arXiv subjects

Lirong Deng

Publications and source records attributed to Lirong Deng.

11 recordsLinked to original sources

arXiv2025

PictOBI-20k: Unveiling Large Multimodal Models in Visual Decipherment for Pictographic Oracle Bone Characters

Deciphering oracle bone characters (OBCs), the oldest attested form of written Chinese, has remained the ultimate, unwavering goal of scholars, offering an irreplaceable key to understanding humanity's early modes of production. Current decipherment methodologies of OBC are primarily constrained by the sporadic nature of archaeological excavations and the limited corpus of inscriptions. With the powerful visual perception capability of large multimodal models (LMMs), the potential of using LMMs for visually deciphering OBCs has increased. In this paper, we introduce PictOBI-20k, a dataset designed to evaluate LMMs on the visual decipherment tasks of pictographic OBCs. It includes 20k meticulously collected OBC and real object images, forming over 15k multi-choice questions. We also conduct subjective annotations to investigate the consistency of the reference point between humans and LMMs in visual reasoning. Experiments indicate that general LMMs possess preliminary visual decipherment skills, and LMMs are not effectively using visual information, while most of the time they are limited by language priors. We hope that our dataset can facilitate the evaluation and optimization of visual attention in future OBC-oriented LMMs. The code and dataset will be available at https://github.com/OBI-Future/PictOBI-20k.

cs.CV↗

arXiv2025

Find Them All: Unveiling MLLMs for Versatile Person Re-identification

Person re-identification (ReID) aims to retrieve images of a target person from the gallery set, with wide applications in medical rehabilitation and public security. However, traditional person ReID models are typically uni-modal, resulting in limited generalizability across heterogeneous data modalities. Recently, the emergence of multi-modal large language models (MLLMs) has shown a promising avenue for addressing this issue. Despite this potential, existing methods merely regard MLLMs as feature extractors or caption generators, leaving their capabilities in person ReID tasks largely unexplored. To bridge this gap, we introduce a novel benchmark for \underline{\textbf{V}}ersatile \underline{\textbf{P}}erson \underline{\textbf{Re}}-\underline{\textbf{ID}}entification, termed VP-ReID. The benchmark includes 257,310 multi-modal queries and gallery images, covering ten diverse person ReID tasks. In addition, we propose two task-oriented evaluation schemes for MLLM-based person ReID. Extensive experiments demonstrate the impressive versatility, effectiveness, and interpretability of MLLMs in various person ReID tasks. Nevertheless, they also have limitations in handling a few modalities, particularly thermal and infrared data. We hope that VP-ReID can facilitate the community in developing more robust and generalizable cross-modal foundation models for person ReID.

cs.CV↗

arXiv2025

Can Large Models Fool the Eye? A New Turing Test for Biological Animation

Evaluating the abilities of large models and manifesting their gaps are challenging. Current benchmarks adopt either ground-truth-based score-form evaluation on static datasets or indistinct textual chatbot-style human preferences collection, which may not provide users with immediate, intuitive, and perceptible feedback on performance differences. In this paper, we introduce BioMotion Arena, a novel framework for evaluating large language models (LLMs) and multimodal large language models (MLLMs) via visual animation. Our methodology draws inspiration from the inherent visual perception of motion patterns characteristic of living organisms that utilizes point-light source imaging to amplify the performance discrepancies between models. Specifically, we employ a pairwise comparison evaluation and collect more than 45k votes for 53 mainstream LLMs and MLLMs on 90 biological motion variants. Data analyses show that the crowd-sourced human votes are in good agreement with those of expert raters, demonstrating the superiority of our BioMotion Arena in offering discriminative feedback. We also find that over 90\% of evaluated models, including the cutting-edge open-source InternVL3 and proprietary Claude-4 series, fail to produce fundamental humanoid point-light groups, much less smooth and biologically plausible motions. This enables BioMotion Arena to serve as a challenging benchmark for performance visualization and a flexible evaluation framework without restrictions on ground-truth.

cs.CV↗

arXiv2025

Supersaturation of odd linear cycles

An $r$-uniform linear cycle of length $\ell$, denoted by $C^r_{\ell}$, is an $r$-graph with $\ell$ edges $e_1,e_2,\dots,e_{\ell}$ where $e_i=\{v_{(r-1)(i-1)},v_{(r-1)(i-1)+1},\dots,v_{(r-1)i}\}$ (here $v_0=v_{(r-1)\ell}$). For $0<\delta<1$ and $n$ sufficiently large, we show that every $n$-vertex $r$-graph $G$ with $n^{r-\delta}$ edges contains at least $n^{(r-1)(2\ell+1)-\delta(2\ell+1+\frac{4\ell-1}{(r-1)(2\ell+1)-3})-o(1)}$ copies of $C^r_{2\ell+1}$. Further, conditioning on the existence of dense high-girth hypergraphs, we show that there exists $n$-vertex $r$-graphs with $n^{r-\delta}$ edges and at most $n^{(r-1)(2\ell+1)-\delta(2\ell+1+\frac{1}{(r-1)\ell-1})+o(1)}$ copies of $C^r_{2\ell+1}$.

math.CO↗

arXiv2024

Residual Deep Reinforcement Learning for Inverter-based Volt-Var Control

A residual deep reinforcement learning (RDRL) approach is proposed by integrating DRL with model-based optimization for inverter-based volt-var control in active distribution networks when the accurate power flow model is unknown. RDRL learns a residual action with a reduced residual action space, based on the action of the model-based approach with an approximate model. RDRL inherits the control capability of the approximate-model-based optimization and enhances the policy optimization capability by residual policy learning. Additionally, it improves the approximation accuracy of the critic and reduces the search difficulties of the actor by reducing residual action space. To address the issues of "too small" or "too large" residual action space of RDRL and further improve the optimization performance, we extend RDRL to a boosting RDRL approach. It selects a much smaller residual action space and learns a residual policy by using the policy of RDRL as a base policy. Simulations demonstrate that RDRL and boosting RDRL improve the optimization performance considerably throughout the learning stage and verify their rationales point-by-point, including 1) inheriting the capability of the approximate model-based optimization, 2) residual policy learning, and 3) learning in a reduced action space.

eess.SY↗

arXiv2022

Reducing Action Space: Reference-Model-Assisted Deep Reinforcement Learning for Inverter-based Volt-Var Control

Reference-model-assisted deep reinforcement learning (DRL) for inverter-based Volt-Var Control (IB-VVC) in active distribution networks is proposed. We investigate that a large action space increases the learning difficulties of DRL and degrades the optimization performance in the process of generating data and training neural networks. To reduce the action space of DRL, we design a reference-model-assisted DRL approach. We introduce definitions of the reference model, reference-model-based optimization, and reference actions. The reference-model-assisted DRL learns the residual actions between the reference actions and optimal actions, rather than learning the optimal actions directly. Since the residual actions are considerably smaller than the optimal actions for a reference model, we can design a smaller action space for the reference-model-assisted DRL. It reduces the learning difficulties of DRL and optimises the performance of the reference-model-assisted DRL approach. It is noteworthy that the reference-model-assisted DRL approach is compatible with any policy gradient DRL algorithms for continuous action problems. This work takes the soft actor-critic algorithm as an example and designs a reference-model-assisted soft actor-critic algorithm. Simulations show that 1) large action space degrades the performance of DRL in the whole training stage, and 2) reference-model-assisted DRL requires fewer iteration times and returns a better optimization performance.

eess.SY↗

arXiv2022

Reducing Learning Difficulties: One-Step Two-Critic Deep Reinforcement Learning for Inverter-based Volt-Var Control

A one-step two-critic deep reinforcement learning (OSTC-DRL) approach for inverter-based volt-var control (IB-VVC) in active distribution networks is proposed in this paper. Firstly, considering IB-VVC can be formulated as a single-period optimization problem, we formulate the IB-VVC as a one-step Markov decision process rather than the standard Markov decision process, which simplifies the DRL learning task. Then we design the one-step actor-critic DRL scheme which is a simplified version of recent DRL algorithms, and it avoids the issue of Q value overestimation successfully. Furthermore, considering two objectives of VVC: minimizing power loss and eliminating voltage violation, we utilize two critics to approximate the rewards of two objectives separately. It simplifies the approximation tasks of each critic, and avoids the interaction effect between two objectives in the learning process of critic. The OSTC-DRL approach integrates the one-step actor-critic DRL scheme and the two-critic technology. Based on the OSTC-DRL, we design two centralized DRL algorithms. Further, we extend the OSTC-DRL to multi-agent OSTC-DRL for decentralized IB-VVC and design two multi-agent DRL algorithms. Simulations demonstrate that the proposed OSTC-DRL has a faster convergence rate and a better control performance, and the multi-agent OSTC-DRL works well for decentralized IB-VVC problems.

cs.AI↗

arXiv2020

Optimal Planning of Integrated Heat and Electricity Systems: a Tightening McCormick Approach

In this paper, we propose a convex planning model of integrated heat and electricity systems considering variable mass flow rates. The main challenge comes from the non-convexity of the bilinear terms in the district heating network, i.e., the product of mass flow rate and nodal temperature. To resolve this issue, we first reformulate the district heating network model through equivalent transformation and variable substitution. It shows that the reformulated model has only one set of nonconvex constraints with reduced bilinear terms and the others are linear constraints. Such a reformulation not only guarantees the optimality but fastens the solving process. To relax the remaining bilinear constraints, we apply McCormick envelopes and further propose a heuristic tightening method to constrict the bounds of the McCormick approach and get a nearby feasible solution. Case studies show that the tightening McCormick method quickly solves the heat-electricity planning problem with acceptable feasibility check and optimality.

math.OC↗

arXiv2020

Community Energy Storage Management for Welfare Optimization Using a Markov Decision Process

In this paper, we address an optimal management problem of community energy storage in the real-time electricity market under a stochastic renewable environment. In a real-time electricity market, complete market information may not be assessable for a strategic participant, hence we propose a paradigm that uses partial information including the forecast of real-time prices and slopes of the aggregate supply curve to model the price impact of storage use in the price-maker storage management problem. As a price maker, the community energy storage can not only earn profits through energy arbitrage but also smooth price trajectories and further influence social welfare. We formulate the problem as a finite-horizon Markov decision process that aims to maximize the energy arbitrage and social welfare of the prosumer-based community. The advance of the management scheme is that the optimal policy has a threshold structure. The structure has an analytic form that can guide the energy storage to charge/discharge by comparing its current marginal value and the expected future marginal value. Case studies indicate that welfare-maximizing storage earns more benefits than profit-maximizing storage. The proposed threshold-based algorithm can guarantee optimality and largely decrease the computational complexity of standard stochastic dynamic programming.

eess.SY↗

arXiv2020

A Linear Branch Flow Model for Radial Distribution Networks and its Application to Reactive Power Optimization and Network Reconfiguration

This paper presents a cold-start linear branch flow model named modified DistFlow. In modified DistFlow, the active and reactive power are replaced by their ratios to voltage magnitude as state variables, so that errors introduced by conventional branch flow linearization approaches due to their complete ignoring of the quadratic term are reduced. Based on the path-branch incidence matrix, branch power flows and nodal voltage magnitudes can be obtained in a non-iterative and explicit manner. Subsequently, the proposed modified DistFlow model is applied to the problem of reactive power optimization and network reconfiguration, transforming it into a mixed-integer quadratic programming (MIQP). Simulations show that the proposed modified DistFlow has a better accuracy than existing cold-start linear branch flow models for distribution networks, and the resulting MIQP model for reactive power optimization and network reconfiguration is much more computationally efficient than existing benchmarks.

eess.SY↗

arXiv2020

A Quadratic Convex Approximation of Optimal Power Flow in Distribution System with Application in Loss Allocation

In this paper, a novel quadratic convex optimal power flow model, namely, MDOPF, is proposed to determine the optimal dispatches of distributed generators. Based on the results of MDOPF, two price mechanisms, distribution locational marginal price (DLMP) and distribution locational price (DLP), are analyzed. For DLMP, an explicit method is developed to calculate the marginal loss that does not require a backward/forward sweep algorithm and thus reduces the computational complexity. However, the marginal loss component in DLMP will cause over-collection of losses (OCL). To address this issue, DLP is defined, which contains two components, the energy cost component and loss component, where the loss component is determined by the proposed loss allocation method (LAM). Numerical tests show that the proposed MDOPF has a better accuracy than existing OPF models based on linear power flow equations. In addition, the proposed marginal loss method and DLMP algorithm have satisfactory accuracy compared with benchmarks provided by ACOPF, and the proposed DLP can eliminate OCL.

eess.SY↗