arXiv subjects

Yan Ma

Publications and source records attributed to Yan Ma.

At least 55 records · Page 3Linked to original sources

arXiv2024

A Model-free Biomimetics Algorithm for Deterministic Partially Observable Markov Decision Process

Partially Observable Markov Decision Process (POMDP) is a mathematical framework for modeling decision-making under uncertainty, where the agent's observations are incomplete and the underlying system dynamics are probabilistic. Solving the POMDP problem within the model-free paradigm is challenging for agents due to the inherent difficulty in accurately identifying and distinguishing between states and observations. We define such a difficult problem as a DETerministic Partially Observable Markov Decision Process (DET-POMDP) problem, which is a specific setting of POMDP. In this problem, states and observations are in a many-to-one relationship. The state is obscured, and its relationship is less apparent to the agent. This creates obstacles for the agent to infer the state through observations. To effectively address this problem, we convert DET-POMDP into a fully observable MDP using a model-free biomimetics algorithm called BIOMAP. BIOMAP is based on the MDP Graph Automaton framework to distinguish authentic environmental information from fraudulent data. Thus, it enhances the agent's ability to develop stable policies against DET-POMDP. The experimental results highlight the superior capabilities of BIOMAP in maintaining operational effectiveness and environmental reparability in the presence of environmental deceptions when compared with existing POMDP solvers. This research opens up new avenues for the deployment of reliable POMDP-based systems in fields that are particularly susceptible to DET-POMDP problems.

eess.SY↗

arXiv2024

MMLongBench-Doc: Benchmarking Long-context Document Understanding with Visualizations

Understanding documents with rich layouts and multi-modal components is a long-standing and practical task. Recent Large Vision-Language Models (LVLMs) have made remarkable strides in various tasks, particularly in single-page document understanding (DU). However, their abilities on long-context DU remain an open problem. This work presents MMLongBench-Doc, a long-context, multi-modal benchmark comprising 1,062 expert-annotated questions. Distinct from previous datasets, it is constructed upon 130 lengthy PDF-formatted documents with an average of 49.4 pages and 20,971 textual tokens. Towards comprehensive evaluation, answers to these questions rely on pieces of evidence from (1) different sources (text, image, chart, table, and layout structure) and (2) various locations (i.e. page number). Moreover, 33.2% of the questions are cross-page questions requiring evidence across multiple pages. 22.8% of the questions are designed to be unanswerable for detecting potential hallucinations. Experiments on 14 LVLMs demonstrate that long-context DU greatly challenges current models. Notably, the best-performing model, GPT-4o, achieves an F1 score of only 42.7%, while the second-best, GPT-4V, scores 31.4%. Furthermore, 12 LVLMs (all except GPT-4o and GPT-4V) even present worse performance than their LLM counterparts which are fed with lossy-parsed OCR documents. These results validate the necessity of future research toward more capable long-context LVLMs. Project Page: https://mayubo2333.github.io/MMLongBench-Doc

cs.CV↗

arXiv2024

An atomic-scale view at Fe4N as hydrogen barrier material

Hydrogen, while a promising sustainable energy carrier, presents challenges such as the embrittlement of materials due to its ability to penetrate and weaken their crystal structures. Here we investigate Fe4N nitride layers, formed on iron through a cost-effective gas nitriding process, as an effective hydrogen permeation barrier. A combination of screening using advanced characterization, density functional theory calculations, and hydrogen permeation analysis reveals that a nitride layer reduces hydrogen diffusion by a factor of 20 at room temperature. This reduction is achieved by creating energetically unfavorable states due to stronger H-binding at the surface and high energy barriers for diffusion. The findings demonstrate the potential of Fe4N as a cost-efficient and easy-to-process solution to protecting metallic materials exposed to hydrogen, with great advantages for large-scale applications.

cond-mat.mtrl-sci↗

arXiv2024

Weak-to-Strong Reasoning

When large language models (LLMs) exceed human-level capabilities, it becomes increasingly challenging to provide full-scale and accurate supervision for these models. Weak-to-strong learning, which leverages a less capable model to unlock the latent abilities of a stronger model, proves valuable in this context. Yet, the efficacy of this approach for complex reasoning tasks is still untested. Furthermore, tackling reasoning tasks under the weak-to-strong setting currently lacks efficient methods to avoid blindly imitating the weak supervisor including its errors. In this paper, we introduce a progressive learning framework that enables the strong model to autonomously refine its training data, without requiring input from either a more advanced model or human-annotated data. This framework begins with supervised fine-tuning on a selective small but high-quality dataset, followed by preference optimization on contrastive samples identified by the strong model itself. Extensive experiments on the GSM8K and MATH datasets demonstrate that our method significantly enhances the reasoning capabilities of Llama2-70b using three separate weak models. This method is further validated in a forward-looking experimental setup, where Llama3-8b-instruct effectively supervises Llama3-70b on the highly challenging OlympicArena dataset. This work paves the way for a more scalable and sophisticated strategy to enhance AI reasoning powers. All relevant code and resources are available in \url{https://github.com/GAIR-NLP/weak-to-strong-reasoning}.

cs.CL↗

arXiv2024

ANOLE: An Open, Autoregressive, Native Large Multimodal Models for Interleaved Image-Text Generation

Previous open-source large multimodal models (LMMs) have faced several limitations: (1) they often lack native integration, requiring adapters to align visual representations with pre-trained large language models (LLMs); (2) many are restricted to single-modal generation; (3) while some support multimodal generation, they rely on separate diffusion models for visual modeling and generation. To mitigate these limitations, we present Anole, an open, autoregressive, native large multimodal model for interleaved image-text generation. We build Anole from Meta AI's Chameleon, adopting an innovative fine-tuning strategy that is both data-efficient and parameter-efficient. Anole demonstrates high-quality, coherent multimodal generation capabilities. We have open-sourced our model, training framework, and instruction tuning data.

cs.CL↗

arXiv2024

MoPS: Modular Story Premise Synthesis for Open-Ended Automatic Story Generation

A story premise succinctly defines a story's main idea, foundation, and trajectory. It serves as the initial trigger in automatic story generation. Existing sources of story premises are limited by a lack of diversity, uneven quality, and high costs that make them difficult to scale. In response, we introduce Modular Story Premise Synthesis (MoPS) which breaks down story premises into modules like background and persona for automated design and generation. MoPS consists of three phases: (1) Precollect a consistent set of candidates for each module to form a nested dictionary. (2) Extract a key path from the nested dictionary as the premise design. (3) Instruct an LLM to integrate the design into a coherent premise sentence. Thorough evaluations demonstrate that our synthesized premises excel in diversity, fascination, completeness, and originality compared to those induced from large language models and captured from public story datasets. Similarly, the extended novels and scripts generated from our premises also exhibit higher quality. In supplementary materials, we provide the MoPS code suite, along with 7.6k generated premises and 1k extended stories. Code: https://github.com/GAIR-NLP/MoPS.

cs.CL↗

arXiv2024

SA-LSPL:Sequence-Aware Long- and Short- Term Preference Learning for next POI recommendation

The next Point of Interest (POI) recommendation aims to recommend the next POI for users at a specific time. As users' check-in records can be viewed as a long sequence, methods based on Recurrent Neural Networks (RNNs) have recently shown good applicability to this task. However, existing methods often struggle to fully explore the spatio-temporal correlations and dependencies at the sequence level, and don't take full consideration for various factors influencing users' preferences. To address these issues, we propose a novel approach called Sequence-Aware Long- and Short-Term Preference Learning (SA-LSPL) for next-POI recommendation. We combine various information features to effectively model users' long-term preferences. Specifically, our proposed model uses a multi-modal embedding module to embed diverse check-in details, taking into account both user's personalized preferences and social influences comprehensively. Additionally, we consider explicit spatio-temporal correlations at the sequence level and implicit sequence dependencies. Furthermore, SA-LSPL learns the spatio-temporal correlations of consecutive and non-consecutive visits in the current check-in sequence, as well as transition dependencies between categories, providing a comprehensive capture of user's short-term preferences. Extensive experiments on two real-world datasets demonstrate the superiority of SA-LSPL over state-of-the-art baseline methods.

cs.CY↗

arXiv2024

Swarm: Cost-Efficient Video Content Distribution with a Peer-to-Peer System

As ByteDance's business expands, the substantial infrastructure expenses associated with centralized Content Delivery Network (CDN) networks have rendered content distribution costs prohibitively high. In response, we embarked on exploring a peer-to-peer (P2P) network as a promising solution to alleviate the escalating costs of content distribution. However, the decentralized nature of P2P often introduces performance challenges, given the diversity and dispersion of peer devices. This study introduces Swarm, ByteDance's innovative hybrid system for video streaming. Swarm seamlessly integrates the robustness of a conventional CDN with the cost-efficiency of a decentralized P2P network. Its primary aim is to provide users with reliable streaming quality while minimizing traffic expenses. To achieve this, Swarm employs a centralized control plane comprised of a tracker cluster, overseeing a data plane with numerous edge residual resources. The tracker also takes on the responsibility of mapping clients to servers. Addressing the performance disparities among individual peer servers, Swarm utilizes our proprietary multipath parallel transmission method for communication between clients and peer servers. Operating stably for six years, Swarm now manages over a hundred thousand peer servers, serving nearly a hundred million users daily and saving the company hundreds of millions of RMB annually. Experimental results affirm that, while significantly cutting costs, Swarm performs on par with traditional CDNs.

cs.NI↗

arXiv2024

Analysis of the $gg\to H\to hh\to4τ$ process in the 2HDM lepton specific model at the LHC

We analyse the signature of a light Higgs boson pair in the 2-Higgs Doublet Model(2HDM) Type-X (or lepton specific) over the parameter spaces allowed by theoretical self-consistency requirements as well as the latest experimental constraints from the Large Hadron Collider (LHC), precision test data and $B$ physics. Over the viable regions of the latter, wherein the Standard Model (SM)-like Higgs boson discovered at the LHC in 2012 is the heavier CP-even state of the 2HDM, $H$, it is found that the SM-like Higgs boson can decay into a pair of the lighter CP-even Higgs boson, $h$, via the process $H\to hh$ with a Branching Ratio (BR) of $5\%-10\%$ or so, (with $2 m_h < m_H =125$ GeV). Furthermore, in the Type-X scenario, the lighter Higgs bosons $h$ can dominantly decay into two $τ$'s due to a large $\tanβ$. Therefore, the pair of lighter Higgs bosons can altogether decay into a 4 $τ$ final state. In order to suppress the huge SM background events, we confine ourself to consider the fraction of signal events with two Same-Sign (SS) $τ$'s further decaying into same sign leptons while the other two $τ$'s decay hadronically. By using Monte Carlo (MC) and Machine Learning (ML) tools, we thus focus on the analysis of the signal process $pp\to H\to hh\to τ^{+}τ^{-}τ^{+}τ^{-}\to \ell v_\ell \ell v_\ell τ_h τ_h$ (where $\ell= e, μ$ and $τ_h$ means a hadronic decay of the $τ$) and explore the feasibility of such a search at the LHC for a collision energy $\sqrt{s}=~\text{14 TeV}$ and a luminosity $\text{300}~\text{fb}^{-1}$

hep-ph↗

arXiv2023

Revisiting DETR Pre-training for Object Detection

Motivated by the remarkable achievements of DETR-based approaches on COCO object detection and segmentation benchmarks, recent endeavors have been directed towards elevating their performance through self-supervised pre-training of Transformers while preserving a frozen backbone. Noteworthy advancements in accuracy have been documented in certain studies. Our investigation delved deeply into a representative approach, DETReg, and its performance assessment in the context of emerging models like $\mathcal{H}$-Deformable-DETR. Regrettably, DETReg proves inadequate in enhancing the performance of robust DETR-based models under full data conditions. To dissect the underlying causes, we conduct extensive experiments on COCO and PASCAL VOC probing elements such as the selection of pre-training datasets and strategies for pre-training target generation. By contrast, we employ an optimized approach named Simple Self-training which leads to marked enhancements through the combination of an improved box predictor and the Objects$365$ benchmark. The culmination of these endeavors results in a remarkable AP score of $59.3\%$ on the COCO val set, outperforming $\mathcal{H}$-Deformable-DETR + Swin-L without pre-training by $1.4\%$. Moreover, a series of synthetic pre-training datasets, generated by merging contemporary image-to-text(LLaVA) and text-to-image (SDXL) models, significantly amplifies object detection capabilities.

cs.CV↗

arXiv2023

Quantification before Selection: Active Dynamics Preference for Robust Reinforcement Learning

Training a robust policy is critical for policy deployment in real-world systems or dealing with unknown dynamics mismatch in different dynamic systems. Domain Randomization~(DR) is a simple and elegant approach that trains a conservative policy to counter different dynamic systems without expert knowledge about the target system parameters. However, existing works reveal that the policy trained through DR tends to be over-conservative and performs poorly in target domains. Our key insight is that dynamic systems with different parameters provide different levels of difficulty for the policy, and the difficulty of behaving well in a system is constantly changing due to the evolution of the policy. If we can actively sample the systems with proper difficulty for the policy on the fly, it will stabilize the training process and prevent the policy from becoming over-conservative or over-optimistic. To operationalize this idea, we introduce Active Dynamics Preference~(ADP), which quantifies the informativeness and density of sampled system parameters. ADP actively selects system parameters with high informativeness and low density. We validate our approach in four robotic locomotion tasks with various discrepancies between the training and testing environments. Extensive results demonstrate that our approach has superior robustness for system inconsistency compared to several baselines.

cs.LG↗

arXiv2023

Open-Ended Diverse Solution Discovery with Regulated Behavior Patterns for Cross-Domain Adaptation

While Reinforcement Learning can achieve impressive results for complex tasks, the learned policies are generally prone to fail in downstream tasks with even minor model mismatch or unexpected perturbations. Recent works have demonstrated that a policy population with diverse behavior characteristics can generalize to downstream environments with various discrepancies. However, such policies might result in catastrophic damage during the deployment in practical scenarios like real-world systems due to the unrestricted behaviors of trained policies. Furthermore, training diverse policies without regulation of the behavior can result in inadequate feasible policies for extrapolating to a wide range of test conditions with dynamics shifts. In this work, we aim to train diverse policies under the regularization of the behavior patterns. We motivate our paradigm by observing the inverse dynamics in the environment with partial state information and propose Diversity in Regulation (DiR) training diverse policies with regulated behaviors to discover desired patterns that benefit the generalization. Considerable empirical results on various variations of different environments indicate that our method attains improvements over other diversity-driven counterparts.

cs.LG↗

arXiv2023

The multiplexed light storage of Orbital Angular Momentum based on atomic ensembles

The improvement of the multi-mode capability of quantum memory can further improve the utilization efficiency of the quantum memory and reduce the requirement of quantum communication for storage units. In this letter, we experimentally investigate the multi-mode light multiplexing storage of orbital angular momentum (OAM) mode based on rubidium vapor, and demultiplexing by a photonic OAM mode splitter which combines a Sagnac loop with two dove prisms. Our results show a mode extinction ratio higher than 80$\%$ at 1 $μ$s of storage time. Meanwhile, two OAM modes have been multiplexing stored and demultiplexed in our experimental configuration. We believe the experimental scheme may provide a possibility for high channel capacity and multi-mode quantum multiplexed quantum storage based on atomic ensembles.

physics.atom-ph↗

arXiv2023

Effect of Pore Formation on Redox-Driven Phase Transformation

When solid-state redox-driven phase transformations are associated with mass loss, vacancies are produced that develop into pores. These pores can influence the kinetics of certain redox and phase transformation steps. We investigated the structural and chemical mechanisms in and at pores in a combined experimental-theoretical study, using the reduction of iron oxide by hydrogen as a model system. The redox product (water) accumulates inside the pores and shifts the local equilibrium at the already reduced material back towards re-oxidation into cubic-Fe1-xO (where x refers to Fe deficiency, space group Fm3-m). This effect helps to understand the sluggish reduction of cubic-Fe1-xO by hydrogen, a key process for future sustainable steelmaking.

cond-mat.mtrl-sci↗

arXiv2022

Evolutionary Action Selection for Gradient-based Policy Learning

Evolutionary Algorithms (EAs) and Deep Reinforcement Learning (DRL) have recently been integrated to take the advantage of the both methods for better exploration and exploitation.The evolutionary part in these hybrid methods maintains a population of policy networks.However, existing methods focus on optimizing the parameters of policy network, which is usually high-dimensional and tricky for EA.In this paper, we shift the target of evolution from high-dimensional parameter space to low-dimensional action space.We propose Evolutionary Action Selection-Twin Delayed Deep Deterministic Policy Gradient (EAS-TD3), a novel hybrid method of EA and DRL.In EAS, we focus on optimizing the action chosen by the policy network and attempt to obtain high-quality actions to promote policy learning through an evolutionary algorithm. We conduct several experiments on challenging continuous control tasks.The result shows that EAS-TD3 shows superior performance over other state-of-art methods.

cs.NE↗

arXiv2022

CIGAN: A Python Package for Handling Class Imbalance using Generative Adversarial Networks

A key challenge in Machine Learning is class imbalance, where the sample size of some classes (majority classes) are much higher than that of the other classes (minority classes). If we were to train a classifier directly on imbalanced data, it is more likely for the classifier to predict a new sample as one of the majority classes. In the extreme case, the classifier could completely ignore the minority classes. This could have serious sociological implications in healthcare, as the minority classes are usually the disease classes (e.g., death or positive clinical test result). In this paper, we introduce a software that uses Generative Adversarial Networks to oversample the minority classes so as to improve downstream classification. To the best of our knowledge, this is the first tool that allows multi-class classification (where the target can have an arbitrary number of classes). The code of the tool is publicly available in our github repository (https://github.com/yuxiaohuang/research/tree/master/gwu/working/cigan/code).

cs.LG↗

arXiv2022

RGBT Salient Object Detection: A Large-scale Dataset and Benchmark

Salient object detection in complex scenes and environments is a challenging research topic. Most works focus on RGB-based salient object detection, which limits its performance of real-life applications when confronted with adverse conditions such as dark environments and complex backgrounds. Taking advantage of RGB and thermal infrared images becomes a new research direction for detecting salient object in complex scenes recently, as thermal infrared spectrum imaging provides the complementary information and has been applied to many computer vision tasks. However, current research for RGBT salient object detection is limited by the lack of a large-scale dataset and comprehensive benchmark. This work contributes such a RGBT image dataset named VT5000, including 5000 spatially aligned RGBT image pairs with ground truth annotations. VT5000 has 11 challenges collected in different scenes and environments for exploring the robustness of algorithms. With this dataset, we propose a powerful baseline approach, which extracts multi-level features within each modality and aggregates these features of all modalities with the attention mechanism, for accurate RGBT salient object detection. Extensive experiments show that the proposed baseline approach outperforms the state-of-the-art methods on VT5000 dataset and other two public datasets. In addition, we carry out a comprehensive analysis of different algorithms of RGBT salient object detection on VT5000 dataset, and then make several valuable conclusions and provide some potential research directions for RGBT salient object detection.

cs.CV↗

arXiv2022

Green steel at its crossroads: hybrid hydrogen-based reduction of iron ores

Iron- and steelmaking cause ~7% of the global CO2 emissions, due to the use of carbon for the reduction of iron ores. Replacing carbon by hydrogen as the reductant offers a pathway to reduce emissions. However, production of hydrogen using renewable energy will remain a bottlenecks, because making the annual crude steel production of 1.8 billion tons sustainable requires a minimum amount of ~97 million tons of green hydrogen per year. Another fundamental aspect to make ironmaking sector more sustainable lies in an optimal utilization of green hydrogen and energy, thus reducing efforts for costly in-process hydrogen recycling. We therefore demonstrate here how the efficiency in hydrogen and energy consumption during iron ore reduction can be dramatically improved by the knowledge-based combination of two technologies: partially reducing the ore at low temperature via solid-state direct reduction (DR) to a kinetically defined degree, and subsequently melting and completely transforming it to iron under a reducing plasma (i.e. via hydrogen plasma reduction, HPR). Results suggest that an optimal transition point between these two technologies occurs where their efficiency in hydrogen utilization is equal. We found that the reduction of hematite through magnetite into wustite via DR is clean and efficient, but it gets sluggish and inefficient when iron forms at the outermost layers of the iron ore pellets. Conversely, HPR starts violent and unstable with arc delocalization, but proceeds smoothly and efficiently when processing semi-reduced oxides. We performed hybrid reduction experiments by partially reducing hematite pellets via DR at 700°C to 38% global reduction (using a standard thermogravimetry system) and subsequently transferring them to HPR, conducted with a gas mixture of Ar-10%H2 in an arc-melting furnace, to achieve conversion into liquid iron.

cond-mat.mtrl-sci↗