arXiv subjects

Zhiyuan Shi

Publications and source records attributed to Zhiyuan Shi.

At least 19 recordsLinked to original sources

arXiv2026

HarnessEval-W: Agentifying the Evaluation of Visual Worlds

A benchmark should deliver more than a scalar score: what makes an evaluation trustworthy is the reasoning that justifies the score. This is especially critical for world models, where judging a rollout requires understanding whether physics, causality, and world state evolve correctly. Humans spot such violations naturally, yet no existing benchmark automates this capability: metrics are computed brute-force, leaving no reasoning chain that can be examined or verified. We introduce HarnessEval-W, an agentified evaluation pipeline that brings the harness paradigm from the LLM ecosystem to world model benchmarking. Rather than applying a fixed rubric, HarnessEval-W interprets the context of each evaluation case, decomposes the evaluation question into measurable subproblems, and spawns specialized sub-agents, each equipped with tailored context and diagnostic tools to reason over its own subproblem. The parent agent then validates the gathered evidence and summarizes it into the final verdict. This hierarchical workflow turns every evaluation into a transparent evidence tree whose complete reasoning chain justifies the result. We apply HarnessEval-W to 18 representative world models over 330 evaluation cases. Its judgments closely align with human preferences while providing verifiable, fine-grained diagnoses of every generated rollout. We open-source the full pipeline as a live benchmark and invite the broad community to contribute to grow new skills and evaluation cases as world models evolve.

cs.CV↗

arXiv2026

GEB-Bench: Abstract Structures Told in Many Voices

Can a model look at a river delta and a lightning bolt and see that they share a structure? We introduce GEB-Bench, a benchmark whose unit is an abstract structural motif--self-reference, a strange loop, a Mobius twist--in the spirit of Godel, Escher, Bach. Each motif is told in several voices: a natural scene whose composition is the structure, a folk story whose telling enacts it through a mechanically checkable form device, a mathematical theorem, and a programmatic skeleton; surface parameters are declared nuisance variables and never scored. Motifs, voices, and the structural changes between them form a small cross-modal category, and GEB-Bench's tasks are its questions. Evaluating twelve open and proprietary models, we find that abstraction failure is lawful. The central finding is a gap between recognition and cross-voice mapping: models identify a structure within one voice far better than they carry it across voices; every model pays this tax, and mapping strong enough to narrow it appears only at the frontier tier. Two patterns support it. Errors align more strongly with the designed formal geometry than with measured perceptual geometries, and frontier models from different vendors converge on the same wrong answers; and surface complexity taxes every model that reads structure, with capacity buying headroom rather than immunity. GEB-Bench is fully generative and released with its pipeline.

cs.CV↗

arXiv2026

Sundial: A Family of Highly Capable Time Series Foundation Models

We introduce Sundial, a family of native, flexible, and scalable time series foundation models. To predict the next-patch's distribution, we propose a TimeFlow Loss based on flow-matching, which facilitates native pre-training of Transformers on continuous-valued time series without discrete tokenization. Conditioned on arbitrary-length time series, our models are pre-trained without specifying any prior distribution and can generate multiple probable predictions, achieving more flexibility in representation learning than using parametric densities. Towards time series foundation models, we leverage minimal but crucial adaptations of Transformers and curate TimeBench with one trillion time points, comprising mostly real-world datasets and synthetic data. By mitigating mode collapse via TimeFlow Loss, we pre-train a family of Sundial models on TimeBench, which achieve unprecedented model capacity and generalization performance. In addition to excellent scalability, Sundial achieves state-of-the-art results on both point and probabilistic forecasting benchmarks with a just-in-time inference speed, i.e., making zero-shot predictions within a few milliseconds. We believe that Sundial's pioneering generative forecasting capability can improve model reliability in real-world decision-making. Code is available at: https://github.com/thuml/Sundial.

cs.LG↗

arXiv2026

HeteroCache: A Dynamic Retrieval Approach to Heterogeneous KV Cache Compression for Long-Context LLM Inference

The linear memory growth of the KV cache poses a significant bottleneck for LLM inference in long-context tasks. Existing static compression methods often fail to preserve globally important information. Although recent dynamic retrieval approaches attempt to address this issue, they typically suffer from coarse-grained caching strategies and incur high I/O overhead. To overcome these limitations, we propose HeteroCache, a training-free dynamic compression framework. Our method is built on two key insights: attention heads exhibit diverse temporal heterogeneity, and there is significant spatial redundancy among heads within the same layer. Guided by these insights, HeteroCache categorizes heads based on stability and similarity, applying a fine-grained weighting strategy that allocates larger cache budgets to heads with rapidly shifting attention to capture context changes. Furthermore, it features a hierarchical storage mechanism where representative heads monitor attention drift to trigger asynchronous, on-demand context retrieval, thereby hiding I/O latency. Experiments demonstrate that HeteroCache achieves state-of-the-art performance on long-context benchmarks and accelerates decoding by up to $3\times$ compared to the original model with a 224K context. Our code is available at https://github.com/ponytaill/HeteroCache.

cs.CL↗

arXiv2025

CoRA: Covariate-Aware Adaptation of Time Series Foundation Models

Time Series Foundation Models (TSFMs) have shown significant impact through their model capacity, scalability, and zero-shot generalization. However, due to the heterogeneity of inter-variate dependencies and the backbone scalability on large-scale multivariate datasets, most TSFMs are typically pre-trained on univariate time series. This limitation renders them oblivious to crucial information from diverse covariates in real-world forecasting tasks. To further enhance the performance of TSFMs, we propose a general covariate-aware adaptation (CoRA) framework for TSFMs. It leverages pre-trained backbones of foundation models while effectively incorporating exogenous covariates from various modalities, including time series, language, and images, to improve the quality of predictions. Technically, CoRA maintains the equivalence of initialization and parameter consistency during adaptation. With preserved backbones of foundation models as frozen feature extractors, the outcome embeddings from foundation models are empirically demonstrated more informative than raw data. Further, CoRA employs a novel Granger Causality Embedding (GCE) to automatically evaluate covariates regarding their causal predictability with respect to the target variate. We incorporate these weighted embeddings with a zero-initialized condition-injection mechanism, avoiding catastrophic forgetting of pre-trained foundation models and gradually integrates exogenous information. Extensive experiments show that CoRA of TSFMs surpasses state-of-the-art covariate-aware deep forecasters with full or few-shot training samples, achieving 31.1% MSE reduction on covariate-aware forecasting. Compared to other adaptation methods, CoRA exhibits strong compatibility with various advanced TSFMs and extends the scope of covariates to other modalities, presenting a practical paradigm for the application of TSFMs.

cs.LG↗

arXiv2025

Invariant measures for stochastic Burgers equation on unbounded domains

In this paper, we investigate the stochastic damped Burgers equation with multiplicative noise defined on the entire real line. We demonstrate the existence and uniqueness of a mild solution to the stochastic damped Burgers equation and establish that the solution is uniformly bounded in time. Furthermore, by employing the uniform estimates on the tails of the solution, we obtain the tightness of a family of probability distributions of the solution. Subsequently, by applying the Krylov-Bogolioubov theorem, we establish the existence of invariant measures.

math.DS↗

arXiv2025

Invariant measures for stochastic Burgers equation in unbounded domains with space-time white noise

In this paper, we investigate the stochastic damped Burgers equation with multiplicative space-time white noise defined on the entire real line. We prove the existence and uniqueness of a mild solution of the stochastic damped Burgers equation in the weighted space and establish that the solution is bounded in probability. Furthermore, by using the Krylov-Bogolioubov theorem, we obtain the existence of invariant measures.

math.DS↗

arXiv2022

Minimizing the programming power of phase change memory by using graphene nanoribbon edge-contact

Nonvolatile phase change random access memory (PCRAM) is regarded as one of promising candidates for emerging mass storage in the era of Big Data. However, relatively high programming energy hurdles the further reduction of power consumption in PCRAM. Utilizing narrow edge-contact of graphene can effectively reduce the active volume of phase change material in each cell, and therefore realize low-power operation. Here, we demonstrate that a write energy can be reduced to about ~53.7 fJ in a cell with ~3 nm-wide graphene nanoribbon (GNR) as edge-contact, whose cross-sectional area is only ~1 nm2. It is found that the cycle endurance exhibits an obvious dependence on the bias polarity in the cell with structure asymmetry. If a positive bias was applied to graphene electrode, the endurance can be extended at least one order longer than the case with reversal of polarity. The work represents a great technological advance for the low power PCRAM and could benefit for in-memory computing in future.

cond-mat.mes-hall↗

arXiv2022

WiFi-based Spatiotemporal Human Action Perception

WiFi-based sensing for human activity recognition (HAR) has recently become a hot topic as it brings great benefits when compared with video-based HAR, such as eliminating the demands of line-of-sight (LOS) and preserving privacy. Making the WiFi signals to 'see' the action, however, is quite coarse and thus still in its infancy. An end-to-end spatiotemporal WiFi signal neural network (STWNN) is proposed to enable WiFi-only sensing in both line-of-sight and non-line-of-sight scenarios. Especially, the 3D convolution module is able to explore the spatiotemporal continuity of WiFi signals, and the feature self-attention module can explicitly maintain dominant features. In addition, a novel 3D representation for WiFi signals is designed to preserve multi-scale spatiotemporal information. Furthermore, a small wireless-vision dataset (WVAR) is synchronously collected to extend the potential of STWNN to 'see' through occlusions. Quantitative and qualitative results on WVAR and the other three public benchmark datasets demonstrate the effectiveness of our approach on both accuracy and shift consistency.

eess.SP↗

arXiv2022

GraSens: A Gabor Residual Anti-aliasing Sensing Framework for Action Recognition using WiFi

WiFi-based human action recognition (HAR) has been regarded as a promising solution in applications such as smart living and remote monitoring due to the pervasive and unobtrusive nature of WiFi signals. However, the efficacy of WiFi signals is prone to be influenced by the change in the ambient environment and varies over different sub-carriers. To remedy this issue, we propose an end-to-end Gabor residual anti-aliasing sensing network (GraSens) to directly recognize the actions using the WiFi signals from the wireless devices in diverse scenarios. In particular, a new Gabor residual block is designed to address the impact of the changing surrounding environment with a focus on learning reliable and robust temporal-frequency representations of WiFi signals. In each block, the Gabor layer is integrated with the anti-aliasing layer in a residual manner to gain the shift-invariant features. Furthermore, fractal temporal and frequency self-attention are proposed in a joint effort to explicitly concentrate on the efficacy of WiFi signals and thus enhance the quality of output features scattered in different subcarriers. Experimental results throughout our wireless-vision action recognition dataset (WVAR) and three public datasets demonstrate that our proposed GraSens scheme outperforms state-of-the-art methods with respect to recognition accuracy.

cs.CV↗

arXiv2022

A Wireless-Vision Dataset for Privacy Preserving Human Activity Recognition

Human Activity Recognition (HAR) has recently received remarkable attention in numerous applications such as assisted living and remote monitoring. Existing solutions based on sensors and vision technologies have obtained achievements but still suffering from considerable limitations in the environmental requirement. Wireless signals like WiFi-based sensing have emerged as a new paradigm since it is convenient and not restricted in the environment. In this paper, a new WiFi-based and video-based neural network (WiNN) is proposed to improve the robustness of activity recognition where the synchronized video serves as the supplement for the wireless data. Moreover, a wireless-vision benchmark (WiVi) is collected for 9 class actions recognition in three different visual conditions, including the scenes without occlusion, with partial occlusion, and with full occlusion. Both machine learning methods - support vector machine (SVM) as well as deep learning methods are used for the accuracy verification of the data set. Our results show that WiVi data set satisfies the primary demand and all three branches in the proposed pipeline keep more than $80\%$ of activity recognition accuracy over multiple action segmentation from 1s to 3s. In particular, WiNN is the most robust method in terms of all the actions on three action segmentation compared to the others.

cs.CV↗

arXiv2019

When Machine Learning Meets Big Data: A Wireless Communication Perspective

We have witnessed an exponential growth in commercial data services, which has lead to the 'big data era'. Machine learning, as one of the most promising artificial intelligence tools of analyzing the deluge of data, has been invoked in many research areas both in academia and industry. The aim of this article is twin-fold. Firstly, we briefly review big data analysis and machine learning, along with their potential applications in next-generation wireless networks. The second goal is to invoke big data analysis to predict the requirements of mobile users and to exploit it for improving the performance of "social network-aware wireless". More particularly, a unified big data aided machine learning framework is proposed, which consists of feature extraction, data modeling and prediction/online refinement. The main benefits of the proposed framework are that by relying on big data which reflects both the spectral and other challenging requirements of the users, we can refine the motivation, problem formulations and methodology of powerful machine learning algorithms in the context of wireless networks. In order to characterize the efficiency of the proposed framework, a pair of intelligent practical applications are provided as case studies: 1) To predict the positioning of drone-mounted areal base stations (BSs) according to the specific tele-traffic requirements by gleaning valuable data from social networks. 2) To predict the content caching requirements of BSs according to the users' preferences by mining data from social networks. Finally, open research opportunities are identified for motivating future investigations.

cs.NI↗

arXiv2017

Learning and Refining of Privileged Information-based RNNs for Action Recognition from Depth Sequences

Existing RNN-based approaches for action recognition from depth sequences require either skeleton joints or hand-crafted depth features as inputs. An end-to-end manner, mapping from raw depth maps to action classes, is non-trivial to design due to the fact that: 1) single channel map lacks texture thus weakens the discriminative power; 2) relatively small set of depth training data. To address these challenges, we propose to learn an RNN driven by privileged information (PI) in three-steps: An encoder is pre-trained to learn a joint embedding of depth appearance and PI (i.e. skeleton joints). The learned embedding layers are then tuned in the learning step, aiming to optimize the network by exploiting PI in a form of multi-task loss. However, exploiting PI as a secondary task provides little help to improve the performance of a primary task (i.e. classification) due to the gap between them. Finally, a bridging matrix is defined to connect two tasks by discovering latent PI in the refining step. Our PI-based classification loss maintains a consistency between latent PI and predicted distribution. The latent PI and network are iteratively estimated and updated in an expectation-maximization procedure. The proposed learning process provides greater discriminative power to model subtle depth difference, while helping avoid overfitting the scarcer training data. Our experiments show significant performance gains over state-of-the-art methods on three public benchmark datasets and our newly collected Blanket dataset.

cs.CV↗

arXiv2017

Weakly Supervised Image Annotation and Segmentation with Objects and Attributes

We propose to model complex visual scenes using a non-parametric Bayesian model learned from weakly labelled images abundant on media sharing sites such as Flickr. Given weak image-level annotations of objects and attributes without locations or associations between them, our model aims to learn the appearance of object and attribute classes as well as their association on each object instance. Once learned, given an image, our model can be deployed to tackle a number of vision problems in a joint and coherent manner, including recognising objects in the scene (automatic object annotation), describing objects using their attributes (attribute prediction and association), and localising and delineating the objects (object detection and semantic segmentation). This is achieved by developing a novel Weakly Supervised Markov Random Field Stacked Indian Buffet Process (WS-MRF-SIBP) that models objects and attributes as latent factors and explicitly captures their correlations within and across superpixels. Extensive experiments on benchmark datasets demonstrate that our weakly supervised model significantly outperforms weakly supervised alternatives and is often comparable with existing strongly supervised models on a variety of tasks including semantic segmentation, automatic image annotation and retrieval based on object-attribute associations.

cs.CV↗

arXiv2017

Bayesian Joint Modelling for Object Localisation in Weakly Labelled Images

We address the problem of localisation of objects as bounding boxes in images and videos with weak labels. This weakly supervised object localisation problem has been tackled in the past using discriminative models where each object class is localised independently from other classes. In this paper, a novel framework based on Bayesian joint topic modelling is proposed, which differs significantly from the existing ones in that: (1) All foreground object classes are modelled jointly in a single generative model that encodes multiple object co-existence so that "explaining away" inference can resolve ambiguity and lead to better learning and localisation. (2) Image backgrounds are shared across classes to better learn varying surroundings and "push out" objects of interest. (3) Our model can be learned with a mixture of weakly labelled and unlabelled data, allowing the large volume of unlabelled images on the Internet to be exploited for learning. Moreover, the Bayesian formulation enables the exploitation of various types of prior knowledge to compensate for the limited supervision offered by weakly labelled data, as well as Bayesian domain adaptation for transfer learning. Extensive experiments on the PASCAL VOC, ImageNet and YouTube-Object videos datasets demonstrate the effectiveness of our Bayesian joint model for weakly supervised object localisation.

cs.CV↗

arXiv2017

Transferring a Semantic Representation for Person Re-Identification and Search

Learning semantic attributes for person re-identification and description-based person search has gained increasing interest due to attributes' great potential as a pose and view-invariant representation. However, existing attribute-centric approaches have thus far underperformed state-of-the-art conventional approaches. This is due to their non-scalable need for extensive domain (camera) specific annotation. In this paper we present a new semantic attribute learning approach for person re-identification and search. Our model is trained on existing fashion photography datasets -- either weakly or strongly labelled. It can then be transferred and adapted to provide a powerful semantic description of surveillance person detections, without requiring any surveillance domain supervision. The resulting representation is useful for both unsupervised and supervised person re-identification, achieving state-of-the-art and near state-of-the-art performance respectively. Furthermore, as a semantic representation it allows description-based person search to be integrated within the same framework.

cs.CV↗

arXiv2017

Bayesian Joint Topic Modelling for Weakly Supervised Object Localisation

We address the problem of localisation of objects as bounding boxes in images with weak labels. This weakly supervised object localisation problem has been tackled in the past using discriminative models where each object class is localised independently from other classes. We propose a novel framework based on Bayesian joint topic modelling. Our framework has three distinctive advantages over previous works: (1) All object classes and image backgrounds are modelled jointly together in a single generative model so that "explaining away" inference can resolve ambiguity and lead to better learning and localisation. (2) The Bayesian formulation of the model enables easy integration of prior knowledge about object appearance to compensate for limited supervision. (3) Our model can be learned with a mixture of weakly labelled and unlabelled data, allowing the large volume of unlabelled images on the Internet to be exploited for learning. Extensive experiments on the challenging VOC dataset demonstrate that our approach outperforms the state-of-the-art competitors.

cs.CV↗

arXiv2017

Transfer Learning by Ranking for Weakly Supervised Object Annotation

Most existing approaches to training object detectors rely on fully supervised learning, which requires the tedious manual annotation of object location in a training set. Recently there has been an increasing interest in developing weakly supervised approach to detector training where the object location is not manually annotated but automatically determined based on binary (weak) labels indicating if a training image contains the object. This is a challenging problem because each image can contain many candidate object locations which partially overlaps the object of interest. Existing approaches focus on how to best utilise the binary labels for object location annotation. In this paper we propose to solve this problem from a very different perspective by casting it as a transfer learning problem. Specifically, we formulate a novel transfer learning based on learning to rank, which effectively transfers a model for automatic annotation of object location from an auxiliary dataset to a target dataset with completely unrelated object categories. We show that our approach outperforms existing state-of-the-art weakly supervised approach to annotating objects in the challenging VOC dataset.

cs.CV↗