arXiv subjects

Zihan Zhang

Publications and source records attributed to Zihan Zhang.

At least 127 records · Page 7Linked to original sources

arXiv2025

Enhancing Character-Level Understanding in LLMs through Token Internal Structure Learning

Tokenization methods like Byte-Pair Encoding (BPE) enhance computational efficiency in large language models (LLMs) but often obscure internal character structures within tokens. This limitation hinders LLMs' ability to predict precise character positions, which is crucial in tasks like Chinese Spelling Correction (CSC) where identifying the positions of misspelled characters accelerates correction processes. We propose Token Internal Position Awareness (TIPA), a method that significantly improves models' ability to capture character positions within tokens by training them on reverse character prediction tasks using the tokenizer's vocabulary. Experiments demonstrate that TIPA enhances position prediction accuracy in LLMs, enabling more precise identification of target characters in original text. Furthermore, when applied to downstream tasks that do not require exact position prediction, TIPA still boosts performance in tasks needing character-level information, validating its versatility and effectiveness.

cs.CL↗

arXiv2025

Generative Models at the Frontier of Compression: A Survey on Generative Face Video Coding

The rise of deep generative models has greatly advanced video compression, reshaping the paradigm of face video coding through their powerful capability for semantic-aware representation and lifelike synthesis. Generative Face Video Coding (GFVC) stands at the forefront of this revolution, which could characterize complex facial dynamics into compact latent codes for bitstream compactness at the encoder side and leverages powerful deep generative models to reconstruct high-fidelity face signal from the compressed latent codes at the decoder side. As such, this well-designed GFVC paradigm could enable high-fidelity face video communication at ultra-low bitrate ranges, far surpassing the capabilities of the latest Versatile Video Coding (VVC) standard. To pioneer foundational research and accelerate the evolution of GFVC, this paper presents the first comprehensive survey of GFVC technologies, systematically bridging critical gaps between theoretical innovation and industrial standardization. In particular, we first review a broad range of existing GFVC methods with different feature representations and optimization strategies, and conduct a thorough benchmarking analysis. In addition, we construct a large-scale GFVC-compressed face video database with subjective Mean Opinion Scores (MOSs) based on human perception, aiming to identify the most appropriate quality metrics tailored to GFVC. Moreover, we summarize the GFVC standardization potentials with a unified high-level syntax and develop a low-complexity GFVC system which are both expected to push forward future practical deployments and applications. Finally, we envision the potential of GFVC in industrial applications and deliberate on the current challenges and future opportunities.

cs.CV↗

arXiv2025

Identity Deepfake Threats to Biometric Authentication Systems: Public and Expert Perspectives

Generative AI (Gen-AI) deepfakes pose a rapidly evolving threat to biometric authentication, yet a significant gap exists between expert understanding of these risks and public perception. This disconnection creates critical vulnerabilities in systems trusted by millions. To bridge this gap, we conducted a comprehensive mixed-method study, surveying 408 professionals across key sectors and conducting in-depth interviews with 37 participants (25 experts, 12 general public [non-experts]). Our findings reveal a paradox: while the public increasingly relies on biometrics for convenience, experts express grave concerns about the spoofing of static modalities like face and voice recognition. We found significant demographic and sector-specific divides in awareness and trust, with finance professionals, for example, showing heightened skepticism. To systematically analyze these threats, we introduce a novel Deepfake Kill Chain model, adapted from Hutchins et al.'s cybersecurity frameworks to map the specific attack vectors used by malicious actors against biometric systems. Based on this model and our empirical findings, we propose a tri-layer mitigation framework that prioritizes dynamic biometric signals (e.g., eye movements), robust privacy-preserving data governance, and targeted educational initiatives. This work provides the first empirically grounded roadmap for defending against AI-generated identity threats by aligning technical safeguards with human-centered insights.

cs.HC↗

arXiv2025

Sharp Gap-Dependent Variance-Aware Regret Bounds for Tabular MDPs

We consider the gap-dependent regret bounds for episodic MDPs. We show that the Monotonic Value Propagation (MVP) algorithm achieves a variance-aware gap-dependent regret bound of $$\tilde{O}\left(\left(\sum_{Δ_h(s,a)>0} \frac{H^2 \log K \land \mathtt{Var}_{\max}^{\text{c}}}{Δ_h(s,a)} +\sum_{Δ_h(s,a)=0}\frac{ H^2 \land \mathtt{Var}_{\max}^{\text{c}}}{Δ_{\mathrm{min}}} + SAH^4 (S \lor H) \right) \log K\right),$$ where $H$ is the planning horizon, $S$ is the number of states, $A$ is the number of actions, and $K$ is the number of episodes. Here, $Δ_h(s,a) =V_h^* (a) - Q_h^* (s, a)$ represents the suboptimality gap and $Δ_{\mathrm{min}} := \min_{Δ_h (s,a) > 0} Δ_h(s,a)$. The term $\mathtt{Var}_{\max}^{\text{c}}$ denotes the maximum conditional total variance, calculated as the maximum over all $(π, h, s)$ tuples of the expected total variance under policy $π$ conditioned on trajectories visiting state $s$ at step $h$. $\mathtt{Var}_{\max}^{\text{c}}$ characterizes the maximum randomness encountered when learning any $(h, s)$ pair. Our result stems from a novel analysis of the weighted sum of the suboptimality gap and can be potentially adapted for other algorithms. To complement the study, we establish a lower bound of $$Ω\left( \sum_{Δ_h(s,a)>0} \frac{H^2 \land \mathtt{Var}_{\max}^{\text{c}}}{Δ_h(s,a)}\cdot \log K\right),$$ demonstrating the necessity of dependence on $\mathtt{Var}_{\max}^{\text{c}}$ even when the maximum unconditional total variance (without conditioning on $(h, s)$) approaches zero.

cs.LG↗

arXiv2025

AG Codes Achieve List-decoding Capacity over Constant-sized Fields

The recently-emerging field of higher order MDS codes has sought to unify a number of concepts in coding theory. Such areas captured by higher order MDS codes include maximally recoverable (MR) tensor codes, codes with optimal list-decoding guarantees, and codes with constrained generator matrices (as in the GM-MDS theorem). By proving these equivalences, Brakensiek-Gopi-Makam showed the existence of optimally list-decodable Reed-Solomon codes over exponential sized fields. Building on this, recent breakthroughs by Guo-Zhang and Alrabiah-Guruswami-Li have shown that randomly punctured Reed-Solomon codes achieve list-decoding capacity (which is a relaxation of optimal list-decodability) over linear size fields. We extend these works by developing a formal theory of relaxed higher order MDS codes. In particular, we show that there are two inequivalent relaxations which we call lower and upper relaxations. The lower relaxation is equivalent to relaxed optimal list-decodable codes and the upper relaxation is equivalent to relaxed MR tensor codes with a single parity check per column. We then generalize the techniques of GZ and AGL to show that both these relaxations can be constructed over constant size fields by randomly puncturing suitable algebraic-geometric codes. For this, we crucially use the generalized GM-MDS theorem for polynomial codes recently proved by Brakensiek-Dhar-Gopi. We obtain the following corollaries from our main result. First, randomly punctured AG codes of rate $R$ achieve list-decoding capacity with list size $O(1/ε)$ and field size $\exp(O(1/ε^2))$. Prior to this work, AG codes were not even known to achieve list-decoding capacity. Second, by randomly puncturing AG codes, we can construct relaxed MR tensor codes with a single parity check per column over constant-sized fields, whereas (non-relaxed) MR tensor codes require exponential field size.

cs.IT↗

arXiv2025

MiMo-VL Technical Report

We open-source MiMo-VL-7B-SFT and MiMo-VL-7B-RL, two powerful vision-language models delivering state-of-the-art performance in both general visual understanding and multimodal reasoning. MiMo-VL-7B-RL outperforms Qwen2.5-VL-7B on 35 out of 40 evaluated tasks, and scores 59.4 on OlympiadBench, surpassing models with up to 78B parameters. For GUI grounding applications, it sets a new standard with 56.1 on OSWorld-G, even outperforming specialized models such as UI-TARS. Our training combines four-stage pre-training (2.4 trillion tokens) with Mixed On-policy Reinforcement Learning (MORL) integrating diverse reward signals. We identify the importance of incorporating high-quality reasoning data with long Chain-of-Thought into pre-training stages, and the benefits of mixed RL despite challenges in simultaneous multi-domain optimization. We also contribute a comprehensive evaluation suite covering 50+ tasks to promote reproducibility and advance the field. The model checkpoints and full evaluation suite are available at https://github.com/XiaomiMiMo/MiMo-VL.

cs.CL↗

arXiv2025

Decay of correlation for edge colorings when $q>3Δ$

We examine various perspectives on the decay of correlation for the uniform distribution over proper $q$-edge colorings of graphs with maximum degree $Δ$. First, we establish the coupling independence property when $q\ge 3Δ$ for general graphs. Together with the work of Chen et al. (2024), this result implies a fully polynomial-time approximation scheme (FPTAS) for counting the number of proper $q$-edge colorings. Next, we prove the strong spatial mixing property on trees, provided that $q> (3+o(1))Δ$. The strong spatial mixing property is derived from the spectral independence property of a version of the weighted edge coloring distribution, which is established using the matrix trickle-down method developed in Abdolazimi, Liu and Oveis Gharan (FOCS, 2021) and Wang, Zhang and Zhang (STOC, 2024). Finally, we show that the weak spatial mixing property holds on trees with maximum degree $Δ$ if and only if $q\ge 2Δ-1$.

cs.DS↗

arXiv2025

AISHELL-5: The First Open-Source In-Car Multi-Channel Multi-Speaker Speech Dataset for Automatic Speech Diarization and Recognition

This paper delineates AISHELL-5, the first open-source in-car multi-channel multi-speaker Mandarin automatic speech recognition (ASR) dataset. AISHLL-5 includes two parts: (1) over 100 hours of multi-channel speech data recorded in an electric vehicle across more than 60 real driving scenarios. This audio data consists of four far-field speech signals captured by microphones located on each car door, as well as near-field signals obtained from high-fidelity headset microphones worn by each speaker. (2) a collection of 40 hours of real-world environmental noise recordings, which supports the in-car speech data simulation. Moreover, we also provide an open-access, reproducible baseline system based on this dataset. This system features a speech frontend model that employs speech source separation to extract each speaker's clean speech from the far-field signals, along with a speech recognition module that accurately transcribes the content of each individual speaker. Experimental results demonstrate the challenges faced by various mainstream ASR models when evaluated on the AISHELL-5. We firmly believe the AISHELL-5 dataset will significantly advance the research on ASR systems under complex driving scenarios by establishing the first publicly available in-car ASR benchmark.

cs.SD↗

arXiv2025

Compressing Human Body Video with Interactive Semantics: A Generative Approach

In this paper, we propose to compress human body video with interactive semantics, which can facilitate video coding to be interactive and controllable by manipulating semantic-level representations embedded in the coded bitstream. In particular, the proposed encoder employs a 3D human model to disentangle nonlinear dynamics and complex motion of human body signal into a series of configurable embeddings, which are controllably edited, compactly compressed, and efficiently transmitted. Moreover, the proposed decoder can evolve the mesh-based motion fields from these decoded semantics to realize the high-quality human body video reconstruction. Experimental results illustrate that the proposed framework can achieve promising compression performance for human body videos at ultra-low bitrate ranges compared with the state-of-the-art video coding standard Versatile Video Coding (VVC) and the latest generative compression schemes. Furthermore, the proposed framework enables interactive human body video coding without any additional pre-/post-manipulation processes, which is expected to shed light on metaverse-related digital human communication in the future.

eess.IV↗

arXiv2025

Settling the Sample Complexity of Online Reinforcement Learning

A central issue lying at the heart of online reinforcement learning (RL) is data efficiency. While a number of recent works achieved asymptotically minimal regret in online RL, the optimality of these results is only guaranteed in a ``large-sample'' regime, imposing enormous burn-in cost in order for their algorithms to operate optimally. How to achieve minimax-optimal regret without incurring any burn-in cost has been an open problem in RL theory. We settle this problem for the context of finite-horizon inhomogeneous Markov decision processes. Specifically, we prove that a modified version of Monotonic Value Propagation (MVP), a model-based algorithm proposed by \cite{zhang2020reinforcement}, achieves a regret on the order of (modulo log factors) \begin{equation*} \min\big\{ \sqrt{SAH^3K}, \,HK \big\}, \end{equation*} where $S$ is the number of states, $A$ is the number of actions, $H$ is the planning horizon, and $K$ is the total number of episodes. This regret matches the minimax lower bound for the entire range of sample size $K\geq 1$, essentially eliminating any burn-in requirement. It also translates to a PAC sample complexity (i.e., the number of episodes needed to yield $\varepsilon$-accuracy) of $\frac{SAH^3}{\varepsilon^2}$ up to log factor, which is minimax-optimal for the full $\varepsilon$-range. Further, we extend our theory to unveil the influences of problem-dependent quantities like the optimal value/cost and certain variances. The key technical innovation lies in the development of a new regret decomposition strategy and a novel analysis paradigm to decouple complicated statistical dependency -- a long-standing challenge facing the analysis of online RL in the sample-hungry regime.

cs.LG↗

arXiv2025

SmoothSegNet: A Global-Local Framework for Liver Tumor Segmentation with Clinical KnowledgeInformed Label Smoothing

Liver cancer is a leading cause of mortality worldwide, and accurate Computed Tomography (CT)-based tumor segmentation is essential for diagnosis and treatment. Manual delineation is time-intensive, prone to variability, and highlights the need for reliable automation. While deep learning has shown promise for automated liver segmentation, precise liver tumor segmentation remains challenging due to the heterogeneous nature of tumors, imprecise tumor margins, and limited labeled data. We present SmoothSegNet, a novel deep learning framework that addresses these challenges with the three key designs: (1) A novel knowledge-informed label smoothing technique that distills knowledge from clinical data to generate smooth labels, which are used to regularize model training, reducing the overfitting risk and enhancing model performance; (2) A global and local segmentation framework that breaks down the main task into two simpler sub-tasks, allowing optimized preprocessing and training for each; and (3) Pre- and post-processing pipelines customized to the challenges of each subtask aimed to enhance tumor visibility and refines tumor boundaries. We apply the proposed model on a challenging HCC-TACE-Seg dataset and show that SmoothSegNet outperformed various benchmarks in segmentation performance, particularly at smaller tumors (<10cm). Our ablation studies show that the three design components complementarily contribute to the model improved performance. Code for the proposed method are available at https://github.com/lingchm/medassist-liver-cancer.

cs.LG↗

arXiv2025

Explicit Folded Reed-Solomon and Multiplicity Codes Achieve Relaxed Generalized Singleton Bounds

In this paper, we prove that explicit FRS codes and multiplicity codes achieve relaxed generalized Singleton bounds for list size $L\ge1.$ Specifically, we show the following: (1) FRS code of length $n$ and rate $R$ over the alphabet $\mathbb{F}_q^s$ with distinct evaluation points is $\left(\frac{L}{L+1}\left(1-\frac{sR}{s-L+1}\right),L\right)$ list-decodable (LD) for list size $L\in[s]$. (2) Multiplicity code of length $n$ and rate $R$ over the alphabet $\mathbb{F}_p^s$ with distinct evaluation points is $\left(\frac{L}{L+1}\left(1-\frac{sR}{s-L+1}\right),L\right)$ LD for list size $L\in[s]$. Choosing $s=Θ(1/ε^2)$ and $L=O(1/ε)$, our results imply that both FRS codes and multiplicity codes achieve LD capacity $1-R-ε$ with optimal list size $O(1/ε)$. This exponentially improves the previous state of the art $(1/ε)^{O(1/ε)}$ established by Kopparty et. al. (FOCS 2018) and Tamo (IEEE TIT, 2024). In particular, our results on FRS codes fully resolve a open problem proposed by Guruswami and Rudra (STOC 2006). Furthermore, our results imply the first explicit constructions of $(1-R-ε,O(1/ε))$ LD codes of rate $R$ with poly-sized alphabets. Our method can also be extended to analyze the list-recoverability (LR) of FRS codes. We provide a tighter radius upper bound that FRS codes cannot be $(\frac{L+1-\ell}{L+1}(1-\frac{mR}{m-1})+o(1),\ell, L)$ LR where $m=\lceil\log_{\ell}{(L+1)}\rceil$. We conjecture this bound is almost tight when $L+1=\ell^a$ for any $a\in\mathbb{N}^{\ge 2}$. To give some evidences, we show FRS codes are $\left(\frac{1}{2}-\frac{sR}{s-2},2,3\right)$ LR, which proves the tightness in the smallest non-trivial case. Our bound refutes the possibility that FRS codes could achieve LR capacity $(1-R-ε, \ell, O(\frac{\ell}ε))$. This implies an intrinsic separation between LD and LR of FRS codes.

cs.IT↗

arXiv2025

A Solver-Aided Hierarchical Language for LLM-Driven CAD Design

Large language models (LLMs) have been enormously successful in solving a wide variety of structured and unstructured generative tasks, but they struggle to generate procedural geometry in Computer Aided Design (CAD). These difficulties arise from an inability to do spatial reasoning and the necessity to guide a model through complex, long range planning to generate complex geometry. We enable generative CAD Design with LLMs through the introduction of a solver-aided, hierarchical domain specific language (DSL) called AIDL, which offloads the spatial reasoning requirements to a geometric constraint solver. Additionally, we show that in the few-shot regime, AIDL outperforms even a language with in-training data (OpenSCAD), both in terms of generating visual results closer to the prompt and creating objects that are easier to post-process and reason about.

cs.CV↗

arXiv2025

Reconstructing short-lived particles using hypergraph representation learning

In collider experiments, the kinematic reconstruction of heavy, short-lived particles is vital for precision tests of the Standard Model and in searches for physics beyond it. Performing kinematic reconstruction in collider events with many final-state jets, such as the all-hadronic decay of top-antitop quark pairs, is challenging. We present HyPER: Hypergraph for Particle Event Reconstruction, a novel architecture based on graph neural networks that uses hypergraph representation learning to build more powerful and efficient representations of collider events. HyPER is used to reconstruct parent particles from sets of final-state objects. Trained and tested on simulation, the HyPER model is shown to perform favorably when compared to existing state-of-the-art reconstruction techniques, while demonstrating superior parameter efficiency. The novel hypergraph approach allows the method to be applied to particle reconstruction in a multitude of different physics processes.

hep-ph↗

arXiv2025

MixDec Sampling: A Soft Link-based Sampling Method of Graph Neural Network for Recommendation

Graph neural networks have been widely used in recent recommender systems, where negative sampling plays an important role. Existing negative sampling methods restrict the relationship between nodes as either hard positive pairs or hard negative pairs. This leads to the loss of structural information, and lacks the mechanism to generate positive pairs for nodes with few neighbors. To overcome limitations, we propose a novel soft link-based sampling method, namely MixDec Sampling, which consists of Mixup Sampling module and Decay Sampling module. The Mixup Sampling augments node features by synthesizing new nodes and soft links, which provides sufficient number of samples for nodes with few neighbors. The Decay Sampling strengthens the digestion of graph structure information by generating soft links for node embedding learning. To the best of our knowledge, we are the first to model sampling relationships between nodes by soft links in GNN-based recommender systems. Extensive experiments demonstrate that the proposed MixDec Sampling can significantly and consistently improve the recommendation performance of several representative GNN-based models on various recommendation benchmarks.

cs.IR↗

arXiv2025

FF7: A Code Package for High-throughput Calculations and Constructing Materials Database

Decades accumulation of theory simulations lead to boom in material database, which combined with machine learning methods has been a valuable driver for the data-intensive material discovery, i.e., the fourth research paradigm. However, construction of segmented databases and data reuse in generic databases with uniform parameters still lack easy-to-use code tools. We herein develop a code package named FF7 (Fast Funnel with 7 modules) to provide command-line based interactive interface for performing customized high-throughput calculations and building your own handy databases. Data correlation studies and material property prediction can progress by built-in installation-free artificial neural network module and various post processing functions are also supported by auxiliary module. This paper shows the usage of FF7 code package and demonstrates its usefulness by example of database driven thermodynamic stability high-throughput calculation and machine learning model for predicting the superconducting critical temperature of clathrate hydrides.

cond-mat.mtrl-sci↗

arXiv2025

Breaking Barriers or Building Dependency? Exploring Team-LLM Collaboration in AI-infused Classroom Debate

Classroom debates are a unique form of collaborative learning characterized by fast-paced, high-intensity interactions that foster critical thinking and teamwork. Despite the recognized importance of debates, the role of AI tools, particularly LLM-based systems, in supporting this dynamic learning environment has been under-explored in HCI. This study addresses this opportunity by investigating the integration of LLM-based AI into real-time classroom debates. Over four weeks, 22 students in a Design History course participated in three rounds of debates with support from ChatGPT. The findings reveal how learners prompted the AI to offer insights, collaboratively processed its outputs, and divided labor in team-AI interactions. The study also surfaces key advantages of AI usage, reducing social anxiety, breaking communication barriers, and providing scaffolding for novices, alongside risks, such as information overload and cognitive dependency, which could limit learners' autonomy. We thereby discuss a set of nuanced implications for future HCI exploration.

cs.HC↗

arXiv2024

High-throughput discovery of robust room-temperature superconductors among complex ternary clathrate hydrides

After the decade-long exhaustive study of binary high-Tc superconducting hydrides, the frontier of this stimulating research field has recently shifted to ternary hydrides with much expanded conformational space in search of coveted room-temperature superconductors. This task, however, presents a formidable challenge due to enormous demands on computational resources. Here, we devise an efficient high-throughput approach using keen material insights and a self-built database to screen for robust ternary hydrides in clathrate structures, which were proven to host highest Tc in binary hydrides, and to estimate Tc by a reliable empirical formula. This approach has made it possible to uncover a diverse set of complex multiple-hydrogen-cage ternary hydrides hosting near or above room-temperature Tc, which are beyond the reach of prevailing structure search methods. This study establishes a distinct paradigm that opens a fresh avenue to enable and accelerate the discovery of promising room-temperature superconductors among unprecedented complex clathrate hydrides.

cond-mat.supr-con↗