-based text embeddings by improving the isotropy of the embedding space. Moreover, our supervised projection method significantly improves the performance of LLM-based embeddings despite requiring a small number of embedding dimensions.

cs.CL↗

arXiv2026

Adaptive Confidence Gating in Multi-Agent Collaboration for Efficient and Optimized Code Generation

While Large Language Models (LLMs) have catalyzed breakthroughs in automated code generation, Small Language Models (SLMs) often encounter reasoning bottlenecks and failure loops when addressing complex logical requirements. To overcome these challenges, we propose DebateCoder, a multi-agent collaborative framework designed to improve the reasoning ability of SLMs (e.g., Pangu-1B) in resource-constrained environments. DebateCoder uses a structured role-playing protocol with three agents: User Agent (A_UA), Technical Agent (A_TA), and Quality Assurance Agent (A_QA). It also includes an Adaptive Confidence Gating mechanism with a 95% threshold to balance accuracy and inference efficiency. In addition, we introduce a multi-turn deliberation module and a reviewer-guided analytical debugging loop for orthogonal pre-generation debate and post-generation refinement. Experiments on HumanEval and MBPP show that DebateCoder achieves 70.12% Pass@1 on HumanEval, outperforming MapCoder while reducing API overhead by about 35%. These results indicate that collaborative protocols can mitigate limitations of small-parameter models and provide a scalable, efficient approach to high-quality automated software engineering.

cs.SE↗

arXiv2026

Joint Diffusion for Universal Hand-Object Grasp Generation

Predicting and generating human hand grasp over objects is critical for animation and robotic tasks. In this work, we focus on generating both the hand and objects in a grasp by a single diffusion model. Our proposed Joint Hand-Object Diffusion (JHOD) models the hand and object in a unified latent representation. It uses the hand-object grasping data to learn to accommodate hand and object to form plausible grasps. Also, to enforce the generalizability over diverse object shapes, it leverages large-scale object datasets to learn an inclusive object latent embedding. With or without a given object as an optional condition, the diffusion model can generate grasps unconditionally or conditional to the object. Compared to the usual practice of learning object-conditioned grasp generation from only hand-object grasp data, our method benefits from more diverse object data used for training to handle grasp generation more universally. According to both qualitative and quantitative experiments, both conditional and unconditional generation of hand grasp achieves good visual plausibility and diversity. With the extra inclusiveness of object representation learned from large-scale object datasets, the proposed method generalizes well to unseen object shapes.

cs.CV↗

arXiv2026

Evaluating the Effect of Retrieval Augmentation on Social Biases

Retrieval Augmented Generation (RAG) has gained popularity as a method for conveniently incorporating novel facts that were not seen during the pre-training stage in Large Language Model (LLM)-based Natural Language Generation (NLG) systems. However, LLMs are known to encode significant levels of unfair social biases. The modulation of these biases by RAG in NLG systems is not well understood. In this paper, we systematically study the relationship between the different components of a RAG system and the social biases presented in the text generated across three languages (i.e. English, Japanese and Chinese) and four social bias types (i.e. gender, race, age and religion). Specifically, using the Bias Question Answering (BBQ) benchmark datasets, we evaluate the social biases in RAG responses from document collections with varying levels of stereotypical biases, employing multiple LLMs used as generators. We find that the biases in document collections are often amplified in the generated responses, even when the generating LLM exhibits a low-level of bias. Our findings raise concerns about the use of RAG as a technique for injecting novel facts into NLG systems and call for careful evaluation of potential social biases in RAG applications before their real-world deployment.

cs.CL↗

arXiv2026

Language Model Representations for Efficient Few-Shot Tabular Classification

The Web is a rich source of structured data in the form of tables, from product catalogs and knowledge bases to scientific datasets. However, the heterogeneity of the structure and semantics of these tables makes it challenging to build a unified method that can effectively leverage the information they contain. Meanwhile, Large language models (LLMs) are becoming an increasingly integral component of web infrastructure for tasks like semantic search. This raises a crucial question: can we leverage these already-deployed LLMs to classify structured data in web-native tables (e.g., product catalogs, knowledge base exports, scientific data portals), avoiding the need for specialized models or extensive retraining? This work investigates a lightweight paradigm, $\textbf{Ta}$ble $\textbf{R}$epresentation with $\textbf{L}$anguage Model~($\textbf{TaRL}$), for few-shot tabular classification that directly utilizes semantic embeddings of individual table rows. We first show that naive application of these embeddings underperforms compared to specialized tabular models. We then demonstrate that their potentials can be unlocked with two key techniques: removing the common component from all embeddings and calibrating the softmax temperature. We show that a simple meta-learner, trained on handcrafted features, can learn to predict an appropriate temperature. This approach achieves performance comparable to state-of-the-art models in low-data regimes ($k \leq 32$) of semantically-rich tables. Our findings demonstrate the viability of reusing existing LLM infrastructure for efficient semantics-driven pathway to reuse existing LLM infrastructure for Web table understanding.

cs.CL↗

arXiv2026

Toward Stable Semi-Supervised Remote Sensing Segmentation via Co-Guidance and Co-Fusion

Semi-supervised remote sensing (RS) image semantic segmentation offers a promising solution to alleviate the burden of exhaustive annotation, yet it fundamentally struggles with pseudo-label drift, a phenomenon where confirmation bias leads to the accumulation of errors during training. In this work, we propose Co2S, a stable semi-supervised RS segmentation framework that synergistically fuses priors from vision-language models and self-supervised models. Specifically, we construct a heterogeneous dual-student architecture comprising two distinct ViT-based vision foundation models initialized with pretrained CLIP and DINOv3 to mitigate error accumulation and pseudo-label drift. To effectively incorporate these distinct priors, an explicit-implicit semantic co-guidance mechanism is introduced that utilizes text embeddings and learnable queries to provide explicit and implicit class-level guidance, respectively, thereby jointly enhancing semantic consistency. Furthermore, a global-local feature collaborative fusion strategy is developed to effectively fuse the global contextual information captured by CLIP with the local details produced by DINOv3, enabling the model to generate highly precise segmentation results. Extensive experiments on six popular datasets demonstrate the superiority of the proposed method, which consistently achieves leading performance across various partition protocols and diverse scenarios. Project page is available at https://xavierjiezou.github.io/Co2S/.

cs.CV↗

arXiv2026

KALE-LM-Chem: Vision and Practice Toward an AI Brain for Chemistry

Recent advancements in large language models (LLMs) have demonstrated strong potential for enabling domain-specific intelligence. In this work, we present our vision for building an AI-powered chemical brain, which frames chemical intelligence around four core capabilities: information extraction, semantic parsing, knowledge-based QA, and reasoning & planning. We argue that domain knowledge and logic are essential pillars for enabling such a system to assist and accelerate scientific discovery. To initiate this effort, we introduce our first generation of large language models for chemistry: KALE-LM-Chem and KALE-LM-Chem-1.5, which have achieved outstanding performance in tasks related to the field of chemistry. We hope that our work serves as a strong starting point, helping to realize more intelligent AI and promoting the advancement of human science and technology, as well as societal development.

cs.AI↗

arXiv2026

Bidirectional Channel-selective Semantic Interaction for Semi-Supervised Medical Segmentation

Semi-supervised medical image segmentation is an effective method for addressing scenarios with limited labeled data. Existing methods mainly rely on frameworks such as mean teacher and dual-stream consistency learning. These approaches often face issues like error accumulation and model structural complexity, while also neglecting the interaction between labeled and unlabeled data streams. To overcome these challenges, we propose a Bidirectional Channel-selective Semantic Interaction~(BCSI) framework for semi-supervised medical image segmentation. First, we propose a Semantic-Spatial Perturbation~(SSP) mechanism, which disturbs the data using two strong augmentation operations and leverages unsupervised learning with pseudo-labels from weak augmentations. Additionally, we employ consistency on the predictions from the two strong augmentations to further improve model stability and robustness. Second, to reduce noise during the interaction between labeled and unlabeled data, we propose a Channel-selective Router~(CR) component, which dynamically selects the most relevant channels for information exchange. This mechanism ensures that only highly relevant features are activated, minimizing unnecessary interference. Finally, the Bidirectional Channel-wise Interaction~(BCI) strategy is employed to supplement additional semantic information and enhance the representation of important channels. Experimental results on multiple benchmarking 3D medical datasets demonstrate that the proposed method outperforms existing semi-supervised approaches.

cs.CV↗

arXiv2026

Rigidity Results for Compact Submanifolds with Pinched Ricci Curvature in Euclidean and Spherical Space Forms

For compact submanifolds in Euclidean and Spherical space forms with Ricci curvature bounded below by a function $α(n,k,H,c)$ of mean curvature, we prove that the submanifold is either isometric to the Einstein Clifford torus, or a topological sphere for the maximal bound $α(n,[\frac{n}{2}],H,c)$, or has up to $k$-th homology groups vanishing. This gives an almost complete (except for the differentiable sphere theorem) characterization of compact submanifolds with pinched Ricci curvature, generalizing celebrated rigidity results obtained by Ejiri, Xu-Tian, Xu-Gu, Xu-Leng-Gu, Vlachos, Dajczer-Vlachos.

math.DG↗

arXiv2026

Spectral point transformer for significant wave height estimation from sea clutter

This paper presents a method for estimating significant wave height (Hs) from sparse S_pectral P_oint using a T_ransformer-based approach (SPT). Based on empirical observations that only a minority of spectral points with strong power contribute to wave energy, the proposed SPT effectively integrates geometric and spectral characteristics of ocean surface waves to estimate Hs through multi-dimensional feature representation. The experiment reveals an intriguing phenomenon: the learned features of SPT align well with physical dispersion relations, where the contribution-score map of selected points is concentrated along dispersion curves. Compared to conventional vision networks that process image sequences and full spectra, SPT demonstrates superior performance in Hs regression while consuming significantly fewer computational resources. On a consumer-grade GPU, SPT completes the training of regression model for 1080 sea clutter image sequences within 4 minutes, showcasing its potential to reduce deployment costs for radar wave-measuring systems. The open-source implementation of SPT will be available at https://github.com/joeyee/spt

eess.SP↗

arXiv2026

Exact Clique Number Manipulation via Edge Interdiction

The Edge Interdiction Clique Problem (EICP) aims to remove at most $k$ edges from a graph so as to minimize the size of the largest clique in the remaining graph. This problem captures a fundamental question in graph manipulation: which edges are structurally critical for preserving large cliques? Such a problem is also motivated by practical applications including protein function maintenance and image matching. The EICP is computationally challenging and belongs to a complexity class beyond NP. Existing approaches rely on general mixed-integer bilevel programming solvers or reformulate the problem into a single-level mixed integer linear program. However, they are still not scalable when the graph size and interdiction budget $k$ grow. To overcome this, we investigate new mixed integer linear formulations, which recast the problem into a sequence of parameterized Edge Blocker Clique Problems (EBCP). This perspective decomposes the original problem into simpler subproblems and enables tighter modeling of clique-related inequalities. Furthermore, we propose a two-stage exact algorithm, \textsc{RLCM}, which first applies problem-specific reduction techniques to shrink the graph and then solves the reduced problem using a tailored branch-and-cut framework. Extensive computational experiments on maximum clique benchmark graphs, large real-world sparse networks, and random graphs demonstrate that \textsc{RLCM} consistently outperforms existing approaches.

cs.DS↗

arXiv2026

Normal Scalar Curvature Inequality on a Class of Austere Submanifolds

In this paper, we establish new normal scalar curvature inequalities on a class of austere submanifolds by proving sharper DDVV-type inequalities on associated austere subspaces. We also provide some examples of austere submanifolds in this class and point out one of them achieves the equality in our normal scalar curvature inequality everywhere. As a byproduct, we obtain a Simons-type gap theorem for closed austere submanifolds in unit spheres which belong to that class.

math.DG↗

arXiv2026

DST-Calib: A Dual-Path, Self-Supervised, Target-Free LiDAR-Camera Extrinsic Calibration Network

LiDAR-camera extrinsic calibration is essential for multi-modal data fusion in robotic perception systems. However, existing approaches typically rely on handcrafted calibration targets (e.g., checkerboards) or specific, static scene types, limiting their adaptability and deployment in real-world autonomous and robotic applications. This article presents the first self-supervised LiDAR-camera extrinsic calibration network that operates in an online fashion and eliminates the need for specific calibration targets. We first identify a significant generalization degradation problem in prior methods, caused by the conventional single-sided data augmentation strategy. To overcome this limitation, we propose a novel double-sided data augmentation technique that generates multi-perspective camera views using estimated depth maps, thereby enhancing robustness and diversity during training. Built upon this augmentation strategy, we design a dual-path, self-supervised calibration framework that reduces the dependence on high-precision ground truth labels and supports fully adaptive online calibration. Furthermore, to improve cross-modal feature association, we replace the traditional dual-branch feature extraction design with a difference map construction process that explicitly correlates LiDAR and camera features. This not only enhances calibration accuracy but also reduces model complexity. Extensive experiments conducted on five public benchmark datasets, as well as our own recorded dataset, demonstrate that the proposed method significantly outperforms existing approaches in terms of generalizability.

cs.RO↗

arXiv2025

A Low Background Beta Detection System using a Time Projection Chamber

In this paper, we present a Time Projection Chamber (TPC) system for low-background beta radiation measurements. The system consists of a TPC with two-dimensional-strip readout Micromegas and an anti-coincidence detector with readout pads for cosmic ray veto. The detector system utilize an AGET-based waveform sampling system for data acquisition. The beta detection capability of the system was verified through experimental test using $^{90}$Sr beta source. Additionally, a dedicated simulation program based on Geant4 was developed to model the entire detection process, including responses to both the beta source and background radiation. Simulation results were compared with experimental data for both beta and background samples, showing good agreements. The simulation samples were utilized to optimize and train classification models for beta and background discrimination. By applying the selected model into test data, the system achieved a background rate of 0.49 $\rm cpm/cm^2$ while retaining more than 55% of $^{90}$Sr beta signals within a 7 cm diameter detection region. Further analysis revealed that approximately 70% of the background originates from environmental gamma radiation, while the remaining contribution mainly comes from intrinsic radioactivity of detector materials, particularly the FR-4 based field cage and readout plane. Based on the knowledge gained from the experiments and simulations, an optimization of the TPC system has been proposed, with simulation predicting a potential reduction of the background rate to 0.0012 $\rm cpm/cm^2$.

physics.ins-det↗

arXiv2025

SeedFold: Scaling Biomolecular Structure Prediction

Highly accurate biomolecular structure prediction is a key component of developing biomolecular foundation models, and one of the most critical aspects of building foundation models is identifying the recipes for scaling the model. In this work, we present SeedFold, a folding model that successfully scales up the model capacity. Our contributions are threefold: first, we identify an effective width-scaling strategy for the Pairformer to increase representation capacity; second, we introduce a novel linear triangular attention that reduces computational complexity to enable efficient scaling; finally, we construct a large-scale distillation dataset to substantially enlarge the training set. Experiments on FoldBench show that SeedFold outperforms AlphaFold3 on most protein-related tasks.

q-bio.BM↗

arXiv2025

Local well-posedness for the quasilinear Schrödinger equations via the generalized energy method

We study the Cauchy problem of quasilinear Schrödinger equations, for which Kenig et al. (Invent Math, 2004; Adv Math, 2006) obtained large data local well-posedness by pseudo-differential techniques and viscosity methods, while Marzuola et al. (Adv Math, 2012; Kyoto J Math, 2014; Arch Ration Mech Anal, 2021) and Ben et al. (Arch Ration Mech Anal, 2024) improved the results by dispersive arguments. In this paper, we introduce a generalized energy method that combines momentum and energy estimates to close the bounds, thereby obtaining our results through viscosity methods. If the data is small, the proof relies mainly on integration by parts and Sobolev embeddings, much like the classical local existence theory for semilinear Schrödinger equations. For large data, the framework remains applicable with the incorporation of certain pseudo-differential tools. In the case of quadratic interactions, we establish low regularity local well-posedness for both small and large data in the same function spaces as in works of Kenig et al. For cubic interactions with small initial data, we recover the low regularity results obtained by Marzuola et al. (Kyoto J Math, 2014).

math.AP↗

arXiv2025

Design, simulation and performance of the resistive-anode PICOSEC Micromegas detector

The PICOSEC Micromegas detector is a Micro-Pattern Gaseous Detector concept developed to achieve tens of picosecond timing resolution for charged particle detection by combining a Cherenkov radiator with a two-stage Micromegas amplification structure. To improve operational robustness, a resistive anode has been implemented using a DLC layer deposited on a Kapton substrate. While this design enhances detector stability, the resistive layer may influence rate capability, signal formation, and detector capacitance, altering timing performance. This work presents a comprehensive study of a resistive design, including an analytical model and finite-element simulations to quantify rate-dependent gain reduction due to ohmic voltage drop on the resistive layer. An analytical solution for the voltage across a finite-size resistive layer is derived, and a numerical model is developed to evaluate gain suppression under intense particle fluxes. The impact of the resistive layer on signal formation is investigated using time-dependent weighting fields and the Garfield++ simulation framework. The contribution of signal components induced by the resistive layer is quantified, and preservation of the signal leading edge is found for surface resistivities above 100 kohm per square. Single-channel resistive-anode prototypes were designed, constructed, and experimentally characterized. Laboratory measurements using single photoelectrons and power spectral density analysis show the predicted reduction in signal amplitude while preserving the leading edge. Muon beam tests with CsI and DLC photocathodes demonstrate a time resolution of 11.5 ps for CsI, comparable to 11.9 ps for the metallic-anode device, showing the suitability of the resistive design for precision timing applications.

physics.ins-det↗

arXiv2025

Phenomenological Ginzburg-Landau theory for triple-Q magnetic orders on a hexagonal lattice

We develop a comprehensive Ginzburg-Landau theory describing triple-Q magnetic orders on hexagonal lattices, focusing on $O(N)$ models with $N=2$ and $N=3$. Through systematic analysis of symmetry-allowed terms in the free energy, we establish complete phase diagrams governed by competing interaction parameters. Our theory reveals distinct magnetic configurations including single-Q, double-Q, and triple-Q states, each characterized by unique symmetry breaking patterns and collective excitations. The framework provides fundamental insights into complex magnetic orders recently observed in materials such as Na$_2$Co$_2$TeO$_6$, where the interplay between geometric frustration and multiple ordering vectors leads to exotic magnetic states. Our results establish clear connections between microscopic interactions, broken symmetries, and experimentally observable properties, offering a powerful tool for understanding and predicting novel magnetic phases in frustrated magnets.

cond-mat.str-el↗