arXiv subjects

Zhiyuan Yao

Publications and source records attributed to Zhiyuan Yao.

At least 55 records · Page 3Linked to original sources

arXiv2025

From FFB Starbursts at Cosmic Dawn to Quenching at Cosmic Morning: Hi-z Galaxy Bimodality

We propose a mass-dependent bimodality in the early evolution of galaxies. The massive track connects the super-bright galaxies at cosmic dawn ($z > 8$) to the super-massive quiescent galaxies and black holes (BHs) at cosmic morning ($z \sim 4 - 7$). The dark-matter halos $> 10^{10.5} {\rm M}_\odot$ at $z = 10$ are expected to undergo feedback-free starbursts (FFB) with high star-formation efficiency in dense star clusters within compact galaxies. The less massive halos avoid FFB and form stars gradually under stellar feedback, possibly leading to the peak star-forming galaxies at cosmic noon ($z \sim 1-3$). The FFB and non-FFB halos originate from $>4σ$ and $2-3σ$ density peaks, respectively. The post-FFB galaxies quench their star formation soon after the FFB phase and remain quiescent due to (a) gas depletion by the FFB starbursts and outflows, (b) compaction events driven by angular-momentum loss in colliding streams within the high-sigma-peak FFB halos, (c) turbulent circum-galactic medium (CGM) that suppresses feeding by cold streams, and (d) BH feedback, being a key for complete quenching. BH feedback is enhanced by FFB-driven BH seeding and growth. It seems capable of disrupting the streams by generating CGM turbulence or photo-heating, but this remains an open challenge. The cosmic-morning quiescent galaxies are expected to be massive, compact, showing signatures of compaction, outflows and AGN, with a comoving number density $\sim 10^{-5} {\rm Mpc}^{-3}$, comparable to the super-bright galaxies at cosmic dawn and the AGN at cosmic morning. Their UV luminosity function is predicted to peak about $M_ {\rm uv} \sim -22$ and contribute $\sim 10\%$ of the galaxies there.

astro-ph.GA↗

arXiv2025

Quantum many-body scars through the lens of correlation matrix

Quantum many-body scars (QMBS) -- rare eigenstates that evade thermalization -- are typically characterized by their low entanglement entropies compared to surrounding thermal eigenstates. However, due to finite-size effects in systems accessible via exact diagonalization, this measure can be ambiguous. To address this limitation, we propose using the correlation matrix spectrum as an alternative probe to identify QMBS. In cases of exact QMBS that either have known analytic expressions or are captured by various frameworks of QMBS, we find that the dimensionality of the null space of the correlation matrix -- an integer value, and thus immune to finite-size effects -- can qualitatively identify QMBS. Beyond serving as a diagnostic tool, the correlation matrix method enables the manipulation of the QMBS subspace. For approximate QMBS, such as those in the PXP model, we observe that the correlation matrix spectrum features numerous approximate zero eigenvalues, thereby distinguishing these states. We demonstrate the effectiveness and utility of this method with several paradigmatic QMBS examples.

cond-mat.quant-gas↗

arXiv2025

BitsAI-Fix: LLM-Driven Approach for Automated Lint Error Resolution in Practice

As enterprise codebases continue to grow in scale and complexity, the volume of lint errors far exceeds engineers' manual remediation capacity, leading to continuous accumulation of technical debt and hindered development efficiency. This paper presents BitsAI-Fix, an automated lint error remediation workflow based on Large Language Models (LLMs), designed to address this critical challenge in industrial-scale environments. BitsAI-Fix employs tree-sitter for context expansion and generates search-and-replace format patches through specially trained LLMs, followed by lint scan re-verification to output final remediation results. Additionally, our approach introduces an innovative progressive reinforcement learning (RL) training strategy that can automatically acquire verifiable training data during the project cold-start phase and continuously iterate the model by collecting online samples through feedback after system deployment. Furthermore, we designed a targeted rule-based reward mechanism that combines format rewards and correctness rewards while penalizing redundant modifications. We also propose a "code diff matching" methodology to continuously track online effectiveness. In production deployment at ByteDance, our solution has supported over 5,000 engineers, resolved more than 12,000 static analysis issues, achieved approximately 85% remediation accuracy, with around 1,000 weekly active adopters. This work demonstrates the practical feasibility of LLM-based code remediation solutions in enterprise environments and serves as a reference for automated code fix in large-scale industrial scenarios.

cs.SE↗

arXiv2025

Open-FinLLMs: Open Multimodal Large Language Models for Financial Applications

Financial LLMs hold promise for advancing financial tasks and domain-specific applications. However, they are limited by scarce corpora, weak multimodal capabilities, and narrow evaluations, making them less suited for real-world application. To address this, we introduce \textit{Open-FinLLMs}, the first open-source multimodal financial LLMs designed to handle diverse tasks across text, tabular, time-series, and chart data, excelling in zero-shot, few-shot, and fine-tuning settings. The suite includes FinLLaMA, pre-trained on a comprehensive 52-billion-token corpus; FinLLaMA-Instruct, fine-tuned with 573K financial instructions; and FinLLaVA, enhanced with 1.43M multimodal tuning pairs for strong cross-modal reasoning. We comprehensively evaluate Open-FinLLMs across 14 financial tasks, 30 datasets, and 4 multimodal tasks in zero-shot, few-shot, and supervised fine-tuning settings, introducing two new multimodal evaluation datasets. Our results show that Open-FinLLMs outperforms afvanced financial and general LLMs such as GPT-4, across financial NLP, decision-making, and multi-modal tasks, highlighting their potential to tackle real-world challenges. To foster innovation and collaboration across academia and industry, we release all codes (https://anonymous.4open.science/r/PIXIU2-0D70/B1D7/LICENSE) and models under OSI-approved licenses.

cs.CL↗

arXiv2024

Collaborative Inference for Large Models with Task Offloading and Early Exiting

In 5G smart cities, edge computing is employed to provide nearby computing services for end devices, and the large-scale models (e.g., GPT and LLaMA) can be deployed at the network edge to boost the service quality. However, due to the constraints of memory size and computing capacity, it is difficult to run these large-scale models on a single edge node. To meet the resource constraints, a large-scale model can be partitioned into multiple sub-models and deployed across multiple edge nodes. Then tasks are offloaded to the edge nodes for collaborative inference. Additionally, we incorporate the early exit mechanism to further accelerate inference. However, the heterogeneous system and dynamic environment will significantly affect the inference efficiency. To address these challenges, we theoretically analyze the coupled relationship between task offloading strategy and confidence thresholds, and develop a distributed algorithm, termed DTO-EE, based on the coupled relationship and convex optimization. DTO-EE enables each edge node to jointly optimize its offloading strategy and the confidence threshold, so as to achieve a promising trade-off between response delay and inference accuracy. The experimental results show that DTO-EE can reduce the average response delay by 21%-41% and improve the inference accuracy by 1%-4%, compared to the baselines.

cs.DC↗

arXiv2024

Effects of Cloud Geometry and Metallicity on Shattering and Coagulation of Cold Gas, and Implications for Cold Streams Penetrating Virial Shocks

Theory and observations reveal that the circumgalactic medium (CGM) and the cosmic web at high redshifts are multiphase, with small clouds of cold gas embedded in a hot, diffuse medium. A proposed mechanism is `shattering' of large, thermally unstable clouds into tiny cloudlets of size lshatter~min(cs*tcool). We study these processes using idealized numerical simulations of thermally unstable gas clouds. We expand upon previous works by exploring the effects of cloud geometry (spheres, streams, and sheets), metallicity, and the inclusion of an ionizing UV background. We find that `shattering' is triggered by clouds losing sonic contact and rapidly imploding, leading to a reflected shock which causes the cloud to re-expand and induces Richtmyer-Meshkov instabilities at its interface. After fragmentation the cloudlets experience a drag force from the surrounding hot gas, leading to recoagulation into larger clouds. We distinguish between `fast' and `slow' coagulation regimes, showing that sheets are always in the `fast' coagulation regime while streams and spheres have a maximum overdensity for rapid coagulation. The critical overdensity for spheres is smaller than for streams, such that the coagulation efficiency increases from spheres to streams to sheets. Surprisingly, lshatter does not appear to be a characteristic clump size even if it is well resolved. Rather, fragmentation continues until the grid scale with a mass distribution of N(>m)~m^{-1}. We apply our results to the case of cold streams feeding massive (Mv>10^{12}Msun) high-z (z>2) galaxies from the cosmic web, finding that streams are likely to shatter upon entering the CGM through the virial shock. This could explain the large clumping factors and covering fractions of cold gas in the CGM around such galaxies, and may be related to galaxy quenching by preventing cold streams from reaching the central galaxy. [abridged]

astro-ph.GA↗

arXiv2024

FinCon: A Synthesized LLM Multi-Agent System with Conceptual Verbal Reinforcement for Enhanced Financial Decision Making

Large language models (LLMs) have demonstrated notable potential in conducting complex tasks and are increasingly utilized in various financial applications. However, high-quality sequential financial investment decision-making remains challenging. These tasks require multiple interactions with a volatile environment for every decision, demanding sufficient intelligence to maximize returns and manage risks. Although LLMs have been used to develop agent systems that surpass human teams and yield impressive investment returns, opportunities to enhance multi-sourced information synthesis and optimize decision-making outcomes through timely experience refinement remain unexplored. Here, we introduce the FinCon, an LLM-based multi-agent framework with CONceptual verbal reinforcement tailored for diverse FINancial tasks. Inspired by effective real-world investment firm organizational structures, FinCon utilizes a manager-analyst communication hierarchy. This structure allows for synchronized cross-functional agent collaboration towards unified goals through natural language interactions and equips each agent with greater memory capacity than humans. Additionally, a risk-control component in FinCon enhances decision quality by episodically initiating a self-critiquing mechanism to update systematic investment beliefs. The conceptualized beliefs serve as verbal reinforcement for the future agent's behavior and can be selectively propagated to the appropriate node that requires knowledge updates. This feature significantly improves performance while reducing unnecessary peer-to-peer communication costs. Moreover, FinCon demonstrates strong generalization capabilities in various financial tasks, including single stock trading and portfolio management.

cs.CL↗

arXiv2024

CatMemo at the FinLLM Challenge Task: Fine-Tuning Large Language Models using Data Fusion in Financial Applications

The integration of Large Language Models (LLMs) into financial analysis has garnered significant attention in the NLP community. This paper presents our solution to IJCAI-2024 FinLLM challenge, investigating the capabilities of LLMs within three critical areas of financial tasks: financial classification, financial text summarization, and single stock trading. We adopted Llama3-8B and Mistral-7B as base models, fine-tuning them through Parameter Efficient Fine-Tuning (PEFT) and Low-Rank Adaptation (LoRA) approaches. To enhance model performance, we combine datasets from task 1 and task 2 for data fusion. Our approach aims to tackle these diverse tasks in a comprehensive and integrated manner, showcasing LLMs' capacity to address diverse and complex financial tasks with improved accuracy and decision-making capabilities.

cs.CE↗

arXiv2024

FinBen: A Holistic Financial Benchmark for Large Language Models

LLMs have transformed NLP and shown promise in various fields, yet their potential in finance is underexplored due to a lack of comprehensive evaluation benchmarks, the rapid development of LLMs, and the complexity of financial tasks. In this paper, we introduce FinBen, the first extensive open-source evaluation benchmark, including 36 datasets spanning 24 financial tasks, covering seven critical aspects: information extraction (IE), textual analysis, question answering (QA), text generation, risk management, forecasting, and decision-making. FinBen offers several key innovations: a broader range of tasks and datasets, the first evaluation of stock trading, novel agent and Retrieval-Augmented Generation (RAG) evaluation, and three novel open-source evaluation datasets for text summarization, question answering, and stock trading. Our evaluation of 15 representative LLMs, including GPT-4, ChatGPT, and the latest Gemini, reveals several key findings: While LLMs excel in IE and textual analysis, they struggle with advanced reasoning and complex tasks like text generation and forecasting. GPT-4 excels in IE and stock trading, while Gemini is better at text generation and forecasting. Instruction-tuned LLMs improve textual analysis but offer limited benefits for complex tasks such as QA. FinBen has been used to host the first financial LLMs shared task at the FinNLP-AgentScen workshop during IJCAI-2024, attracting 12 teams. Their novel solutions outperformed GPT-4, showcasing FinBen's potential to drive innovation in financial LLMs. All datasets, results, and codes are released for the research community: https://github.com/The-FinAI/PIXIU.

cs.CL↗

arXiv2024

Reinforcement Learning in Agent-Based Market Simulation: Unveiling Realistic Stylized Facts and Behavior

Investors and regulators can greatly benefit from a realistic market simulator that enables them to anticipate the consequences of their decisions in real markets. However, traditional rule-based market simulators often fall short in accurately capturing the dynamic behavior of market participants, particularly in response to external market impact events or changes in the behavior of other participants. In this study, we explore an agent-based simulation framework employing reinforcement learning (RL) agents. We present the implementation details of these RL agents and demonstrate that the simulated market exhibits realistic stylized facts observed in real-world markets. Furthermore, we investigate the behavior of RL agents when confronted with external market impacts, such as a flash crash. Our findings shed light on the effectiveness and adaptability of RL-based agents within the simulation, offering insights into their response to significant market events.

q-fin.TR↗

arXiv2024

Control in Stochastic Environment with Delays: A Model-based Reinforcement Learning Approach

In this paper we are introducing a new reinforcement learning method for control problems in environments with delayed feedback. Specifically, our method employs stochastic planning, versus previous methods that used deterministic planning. This allows us to embed risk preference in the policy optimization problem. We show that this formulation can recover the optimal policy for problems with deterministic transitions. We contrast our policy with two prior methods from literature. We apply the methodology to simple tasks to understand its features. Then, we compare the performance of the methods in controlling multiple Atari games.

cs.LG↗

arXiv2024

Develop End-to-End Anomaly Detection System

Anomaly detection plays a crucial role in ensuring network robustness. However, implementing intelligent alerting systems becomes a challenge when considering scenarios in which anomalies can be caused by both malicious and non-malicious events, leading to the difficulty of determining anomaly patterns. The lack of labeled data in the computer networking domain further exacerbates this issue, impeding the development of robust models capable of handling real-world scenarios. To address this challenge, in this paper, we propose an end-to-end anomaly detection model development pipeline. This framework makes it possible to consume user feedback and enable continuous user-centric model performance evaluation and optimization. We demonstrate the efficacy of the framework by way of introducing and bench-marking a new forecasting model -- named \emph{Lachesis} -- on a real-world networking problem. Experiments have demonstrated the robustness and effectiveness of the two proposed versions of \emph{Lachesis} compared with other models proposed in the literature. Our findings underscore the potential for improving the performance of data-driven products over their life cycles through a harmonized integration of user feedback and iterative development.

cs.NI↗

arXiv2023

A Hybrid Approach for Smart Alert Generation

Anomaly detection is an important task in network management. However, deploying intelligent alert systems in real-world large-scale networking systems is challenging when we take into account (i) scalability, (ii) data heterogeneity, and (iii) generalizability and maintainability. In this paper, we propose a hybrid model for an alert system that combines statistical models with a whitelist mechanism to tackle these challenges and reduce false positive alerts. The statistical models take advantage of a large database to detect anomalies in time-series data, while the whitelist filters out persistently alerted nodes to further reduce false positives. Our model is validated using qualitative data from customer support cases. Future work includes more feature engineering and input data, as well as including human feedback in the model development process.

cs.NI↗

arXiv2023

Probing quantum many-body correlations by universal ramping dynamics

Ramping a physical parameter is one of the most common experimental protocols in studying a quantum system, and ramping dynamics has been widely used in preparing a quantum state and probing physical properties. Here, we present a novel method of probing quantum many-body correlation by ramping dynamics. We ramp a Hamiltonian parameter to the same target value from different initial values and with different velocities, and we show that the first-order correction on the finite ramping velocity is universal and path-independent, revealing a novel quantum many-body correlation function of the equilibrium phases at the target values. We term this method as the non-adiabatic linear response since this is the leading order correction beyond the adiabatic limit. We demonstrate this method experimentally by studying the Bose-Hubbard model with ultracold atoms in three-dimensional optical lattices. Unlike the conventional linear response that reveals whether the quasi-particle dispersion of a quantum phase is gapped or gapless, this probe is more sensitive to whether the quasi-particle lifetime is long enough such that the quantum phase possesses a well-defined quasi-particle description. In the Bose-Hubbard model, this non-adiabatic linear response is significant in the quantum critical regime where well-defined quasi-particles are absent. And in contrast, this response is vanishingly small in both superfluid and Mott insulators which possess well-defined quasi-particles. Because our proposal uses the most common experimental protocol, we envision that our method can find broad applications in probing various quantum systems.

cond-mat.quant-gas↗

arXiv2022

Bounding Entanglement Entropy Using Zeros of Local Correlation Matrices

Correlation functions and entanglement are two different aspects to characterize quantum many-body states. While many correlation functions are experimentally accessible, entanglement entropy (EE), the simplest characterization of quantum entanglement, is usually difficult to measure. In this Letter, we propose a protocol to bound EE by local measurements. This protocol utilizes local correlation matrices and focuses on their (approximate) zero eigenvalues. Given a quantum state, each (approximate) zero eigenvalue can be used to define a set of local projection operators. An auxiliary Hamiltonian can then be constructed by summing these projectors. When the construction only involves projectors of zero eigenvalues, we prove the EE of a subsystem is bounded by the ground-state degeneracy of the auxiliary Hamiltonian on this subsystem. When projectors from nonzero eigenvalues are included, we show the EE can be bounded by a thermal entropy of the subsystem. Our protocol can be applied experimentally to investigate exotic quantum many-body states prepared in quantum simulators.

cond-mat.str-el↗

arXiv2022

Learning Distributed and Fair Policies for Network Load Balancing as Markov Potential Game

This paper investigates the network load balancing problem in data centers (DCs) where multiple load balancers (LBs) are deployed, using the multi-agent reinforcement learning (MARL) framework. The challenges of this problem consist of the heterogeneous processing architecture and dynamic environments, as well as limited and partial observability of each LB agent in distributed networking systems, which can largely degrade the performance of in-production load balancing algorithms in real-world setups. Centralised-training-decentralised-execution (CTDE) RL scheme has been proposed to improve MARL performance, yet it incurs -- especially in distributed networking systems, which prefer distributed and plug-and-play design scheme -- additional communication and management overhead among agents. We formulate the multi-agent load balancing problem as a Markov potential game, with a carefully and properly designed workload distribution fairness as the potential function. A fully distributed MARL algorithm is proposed to approximate the Nash equilibrium of the game. Experimental evaluations involve both an event-driven simulator and real-world system, where the proposed MARL load balancing algorithm shows close-to-optimal performance in simulations, and superior results over in-production LBs in the real-world system.

cs.AI↗

arXiv2022

Efficient Data-Driven Network Functions

Cloud environments require dynamic and adaptive networking policies. It is preferred to use heuristics over advanced learning algorithms in Virtual Network Functions (VNFs) in production becuase of high-performance constraints. This paper proposes Aquarius to passively yet efficiently gather observations and enable the use of machine learning to collect, infer, and supply accurate networking state information-without incurring additional signalling and management overhead. This paper illustrates the use of Aquarius with a traffic classifier, an autoscaling system, and a load balancer-and demonstrates the use of three different machine learning paradigms-unsupervised, supervised, and reinforcement learning, within Aquarius, for inferring network state. Testbed evaluations show that Aquarius increases network state visibility and brings notable performance gains with low overhead.

cs.NI↗

arXiv2022

Multi-Agent Reinforcement Learning for Network Load Balancing in Data Center

This paper presents the network load balancing problem, a challenging real-world task for multi-agent reinforcement learning (MARL) methods. Traditional heuristic solutions like Weighted-Cost Multi-Path (WCMP) and Local Shortest Queue (LSQ) are less flexible to the changing workload distributions and arrival rates, with a poor balance among multiple load balancers. The cooperative network load balancing task is formulated as a Dec-POMDP problem, which naturally induces the MARL methods. To bridge the reality gap for applying learning-based methods, all methods are directly trained and evaluated on an emulation system from moderate-to large-scale. Experiments on realistic testbeds show that the independent and "selfish" load balancing strategies are not necessarily the globally optimal ones, while the proposed MARL solution has a superior performance over different realistic settings. Additionally, the potential difficulties of MARL methods for network load balancing are analysed, which helps to draw the attention of the learning and network communities to such challenges.

cs.DC↗