PyTorchKR Papers

최신 논문

총 4,899편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2412.12153공식 코드 · Papers with Code

Revisiting Weight Averaging for Model Merging

Model merging aims to build a multi-task learner by combining the parameters of individually fine-tuned models without additional training. While a straightforward approach is to average model parameters across tasks, this often results in suboptimal performance due to interference among parameters across tasks. In this paper, we present intriguing results that weight averaging implicitly induces task vectors centered around the weight averaging itself and that applying a low-rank approximation to these centered task vectors significantly improves merging performance. Our analysis shows that centering the task vectors effectively separates core task-specific knowledge and nuisance noise within the fine-tuned parameters into the top and lower singular vectors, respectively, allowing us to reduce inter-task interference through its low-rank approximation. We evaluate our method on eight image classification tasks, demonstrating that it outperforms prior methods by a significant margin, narrowing the performance gap with traditional multi-task learning to within 1-3%

arXiv:2412.13862

Energy-Based Preference Model Offers Better Offline Alignment than the Bradley-Terry Preference Model

Since the debut of DPO, it has been shown that aligning a target LLM with human preferences via the KL-constrained RLHF loss is mathematically equivalent to a special kind of reward modeling task. Concretely, the task requires: 1) using the target LLM to parameterize the reward model, and 2) tuning the reward model so that it has a 1:1 linear relationship with the true reward. However, we identify a significant issue: the DPO loss might have multiple minimizers, of which only one satisfies the required linearity condition. The problem arises from a well-known issue of the underlying Bradley-Terry preference model: it does not always have a unique maximum likelihood estimator (MLE). Consequently,the minimizer of the RLHF loss might be unattainable because it is merely one among many minimizers of the DPO loss. As a better alternative, we propose an energy-based model (EBM) that always has a unique MLE, inherently satisfying the linearity requirement. To approximate the MLE in practice, we propose a contrastive loss named Energy Preference Alignment (EPA), wherein each positive sample is contrasted against one or more strong negatives as well as many free weak negatives. Theoretical properties of our EBM enable the approximation error of EPA to almost surely vanish when a sufficient number of negatives are used. Empirically, we demonstrate that EPA consistently delivers better performance on open benchmarks compared to DPO, thereby showing the superiority of our EBM.

arXiv:2412.13244

iRBSM: A Deep Implicit 3D Breast Shape Model

We present the first deep implicit 3D shape model of the female breast, building upon and improving the recently proposed Regensburg Breast Shape Model (RBSM). Compared to its PCA-based predecessor, our model employs implicit neural representations; hence, it can be trained on raw 3D breast scans and eliminates the need for computationally demanding non-rigid registration -- a task that is particularly difficult for feature-less breast shapes. The resulting model, dubbed iRBSM, captures detailed surface geometry including fine structures such as nipples and belly buttons, is highly expressive, and outperforms the RBSM on different surface reconstruction tasks. Finally, leveraging the iRBSM, we present a prototype application to 3D reconstruct breast shapes from just a single image. Model and code publicly available at https://rbsm.re-mic.de/implicit.

arXiv:2412.13630

Policy Decorator: Model-Agnostic Online Refinement for Large Policy Model

Recent advancements in robot learning have used imitation learning with large models and extensive demonstrations to develop effective policies. However, these models are often limited by the quantity, quality, and diversity of demonstrations. This paper explores improving offline-trained imitation learning models through online interactions with the environment. We introduce Policy Decorator, which uses a model-agnostic residual policy to refine large imitation learning models during online interactions. By implementing controlled exploration strategies, Policy Decorator enables stable, sample-efficient online learning. Our evaluation spans eight tasks across two benchmarks-ManiSkill and Adroit-and involves two state-of-the-art imitation learning models (Behavior Transformer and Diffusion Policy). The results show Policy Decorator effectively improves the offline-trained policies and preserves the smooth motion of imitation learning models, avoiding the erratic behaviors of pure RL policies. See our project page (https://policydecorator.github.io) for videos.

arXiv:2412.13437

Deploying Foundation Model Powered Agent Services: A Survey

Foundation model (FM) powered agent services are regarded as a promising solution to develop intelligent and personalized applications for advancing toward Artificial General Intelligence (AGI). To achieve high reliability and scalability in deploying these agent services, it is essential to collaboratively optimize computational and communication resources, thereby ensuring effective resource allocation and seamless service delivery. In pursuit of this vision, this paper proposes a unified framework aimed at providing a comprehensive survey on deploying FM-based agent services across heterogeneous devices, with the emphasis on the integration of model and resource optimization to establish a robust infrastructure for these services. Particularly, this paper begins with exploring various low-level optimization strategies during inference and studies approaches that enhance system scalability, such as parallelism techniques and resource scaling methods. The paper then discusses several prominent FMs and investigates research efforts focused on inference acceleration, including techniques such as model compression and token reduction. Moreover, the paper also investigates critical components for constructing agent services and highlights notable intelligent applications. Finally, the paper presents potential research directions for developing real-time agent services with high Quality of Service (QoS).

arXiv:2412.13432공식 코드 · Papers with Code

Large Language Model Enhanced Recommender Systems: A Survey

Large Language Model (LLM) has transformative potential in various domains, including recommender systems (RS). There have been a handful of research that focuses on empowering the RS by LLM. However, previous efforts mainly focus on LLM as RS, which may face the challenge of intolerant inference costs by LLM. Recently, the integration of LLM into RS, known as LLM-Enhanced Recommender Systems (LLMERS), has garnered significant interest due to its potential to address latency and memory constraints in real-world applications. This paper presents a comprehensive survey of the latest research efforts aimed at leveraging LLM to enhance RS capabilities. We identify a critical shift in the field with the move towards incorporating LLM into the online system, notably by avoiding their use during inference. Our survey categorizes the existing LLMERS approaches into three primary types based on the component of the RS model being augmented: Knowledge Enhancement, Interaction Enhancement, and Model Enhancement. We provide an in-depth analysis of each category, discussing the methodologies, challenges, and contributions of recent studies. Furthermore, we highlight several promising research directions that could further advance the field of LLMERS.

arXiv:2412.13858

IDEQ: an improved diffusion model for the TSP

We investigate diffusion models to solve the Traveling Salesman Problem. Building on the recent DIFUSCO and T2TCO approaches, we propose IDEQ. IDEQ improves the quality of the solutions by leveraging the constrained structure of the state space of the TSP. Another key component of IDEQ consists in replacing the last stages of DIFUSCO curriculum learning by considering a uniform distribution over the Hamiltonian tours whose orbits by the 2-opt operator converge to the optimal solution as the training objective. Our experiments show that IDEQ improves the state of the art for such neural network based techniques on synthetic instances. More importantly, our experiments show that IDEQ performs very well on the instances of the TSPlib, a reference benchmark in the TSP community: it closely matches the performance of the best heuristics, LKH3, being even able to obtain better solutions than LKH3 on 2 instances of the TSPlib defined on 1577 and 3795 cities. IDEQ obtains 0.3% optimality gap on TSP instances made of 500 cities, and 0.5% on TSP instances with 1000 cities. This sets a new SOTA for neural based methods solving the TSP. Moreover, IDEQ exhibits a lower variance and better scales-up with the number of cities with regards to DIFUSCO and T2TCO.

arXiv:2412.13844

CRM: Retrieval Model with Controllable Condition

Recommendation systems (RecSys) are designed to connect users with relevant items from a vast pool of candidates while aligning with the business goals of the platform. A typical industrial RecSys is composed of two main stages, retrieval and ranking: (1) the retrieval stage aims at searching hundreds of item candidates satisfied user interests; (2) based on the retrieved items, the ranking stage aims at selecting the best dozen items by multiple targets estimation for each item candidate, including classification and regression targets. Compared with ranking model, the retrieval model absence of item candidate information during inference, therefore retrieval models are often trained by classification target only (e.g., click-through rate), but failed to incorporate regression target (e.g., the expected watch-time), which limit the effectiveness of retrieval. In this paper, we propose the Controllable Retrieval Model (CRM), which integrates regression information as conditional features into the two-tower retrieval paradigm. This modification enables the retrieval stage could fulfill the target gap with ranking model, enhancing the retrieval model ability to search item candidates satisfied the user interests and condition effectively. We validate the effectiveness of CRM through real-world A/B testing and demonstrate its successful deployment in Kuaishou short-video recommendation system, which serves over 400 million users.

arXiv:2412.12244공식 코드 · Papers with Code

ChronoFlow: A Data-Driven Model for Gyrochronology

Gyrochronology is a technique for constraining stellar ages using rotation periods, which change over a star's main sequence lifetime due to magnetic braking. This technique shows promise for main sequence FGKM stars, where other methods are imprecise. However, the observed dispersion in rotation rates for similar coeval stars has historically been difficult to characterize. To properly understand this complexity, we have assembled the largest standardized data catalog of rotators in open clusters to date, consisting of $\approx$8,000 stars across 30 open clusters/associations spanning ages of 1.5 Myr to 4 Gyr. We have also developed ChronoFlow: a flexible data-driven model which accurately captures observed rotational dispersion. We show that ChronoFlow can be used to accurately forward model rotational evolution, and to infer both cluster and individual stellar ages. We recover cluster ages with a statistical uncertainty of 0.06 dex ($\approx$15%), and individual stellar ages with a statistical uncertainty of 0.7 dex. Additionally, we conducted robust systematic tests to analyze the impact of extinction models, cluster membership, and calibration ages. These contribute an additional 0.06 dex of uncertainty in cluster age estimates, resulting in a total error budget of 0.08 dex. We apply ChronoFlow to estimate ages for M34, NGC 2516, NGC 6709, and the Theia 456 stellar stream. Our results show that ChronoFlow can precisely estimate the ages of coeval stellar populations, and constrain ages for individual stars. Furthermore, its predictions may be used to inform physical spin down models. ChronoFlow is publicly available at https://github.com/philvanlane/chronoflow.

arXiv:2412.13158

On Model Extrapolation in Marginal Shapley Values

As the use of complex machine learning models continues to grow, so does the need for reliable explainability methods. One of the most popular methods for model explainability is based on Shapley values. There are two most commonly used approaches to calculating Shapley values which produce different results when features are correlated, conditional and marginal. In our previous work, it was demonstrated that the conditional approach is fundamentally flawed due to implicit assumptions of causality. However, it is a well-known fact that marginal approach to calculating Shapley values leads to model extrapolation where it might not be well defined. In this paper we explore the impacts of model extrapolation on Shapley values in the case of a simple linear spline model. Furthermore, we propose an approach which while using marginal averaging avoids model extrapolation and with addition of causal information replicates causal Shapley values. Finally, we demonstrate our method on the real data example.

arXiv:2412.12468

Transferable and Forecastable User Targeting Foundation Model

User targeting, the process of selecting targeted users from a pool of candidates for non-expert marketers, has garnered substantial attention with the advancements in digital marketing. However, existing user targeting methods encounter two significant challenges: (i) Poor cross-domain and cross-scenario transferability and generalization, and (ii) Insufficient forecastability in real-world applications. These limitations hinder their applicability across diverse industrial scenarios. In this work, we propose FIND, an industrial-grade, transferable, and forecastable user targeting foundation model. To enhance cross-domain transferability, our framework integrates heterogeneous multi-scenario user data, aligning them with one-sentence targeting demand inputs through contrastive pre-training. For improved forecastability, the text description of each user is derived based on anticipated future behaviors, while user representations are constructed from historical information. Experimental results demonstrate that our approach significantly outperforms existing baselines in cross-domain, real-world user targeting scenarios, showcasing the superior capabilities of FIND. Moreover, our method has been successfully deployed on the Alipay platform and is widely utilized across various scenarios.

arXiv:2412.12755

Progressive Monitoring of Generative Model Training Evolution

While deep generative models (DGMs) have gained popularity, their susceptibility to biases and other inefficiencies that lead to undesirable outcomes remains an issue. With their growing complexity, there is a critical need for early detection of issues to achieve desired results and optimize resources. Hence, we introduce a progressive analysis framework to monitor the training process of DGMs. Our method utilizes dimensionality reduction techniques to facilitate the inspection of latent representations, the generated and real distributions, and their evolution across training iterations. This monitoring allows us to pause and fix the training method if the representations or distributions progress undesirably. This approach allows for the analysis of a models' training dynamics and the timely identification of biases and failures, minimizing computational loads. We demonstrate how our method supports identifying and mitigating biases early in training a Generative Adversarial Network (GAN) and improving the quality of the generated data distribution.

arXiv:2412.10457

Explaining Model Overfitting in CNNs via GMM Clustering

Convolutional Neural Networks (CNNs) have demonstrated remarkable prowess in the field of computer vision. However, their opaque decision-making processes pose significant challenges for practical applications. In this study, we provide quantitative metrics for assessing CNN filters by clustering the feature maps corresponding to individual filters in the model via Gaussian Mixture Model (GMM). By analyzing the clustering results, we screen out some anomaly filters associated with outlier samples. We further analyze the relationship between the anomaly filters and model overfitting, proposing three hypotheses. This method is universally applicable across diverse CNN architectures without modifications, as evidenced by its successful application to models like AlexNet and LeNet-5. We present three meticulously designed experiments demonstrating our hypotheses from the perspectives of model behavior, dataset characteristics, and filter impacts. Through this work, we offer a novel perspective for evaluating the CNN performance and gain new insights into the operational behavior of model overfitting.

arXiv:2412.10416

SUPERMERGE: An Approach For Gradient-Based Model Merging

Large language models, such as ChatGPT, Claude, or LLaMA, are gigantic, monolithic, and possess the superpower to simultaneously support thousands of tasks. However, high-throughput applications often prefer smaller task-specific models because of their lower latency and cost. One challenge of using task-specific models is the incremental need for solving newer tasks after the model is already deployed for existing tasks. A straightforward solution requires fine-tuning the model again for both existing and new tasks, which is computationally expensive and time-consuming. To address this issue, we propose a model merging based approach called SUPERMERGE. SUPERMERGE is a gradient-based method to systematically merge several fine-tuned models trained on existing and new tasks. SUPERMERGE is designed to be lightweight and fast, and the merged model achieves similar performance to fully fine-tuned models on all tasks. Furthermore, we proposed a hierarchical model merging strategy to reduce the peak space requirement without sacrificing the performance of the merged model. We experimentally demonstrate that SUPERMERGE outperforms existing model merging methods on common natural language processing and computer vision tasks.

arXiv:2412.10411

Pre-trained protein language model for codon optimization

Motivation: Codon optimization of Open Reading Frame (ORF) sequences is essential for enhancing mRNA stability and expression in applications like mRNA vaccines, where codon choice can significantly impact protein yield which directly impacts immune strength. In this work, we investigate the use of a pre-trained protein language model (PPLM) for getting a rich representation of amino acids which could be utilized for codon optimization. This leaves us with a simpler fine-tuning task over PPLM in optimizing ORF sequences. Results: The ORFs generated by our proposed models outperformed their natural counterparts encoding the same proteins on computational metrics for stability and expression. They also demonstrated enhanced performance against the benchmark ORFs used in mRNA vaccines for the SARS-CoV-2 viral spike protein and the varicella-zoster virus (VZV). These results highlight the potential of adapting PPLM for designing ORFs tailored to encode target antigens in mRNA vaccines.

arXiv:2412.10858

CRENER: A Character Relation Enhanced Chinese NER Model

Chinese Named Entity Recognition (NER) is an important task in information extraction, which has a significant impact on downstream applications. Due to the lack of natural separators in Chinese, previous NER methods mostly relied on external dictionaries to enrich the semantic and boundary information of Chinese words. However, such methods may introduce noise that affects the accuracy of named entity recognition. To this end, we propose a character relation enhanced Chinese NER model (CRENER). This model defines four types of tags that reflect the relationships between characters, and proposes a fine-grained modeling of the relationships between characters based on three types of relationships: adjacency relations between characters, relations between characters and tags, and relations between tags, to more accurately identify entity boundaries and improve Chinese NER accuracy. Specifically, we transform the Chinese NER task into a character-character relationship classification task, ensuring the accuracy of entity boundary recognition through joint modeling of relation tags. To enhance the model's ability to understand contextual information, WRENER further constructed an adapted transformer encoder that combines unscaled direction-aware and distance-aware masked self-attention mechanisms. Moreover, a relationship representation enhancement module was constructed to model predefined relationship tags, effectively mining the relationship representations between characters and tags. Experiments conducted on four well-known Chinese NER benchmark datasets have shown that the proposed model outperforms state-of-the-art baselines. The ablation experiment also demonstrated the effectiveness of the proposed model.

arXiv:2412.11016

A Contextualized BERT model for Knowledge Graph Completion

Knowledge graphs (KGs) are valuable for representing structured, interconnected information across domains, enabling tasks like semantic search, recommendation systems and inference. A pertinent challenge with KGs, however, is that many entities (i.e., heads, tails) or relationships are unknown. Knowledge Graph Completion (KGC) addresses this by predicting these missing nodes or links, enhancing the graph's informational depth and utility. Traditional methods like TransE and ComplEx predict tail entities but struggle with unseen entities. Textual-based models leverage additional semantics but come with high computational costs, semantic inconsistencies, and data imbalance issues. Recent LLM-based models show improvement but overlook contextual information and rely heavily on entity descriptions. In this study, we introduce a contextualized BERT model for KGC that overcomes these limitations by utilizing the contextual information from neighbouring entities and relationships to predict tail entities. Our model eliminates the need for entity descriptions and negative triplet sampling, reducing computational demands while improving performance. Our model outperforms state-of-the-art methods on standard datasets, improving Hit@1 by 5.3% and 4.88% on FB15k-237 and WN18RR respectively, setting a new benchmark in KGC.

arXiv:2412.11006공식 코드 · Papers with Code

Entropy-Regularized Process Reward Model

Large language models (LLMs) have shown promise in performing complex multi-step reasoning, yet they continue to struggle with mathematical reasoning, often making systematic errors. A promising solution is reinforcement learning (RL) guided by reward models, particularly those focusing on process rewards, which score each intermediate step rather than solely evaluating the final outcome. This approach is more effective at guiding policy models towards correct reasoning trajectories. In this work, we propose an entropy-regularized process reward model (ER-PRM) that integrates KL-regularized Markov Decision Processes (MDP) to balance policy optimization with the need to prevent the policy from shifting too far from its initial distribution. We derive a novel reward construction method based on the theoretical results. Our theoretical analysis shows that we could derive the optimal reward model from the initial policy sampling. Our empirical experiments on the MATH and GSM8K benchmarks demonstrate that ER-PRM consistently outperforms existing process reward models, achieving 1% improvement on GSM8K and 2-3% improvement on MATH under best-of-N evaluation, and more than 1% improvement under RLHF. These results highlight the efficacy of entropy-regularization in enhancing LLMs' reasoning capabilities.

arXiv:2412.11483

"They've Stolen My GPL-Licensed Model!": Toward Standardized and Transparent Model Licensing

As model parameter sizes reach the billion-level range and their training consumes zettaFLOPs of computation, components reuse and collaborative development are become increasingly prevalent in the Machine Learning (ML) community. These components, including models, software, and datasets, may originate from various sources and be published under different licenses, which govern the use and distribution of licensed works and their derivatives. However, commonly chosen licenses, such as GPL and Apache, are software-specific and are not clearly defined or bounded in the context of model publishing. Meanwhile, the reused components may also have free-content licenses and model licenses, which pose a potential risk of license noncompliance and rights infringement within the model production workflow. In this paper, we propose addressing the above challenges along two lines: 1) For license analysis, we have developed a new vocabulary for ML workflow management and encoded license rules to enable ontological reasoning for analyzing rights granting and compliance issues. 2) For standardized model publishing, we have drafted a set of model licenses that provide flexible options to meet the diverse needs of model publishing. Our analysis tool is built on Turtle language and Notation3 reasoning engine, envisioned as a first step toward Linked Open Model Production Data. We have also encoded our proposed model licenses into rules and demonstrated the effects of GPL and other commonly used licenses in model publishing, along with the flexibility advantages of our licenses, through comparisons and experiments.

arXiv:2412.10824

Diffusion Model from Scratch

Diffusion generative models are currently the most popular generative models. However, their underlying modeling process is quite complex, and starting directly with the seminal paper Denoising Diffusion Probability Model (DDPM) can be challenging. This paper aims to assist readers in building a foundational understanding of generative models by tracing the evolution from VAEs to DDPM through detailed mathematical derivations and a problem-oriented analytical approach. It also explores the core ideas and improvement strategies of current mainstream methodologies, providing guidance for undergraduate and graduate students interested in learning about diffusion models.