arXiv subjects

Yifan Xu

Publications and source records attributed to Yifan Xu.

At least 109 records · Page 6Linked to original sources

arXiv2024

AndroidLab: Training and Systematic Benchmarking of Android Autonomous Agents

Autonomous agents have become increasingly important for interacting with the real world. Android agents, in particular, have been recently a frequently-mentioned interaction method. However, existing studies for training and evaluating Android agents lack systematic research on both open-source and closed-source models. In this work, we propose AndroidLab as a systematic Android agent framework. It includes an operation environment with different modalities, action space, and a reproducible benchmark. It supports both large language models (LLMs) and multimodal models (LMMs) in the same action space. AndroidLab benchmark includes predefined Android virtual devices and 138 tasks across nine apps built on these devices. By using the AndroidLab environment, we develop an Android Instruction dataset and train six open-source LLMs and LMMs, lifting the average success rates from 4.59% to 21.50% for LLMs and from 1.93% to 13.28% for LMMs. AndroidLab is open-sourced and publicly available at https://github.com/THUDM/Android-Lab.

cs.AI↗

arXiv2024

CSP-Net: Common Spatial Pattern Empowered Neural Networks for EEG-Based Motor Imagery Classification

Electroencephalogram-based motor imagery (MI) classification is an important paradigm of non-invasive brain-computer interfaces. Common spatial pattern (CSP), which exploits different energy distributions on the scalp while performing different MI tasks, is very popular in MI classification. Convolutional neural networks (CNNs) have also achieved great success, due to their powerful learning capabilities. This paper proposes two CSP-empowered neural networks (CSP-Nets), which integrate knowledge-driven CSP filters with data-driven CNNs to enhance the performance in MI classification. CSP-Net-1 directly adds a CSP layer before a CNN to improve the input discriminability. CSP-Net-2 replaces a convolutional layer in CNN with a CSP layer. The CSP layer parameters in both CSP-Nets are initialized with CSP filters designed from the training data. During training, they can either be kept fixed or optimized using gradient descent. Experiments on four public MI datasets demonstrated that the two CSP-Nets consistently improved over their CNN backbones, in both within-subject and cross-subject classifications. They are particularly useful when the number of training samples is very small. Our work demonstrates the advantage of integrating knowledge-driven traditional machine learning with data-driven deep learning in EEG-based brain-computer interfaces.

eess.SP↗

arXiv2024

AutoGLM: Autonomous Foundation Agents for GUIs

We present AutoGLM, a new series in the ChatGLM family, designed to serve as foundation agents for autonomous control of digital devices through Graphical User Interfaces (GUIs). While foundation models excel at acquiring human knowledge, they often struggle with decision-making in dynamic real-world environments, limiting their progress toward artificial general intelligence. This limitation underscores the importance of developing foundation agents capable of learning through autonomous environmental interactions by reinforcing existing models. Focusing on Web Browser and Phone as representative GUI scenarios, we have developed AutoGLM as a practical foundation agent system for real-world GUI interactions. Our approach integrates a comprehensive suite of techniques and infrastructures to create deployable agent systems suitable for user delivery. Through this development, we have derived two key insights: First, the design of an appropriate "intermediate interface" for GUI control is crucial, enabling the separation of planning and grounding behaviors, which require distinct optimization for flexibility and accuracy respectively. Second, we have developed a novel progressive training framework that enables self-evolving online curriculum reinforcement learning for AutoGLM. Our evaluations demonstrate AutoGLM's effectiveness across multiple domains. For web browsing, AutoGLM achieves a 55.2% success rate on VAB-WebArena-Lite (improving to 59.1% with a second attempt) and 96.2% on OpenTable evaluation tasks. In Android device control, AutoGLM attains a 36.2% success rate on AndroidLab (VAB-Mobile) and 89.7% on common tasks in popular Chinese APPs.

cs.HC↗

arXiv2024

MapCraft: Dissecting and Designing Custom Geo-Infographics

Geographic infographics are increasingly utilized across various domains to convey spatially relevant information effectively. However, creating these infographics typically requires substantial expertise in design and visualization, as well as proficiency with specialized tools, which can deter many potential creators. To address this barrier, our research analyzed and categorized 118 geographic infographics and sketches designed by 8 experts, leading to the development of a structured design space encompassing four critical dimensions: basic map representations, encoding channels, label design and placement, and highlighting techniques. Based on this design space, we developed a web-based authoring tool that allows users to explore and apply these dimensions interactively. The tool's effectiveness was evaluated through a user study involving 12 participants without prior design experience. Participants were first required manually to create geographic infographics using provided datasets, then utilize our authoring tool to recreate and refine their initial drafts. We also conducted pre- and post-use assessments of the participants' knowledge of geographic infographic design. The findings revealed significant improvements in understanding and applying information encoding channels, highlighting techniques, and label design and placement strategies. These results demonstrate the tool's dual capacity to assist users in creating geographics while educating them on key visualization strategies. Our tool, therefore, empowers a broader audience, including those with limited design and visualization backgrounds, to effectively create and utilize geo-infographics.

cs.HC↗

arXiv2024

Enhancing Performance and Scalability of Large-Scale Recommendation Systems with Jagged Flash Attention

The integration of hardware accelerators has significantly advanced the capabilities of modern recommendation systems, enabling the exploration of complex ranking paradigms previously deemed impractical. However, the GPU-based computational costs present substantial challenges. In this paper, we demonstrate our development of an efficiency-driven approach to explore these paradigms, moving beyond traditional reliance on native PyTorch modules. We address the specific challenges posed by ranking models' dependence on categorical features, which vary in length and complicate GPU utilization. We introduce Jagged Feature Interaction Kernels, a novel method designed to extract fine-grained insights from long categorical features through efficient handling of dynamically sized tensors. We further enhance the performance of attention mechanisms by integrating Jagged tensors with Flash Attention. Our novel Jagged Flash Attention achieves up to 9x speedup and 22x memory reduction compared to dense attention. Notably, it also outperforms dense flash attention, with up to 3x speedup and 53% more memory efficiency. In production models, we observe 10% QPS improvement and 18% memory savings, enabling us to scale our recommendation systems with longer features and more complex architectures.

cs.LG↗

arXiv2024

Point2Graph: An End-to-end Point Cloud-based 3D Open-Vocabulary Scene Graph for Robot Navigation

Current open-vocabulary scene graph generation algorithms highly rely on both 3D scene point cloud data and posed RGB-D images and thus have limited applications in scenarios where RGB-D images or camera poses are not readily available. To solve this problem, we propose Point2Graph, a novel end-to-end point cloud-based 3D open-vocabulary scene graph generation framework in which the requirement of posed RGB-D image series is eliminated. This hierarchical framework contains room and object detection/segmentation and open-vocabulary classification. For the room layer, we leverage the advantage of merging the geometry-based border detection algorithm with the learning-based region detection to segment rooms and create a "Snap-Lookup" framework for open-vocabulary room classification. In addition, we create an end-to-end pipeline for the object layer to detect and classify 3D objects based solely on 3D point cloud data. Our evaluation results show that our framework can outperform the current state-of-the-art (SOTA) open-vocabulary object and room segmentation and classification algorithm on widely used real-scene datasets.

cs.RO↗

arXiv2024

First Indication of Solar $^8$B Neutrino Flux through Coherent Elastic Neutrino-Nucleus Scattering in PandaX-4T

The PandaX-4T liquid xenon detector at the China Jinping Underground Laboratory is used to measure the solar $^8$B neutrino flux by detecting neutrinos through coherent scattering with xenon nuclei. Data samples requiring the coincidence of scintillation and ionization signals (paired), as well as unpaired ionization-only signals (US2), are selected with energy threshold of approximately 1.1 keV (0.33 keV) nuclear recoil energy. Combining the commissioning run and the first science run of PandaX-4T, a total exposure of 1.20 and 1.04 tonne$\cdot$year are collected for the paired and US2, respectively. After unblinding, 3 and 332 events are observed with an expectation of 2.8$\pm$0.5 and 251$\pm$32 background events, for the paired and US2 data, respectively. A combined analysis yields a best-fit $^8$B neutrino signal of 3.5 (75) events from the paired (US2) data sample, with $\sim$37\% uncertainty, and the background-only hypothesis is disfavored at 2.64$σ$ significance. This gives a solar $^8$B neutrino flux of ($8.4\pm3.1$)$\times$10$^6$ cm$^{-2}$s$^{-1}$, consistent with the standard solar model prediction. It is also the first indication of solar $^8$B neutrino ``fog'' in a dark matter direct detection experiment.

hep-ex↗

arXiv2024

AlignBench: Benchmarking Chinese Alignment of Large Language Models

Alignment has become a critical step for instruction-tuned Large Language Models (LLMs) to become helpful assistants. However, the effective evaluation of alignment for emerging Chinese LLMs is still largely unexplored. To fill in this gap, we introduce AlignBench, a comprehensive multi-dimensional benchmark for evaluating LLMs' alignment in Chinese. We design a human-in-the-loop data curation pipeline, containing eight main categories, 683 real-scenario rooted queries and corresponding human verified references. To ensure the correctness of references, each knowledge-intensive query is accompanied with evidences collected from reliable web sources (including URLs and quotations) by our annotators. For automatic evaluation, our benchmark employs a rule-calibrated multi-dimensional LLM-as-Judge~\cite{zheng2023judging} approach with Chain-of-Thought to generate explanations and final ratings, ensuring high reliability and interpretability. All evaluation code, data, and LLM generations are available at \url{https://github.com/THUDM/AlignBench}. Since its release, AlignBench has been adopted by top (Chinese) LLMs for evaluating their alignment capabilities in Chinese, including ChatGLM, Qwen, DeepSeek, Yi, Baichuan, and Abab.

cs.CL↗

arXiv2024

VisualAgentBench: Towards Large Multimodal Models as Visual Foundation Agents

Large Multimodal Models (LMMs) have ushered in a new era in artificial intelligence, merging capabilities in both language and vision to form highly capable Visual Foundation Agents. These agents are postulated to excel across a myriad of tasks, potentially approaching general artificial intelligence. However, existing benchmarks fail to sufficiently challenge or showcase the full potential of LMMs in complex, real-world environments. To address this gap, we introduce VisualAgentBench (VAB), a comprehensive and pioneering benchmark specifically designed to train and evaluate LMMs as visual foundation agents across diverse scenarios, including Embodied, Graphical User Interface, and Visual Design, with tasks formulated to probe the depth of LMMs' understanding and interaction capabilities. Through rigorous testing across nine proprietary LMM APIs and eight open models, we demonstrate the considerable yet still developing agent capabilities of these models. Additionally, VAB constructs a trajectory training set constructed through hybrid methods including Program-based Solvers, LMM Agent Bootstrapping, and Human Demonstrations, promoting substantial performance improvements in LMMs through behavior cloning. Our work not only aims to benchmark existing models but also provides a solid foundation for future development into visual foundation agents. Code, train \& test data, and part of fine-tuned open LMMs are available at \url{https://github.com/THUDM/VisualAgentBench}.

cs.AI↗

arXiv2024

ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools

We introduce ChatGLM, an evolving family of large language models that we have been developing over time. This report primarily focuses on the GLM-4 language series, which includes GLM-4, GLM-4-Air, and GLM-4-9B. They represent our most capable models that are trained with all the insights and lessons gained from the preceding three generations of ChatGLM. To date, the GLM-4 models are pre-trained on ten trillions of tokens mostly in Chinese and English, along with a small set of corpus from 24 languages, and aligned primarily for Chinese and English usage. The high-quality alignment is achieved via a multi-stage post-training process, which involves supervised fine-tuning and learning from human feedback. Evaluations show that GLM-4 1) closely rivals or outperforms GPT-4 in terms of general metrics such as MMLU, GSM8K, MATH, BBH, GPQA, and HumanEval, 2) gets close to GPT-4-Turbo in instruction following as measured by IFEval, 3) matches GPT-4 Turbo (128K) and Claude 3 for long context tasks, and 4) outperforms GPT-4 in Chinese alignments as measured by AlignBench. The GLM-4 All Tools model is further aligned to understand user intent and autonomously decide when and which tool(s) touse -- including web browser, Python interpreter, text-to-image model, and user-defined functions -- to effectively complete complex tasks. In practical applications, it matches and even surpasses GPT-4 All Tools in tasks like accessing online information via web browsing and solving math problems using Python interpreter. Over the course, we have open-sourced a series of models, including ChatGLM-6B (three generations), GLM-4-9B (128K, 1M), GLM-4V-9B, WebGLM, and CodeGeeX, attracting over 10 million downloads on Hugging face in the year 2023 alone. The open models can be accessed through https://github.com/THUDM and https://huggingface.co/THUDM.

cs.CL↗

arXiv2024

Async Learned User Embeddings for Ads Delivery Optimization

In recommendation systems, high-quality user embeddings can capture subtle preferences, enable precise similarity calculations, and adapt to changing preferences over time to maintain relevance. The effectiveness of recommendation systems depends on the quality of user embedding. We propose to asynchronously learn high fidelity user embeddings for billions of users each day from sequence based multimodal user activities through a Transformer-like large scale feature learning module. The async learned user representations embeddings (ALURE) are further converted to user similarity graphs through graph learning and then combined with user realtime activities to retrieval highly related ads candidates for the ads delivery system. Our method shows significant gains in both offline and online experiments.

cs.IR↗

arXiv2024

Banal Deception Human-AI Ecosystems: A Study of People's Perceptions of LLM-generated Deceptive Behaviour

Large language models (LLMs) can provide users with false, inaccurate, or misleading information, and we consider the output of this type of information as what Natale (2021) calls `banal' deceptive behaviour. Here, we investigate peoples' perceptions of ChatGPT-generated deceptive behaviour and how this affects peoples' own behaviour and trust. To do this, we use a mixed-methods approach comprising of (i) an online survey with 220 participants and (ii) semi-structured interviews with 12 participants. Our results show that (i) the most common types of deceptive information encountered were over-simplifications and outdated information; (ii) humans' perceptions of trust and `worthiness' of talking to ChatGPT are impacted by `banal' deceptive behaviour; (iii) the perceived responsibility for deception is influenced by education level and the frequency of deceptive information; and (iv) users become more cautious after encountering deceptive information, but they come to trust the technology more when they identify advantages of using it. Our findings contribute to the understanding of human-AI interaction dynamics in the context of \textit{Deceptive AI Ecosystems}, and highlight the importance of user-centric approaches to mitigating the potential harms of deceptive AI technologies.

cs.CY↗

arXiv2024

Socially-Aware Shared Control Navigation for Assistive Mobile Robots in the Built Environment

As the number of Persons with Disabilities (PWD), particularly those with one or more physical impairments, increases, there is an increasing demand for assistive robotic technologies that can support independent mobility in the built environment and reduce the burden on caregivers. Current assistive mobility platforms (e.g., robotic wheelchairs) often fail to incorporate user preferences and control, leading to reduced trust and efficiency. Existing shared control algorithms do not allow the incorporation of the user control preferences inside the navigation framework or the path planning algorithm. In addition, existing dynamic local planner algorithms for robotic wheelchairs do not take into account the social spaces of people, potentially leading such platforms to infringe upon these areas and cause discomfort. To address these concerns, this work introduces a novel socially-aware shared autonomy-based navigation system for assistive mobile robotic platforms. Our navigation framework comprises a Global Planner and a Local Planner. To implement the Global Planner, the proposed approach introduces a novel User Preference Field (UPF) theory within its global planning framework, explicitly acknowledging user preferences to adeptly navigate away from congested areas. For the Local Planner, we propose a Socially-aware Shared Control-based Model Predictive Control with Dynamic Control Barrier Function (SS-MPC-DCBF) to adjust movements in real-time, integrating user preferences for safer, more autonomous navigation. Evaluation results show that our Global Planner aligns closely with user preferences compared to baselines, and our Local Planner demonstrates enhanced safety and efficiency in dynamic and static scenarios. This integrated approach fosters trust and autonomy, crucial for the acceptance of assistive mobility technologies in the built environment.

cs.RO↗

arXiv2024

Libra: Building Decoupled Vision System on Large Language Models

In this work, we introduce Libra, a prototype model with a decoupled vision system on a large language model (LLM). The decoupled vision system decouples inner-modal modeling and cross-modal interaction, yielding unique visual information modeling and effective cross-modal comprehension. Libra is trained through discrete auto-regressive modeling on both vision and language inputs. Specifically, we incorporate a routed visual expert with a cross-modal bridge module into a pretrained LLM to route the vision and language flows during attention computing to enable different attention patterns in inner-modal modeling and cross-modal interaction scenarios. Experimental results demonstrate that the dedicated design of Libra achieves a strong MLLM baseline that rivals existing works in the image-to-text scenario with merely 50 million training data, providing a new perspective for future multimodal foundation models. Code is available at https://github.com/YifanXu74/Libra.

cs.CV↗

arXiv2024

A Cause-Effect Look at Alleviating Hallucination of Knowledge-grounded Dialogue Generation

Empowered by the large-scale pretrained language models, existing dialogue systems have demonstrated impressive performance conducting fluent and natural-sounding conversations. However, they are still plagued by the hallucination problem, causing unpredictable factual errors in the generated responses. Recently, knowledge-grounded dialogue generation models, that intentionally invoke external knowledge resources to more informative responses, are also proven to be effective in reducing hallucination. Following the idea of getting high-quality knowledge, a few efforts have achieved pretty good performance on this issue. As some inevitable knowledge noises may also lead to hallucinations, it is emergent to investigate the reason and future directions for building noise-tolerant methods in KGD tasks. In this paper, we analyze the causal story behind this problem with counterfactual reasoning methods. Based on the causal effect analysis, we propose a possible solution for alleviating the hallucination in KGD by exploiting the dialogue-knowledge interaction. Experimental results of our example implementation show that this method can reduce hallucination without disrupting other dialogue performance, while keeping adaptive to different generation models. We hope our efforts can support and call for more attention to developing lightweight techniques towards robust and trusty dialogue systems.

cs.CL↗

arXiv2024

ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline

Large language models (LLMs) have shown excellent mastering of human language, but still struggle in real-world applications that require mathematical problem-solving. While many strategies and datasets to enhance LLMs' mathematics are developed, it remains a challenge to simultaneously maintain and improve both language and mathematical capabilities in deployed LLM systems.In this work, we tailor the Self-Critique pipeline, which addresses the challenge in the feedback learning stage of LLM alignment. We first train a general Math-Critique model from the LLM itself to provide feedback signals. Then, we sequentially employ rejective fine-tuning and direct preference optimization over the LLM's own generations for data collection. Based on ChatGLM3-32B, we conduct a series of experiments on both academic and our newly created challenging dataset, MathUserEval. Results show that our pipeline significantly enhances the LLM's mathematical problem-solving while still improving its language ability, outperforming LLMs that could be two times larger. Related techniques have been deployed to ChatGLM\footnote{\url{https://chatglm.cn}}, an online serving LLM. Related evaluation dataset and scripts are released at \url{https://github.com/THUDM/ChatGLM-Math}.

cs.CL↗

arXiv2024

Spike-driven Transformer V2: Meta Spiking Neural Network Architecture Inspiring the Design of Next-generation Neuromorphic Chips

Neuromorphic computing, which exploits Spiking Neural Networks (SNNs) on neuromorphic chips, is a promising energy-efficient alternative to traditional AI. CNN-based SNNs are the current mainstream of neuromorphic computing. By contrast, no neuromorphic chips are designed especially for Transformer-based SNNs, which have just emerged, and their performance is only on par with CNN-based SNNs, offering no distinct advantage. In this work, we propose a general Transformer-based SNN architecture, termed as ``Meta-SpikeFormer", whose goals are: 1) Lower-power, supports the spike-driven paradigm that there is only sparse addition in the network; 2) Versatility, handles various vision tasks; 3) High-performance, shows overwhelming performance advantages over CNN-based SNNs; 4) Meta-architecture, provides inspiration for future next-generation Transformer-based neuromorphic chip designs. Specifically, we extend the Spike-driven Transformer in \citet{yao2023spike} into a meta architecture, and explore the impact of structure, spike-driven self-attention, and skip connection on its performance. On ImageNet-1K, Meta-SpikeFormer achieves 80.0\% top-1 accuracy (55M), surpassing the current state-of-the-art (SOTA) SNN baselines (66M) by 3.7\%. This is the first direct training SNN backbone that can simultaneously supports classification, detection, and segmentation, obtaining SOTA results in SNNs. Finally, we discuss the inspiration of the meta SNN architecture for neuromorphic chip design. Source code and models are available at \url{https://github.com/BICLab/Spike-Driven-Transformer-V2}.

cs.NE↗

arXiv2024

3DMIT: 3D Multi-modal Instruction Tuning for Scene Understanding

The remarkable potential of multi-modal large language models (MLLMs) in comprehending both vision and language information has been widely acknowledged. However, the scarcity of 3D scenes-language pairs in comparison to their 2D counterparts, coupled with the inadequacy of existing approaches in understanding of 3D scenes by LLMs, poses a significant challenge. In response, we collect and construct an extensive dataset comprising 75K instruction-response pairs tailored for 3D scenes. This dataset addresses tasks related to 3D VQA, 3D grounding, and 3D conversation. To further enhance the integration of 3D spatial information into LLMs, we introduce a novel and efficient prompt tuning paradigm, 3DMIT. This paradigm eliminates the alignment stage between 3D scenes and language and extends the instruction prompt with the 3D modality information including the entire scene and segmented objects. We evaluate the effectiveness of our method across diverse tasks in the 3D scene domain and find that our approach serves as a strategic means to enrich LLMs' comprehension of the 3D world. Our code is available at https://github.com/staymylove/3DMIT.

cs.CV↗