arXiv subjects

Yi Ren

Publications and source records attributed to Yi Ren.

At least 73 records · Page 4Linked to original sources

arXiv2025

A Time-Triggered Communication Method Based on Urgency-Based Scheduler in Time-Sensitive Networking

The development of the automotive industry and automation has led to a growing demand for time-critical systems to have low latency and jitter for critical traffic. To address this issue, the IEEE 802.1 Time-Sensitive Networking (TSN) task group proposed the Time-Aware Shaper (TAS) to implement Time-Triggered (TT) communication, enabling deterministic transmission by assigning specific time windows to each stream. While Fixed Routing and Waiting-Allowed (FR-WA) scheduling algorithms offer flexibility, they suffer from inefficiencies in solution time and scalability. This study analyzes TAS implementation challenges, emphasizing how network scale expansion increases computational constraints. We propose an Urgency-Based Scheduler method (TT-UBS) to address these limitations to enhance deterministic transmission and computational efficiency under anomalies. A novel scheduling algorithm for TT-UBS parameter determination is developed, alongside simulations and comparative evaluations. Results show that TT-UBS guarantees deterministic traffic delivery while reducing solution time by 98.22% in test scenarios compared to traditional approaches. The methodology is extended to other scheduling algorithms to assess efficiency improvements. This advancement supports TSN's application in mission-critical systems by optimizing time-triggered communication performance and enabling reliable network deployment. The framework demonstrates significant potential for real-time in-vehicle networks requiring latency and jitter control.

cs.NI↗

arXiv2025

Learning Dynamics of LLM Finetuning

Learning dynamics, which describes how the learning of specific training examples influences the model's predictions on other examples, gives us a powerful tool for understanding the behavior of deep learning systems. We study the learning dynamics of large language models during different types of finetuning, by analyzing the step-wise decomposition of how influence accumulates among different potential responses. Our framework allows a uniform interpretation of many interesting observations about the training of popular algorithms for both instruction tuning and preference tuning. In particular, we propose a hypothetical explanation of why specific types of hallucination are strengthened after finetuning, e.g., the model might use phrases or facts in the response for question B to answer question A, or the model might keep repeating similar simple phrases when generating responses. We also extend our framework and highlight a unique "squeezing effect" to explain a previously observed phenomenon in off-policy direct preference optimization (DPO), where running DPO for too long makes even the desired outputs less likely. This framework also provides insights into where the benefits of on-policy DPO and other variants come from. The analysis not only provides a novel perspective of understanding LLM's finetuning but also inspires a simple, effective method to improve alignment performance.

cs.LG↗

arXiv2025

SViP: Sequencing Bimanual Visuomotor Policies with Object-Centric Motion Primitives

Imitation learning (IL), particularly when leveraging high-dimensional visual inputs for policy training, has proven intuitive and effective in complex bimanual manipulation tasks. Nonetheless, the generalization capability of visuomotor policies remains limited, especially when small demonstration datasets are available. Accumulated errors in visuomotor policies significantly hinder their ability to complete long-horizon tasks. To address these limitations, we propose SViP, a framework that seamlessly integrates visuomotor policies into task and motion planning (TAMP). SViP partitions human demonstrations into bimanual and unimanual operations using a semantic scene graph monitor. Continuous decision variables from the key scene graph are employed to train a switching condition generator. This generator produces parameterized scripted primitives that ensure reliable performance even when encountering out-of-the-distribution observations. Using only 20 real-world demonstrations, we show that SViP enables visuomotor policies to generalize across out-of-distribution initial conditions without requiring object pose estimators. For previously unseen tasks, SViP automatically discovers effective solutions to achieve the goal, leveraging constraint modeling in TAMP formulism. In real-world experiments, SViP outperforms state-of-the-art generative IL methods, indicating wider applicability for more complex tasks. Project website: https://sites.google.com/view/svip-bimanual

cs.RO↗

arXiv2025

On the Effect of Negative Gradient in Group Relative Deep Reinforcement Optimization

Reinforcement learning (RL) has become popular in enhancing the reasoning capabilities of large language models (LLMs), with Group Relative Policy Optimization (GRPO) emerging as a widely used algorithm in recent systems. Despite GRPO's widespread adoption, we identify a previously unrecognized phenomenon we term Lazy Likelihood Displacement (LLD), wherein the likelihood of correct responses marginally increases or even decreases during training. This behavior mirrors a recently discovered misalignment issue in Direct Preference Optimization (DPO), attributed to the influence of negative gradients. We provide a theoretical analysis of GRPO's learning dynamic, identifying the source of LLD as the naive penalization of all tokens in incorrect responses with the same strength. To address this, we develop a method called NTHR, which downweights penalties on tokens contributing to the LLD. Unlike prior DPO-based approaches, NTHR takes advantage of GRPO's group-based structure, using correct responses as anchors to identify influential tokens. Experiments on math reasoning benchmarks demonstrate that NTHR effectively mitigates LLD, yielding consistent performance gains across models ranging from 0.5B to 3B parameters.

cs.LG↗

arXiv2025

Chain-Talker: Chain Understanding and Rendering for Empathetic Conversational Speech Synthesis

Conversational Speech Synthesis (CSS) aims to align synthesized speech with the emotional and stylistic context of user-agent interactions to achieve empathy. Current generative CSS models face interpretability limitations due to insufficient emotional perception and redundant discrete speech coding. To address the above issues, we present Chain-Talker, a three-stage framework mimicking human cognition: Emotion Understanding derives context-aware emotion descriptors from dialogue history; Semantic Understanding generates compact semantic codes via serialized prediction; and Empathetic Rendering synthesizes expressive speech by integrating both components. To support emotion modeling, we develop CSS-EmCap, an LLM-driven automated pipeline for generating precise conversational speech emotion captions. Experiments on three benchmark datasets demonstrate that Chain-Talker produces more expressive and empathetic speech than existing methods, with CSS-EmCap contributing to reliable emotion modeling. The code and demos are available at: https://github.com/AI-S2-Lab/Chain-Talker.

cs.SD↗

arXiv2025

The Isaac Newton Telescope Monitoring Survey of Local Group Dwarf Galaxies. VII. Long-Period Variable Stars in the Nearest Starburst Dwarf Galaxy, IC 10

To identify long-period variable (LPV) stars in IC10 - the nearest starburst galaxy of the Local Group (LG) - we conducted an optical monitoring survey using the 2.5-m Isaac Newton Telescope (INT) with the wide-field camera (WFC) in the i-band and V-band from 2015 to 2017. We created a photometric catalog for 53,579 stars within the area of CCD4 of WFC ($\sim$ 0.07 deg$^2$ corresponding to 13.5 kpc$^2$ at the distance of IC10), of which we classified 536 and 380 stars as long-period variable candidates (LPVs), mostly asymptotic giant branch stars (AGBs) and red supergiants (RSGs), within CCD4 and two half-light radii of IC10, respectively. By comparing our output catalog to the catalogs from Pan-STARRS, Spitzer Space Telescope, Hubble Space Telescope (HST), and carbon stars from the Canada-France-Hawai'i Telescope (CFHT) survey, we determined the success of our detection method. We recovered $\sim$ 73% of Spitzer's sources in our catalog and demonstrated that our survey successfully identified 43% of the variable stars found with Spitzer, and also retrieved 40% of the extremely dusty AGB stars among the Spitzer variables. In addition, we successfully identified $\sim$ 70% of HST variables in our catalog. Furthermore, we found all the confirmed LPVs that Gaia DR3 detected in IC10 among our identified LPVs. This paper is the first in a series on IC10, presenting the variable star survey methodology and the photometric catalog, available to the public through the Centre de Données Astronomiques de Strasbourg.

astro-ph.GA↗

arXiv2025

Learning 6-DoF Fine-grained Grasp Detection Based on Part Affordance Grounding

Robotic grasping is a fundamental ability for a robot to interact with the environment. Current methods focus on how to obtain a stable and reliable grasping pose in object level, while little work has been studied on part (shape)-wise grasping which is related to fine-grained grasping and robotic affordance. Parts can be seen as atomic elements to compose an object, which contains rich semantic knowledge and a strong correlation with affordance. However, lacking a large part-wise 3D robotic dataset limits the development of part representation learning and downstream applications. In this paper, we propose a new large Language-guided SHape grAsPing datasEt (named LangSHAPE) to promote 3D part-level affordance and grasping ability learning. From the perspective of robotic cognition, we design a two-stage fine-grained robotic grasping framework (named LangPartGPD), including a novel 3D part language grounding model and a part-aware grasp pose detection model, in which explicit language input from human or large language models (LLMs) could guide a robot to generate part-level 6-DoF grasping pose with textual explanation. Our method combines the advantages of human-robot collaboration and LLMs' planning ability using explicit language as a symbolic intermediate. To evaluate the effectiveness of our proposed method, we perform 3D part grounding and fine-grained grasp detection experiments on both simulation and physical robot settings, following language instructions across different degrees of textual complexity. Results show our method achieves competitive performance in 3D geometry fine-grained grounding, object affordance inference, and 3D part-aware grasping tasks. Our dataset and code are available on our project website https://sites.google.com/view/lang-shape

cs.RO↗

arXiv2025

Metallicity Distribution in M31 and M33 Based on the Tip-RGB Near-infrared Color

This study investigates the metallicity distribution in M31 and M33 by using the near-infrared color index $J-K$ of tip-red-giant-branch (TRGB) of the member stars from \cite{2021ApJ...907...18R} after removing the foreground dwarf stars by the near-infrared $J-H/H-K$ diagram as well as the Gaia astrometric measurements. We employ the Voronoi binning technique to divide the galaxy into sub-regions, the PN method to determine the TRGB position in the $J-K$/$K$ diagram, and the bootstrap method to estimate the uncertainties. The TRGB positions are calculated to be $J-K = 1.195 \pm 0.002$ and $1.100 \pm 0.003$, and $K = 17.615 \pm 0.007$ and $18.185 \pm 0.053$ for M31 and M33 respectively as an entity, consistent with previous results. The $J-K$ gradient of M31 is found to be $-$0.0055 kpc$^{-1}$ at $R_{\rm GC}=(0, 24)$ kpc and $-$0.0002 kpc$^{-1}$ at $R_{\rm GC}=(24, 150)$ kpc. Notably, two dust rings are evident at 11.5 kpc and 14.5 kpc, in agreement with previous studies. The $J-K$ of M33 is analyzed in four directions and generally shows a trend of gradually decreasing from the inside to the outside, however, it shows an increasing trend of 0.022 kpc$^{-1}$ in the inner 0$-$2 kpc in the west. Through the color$-$metallicity relation of TRGB, the metallicity gradient of M31 turns to be $-0.040 \pm 0.0012$ dex kpc$^{-1}$ with $R_{\rm GC}<30$ kpc and $-0.001 \pm 0.0002$ dex kpc$^{-1}$ with $R_{\rm GC}>30$ kpc, and for M33, $-$0.269 $\pm$ 0.0206 dex kpc$^{-1}$ with $R_{\rm GC}<9$ kpc.

astro-ph.GA↗

arXiv2025

NTIRE 2025 Challenge on Day and Night Raindrop Removal for Dual-Focused Images: Methods and Results

This paper reviews the NTIRE 2025 Challenge on Day and Night Raindrop Removal for Dual-Focused Images. This challenge received a wide range of impressive solutions, which are developed and evaluated using our collected real-world Raindrop Clarity dataset. Unlike existing deraining datasets, our Raindrop Clarity dataset is more diverse and challenging in degradation types and contents, which includes day raindrop-focused, day background-focused, night raindrop-focused, and night background-focused degradations. This dataset is divided into three subsets for competition: 14,139 images for training, 240 images for validation, and 731 images for testing. The primary objective of this challenge is to establish a new and powerful benchmark for the task of removing raindrops under varying lighting and focus conditions. There are a total of 361 participants in the competition, and 32 teams submitting valid solutions and fact sheets for the final testing phase. These submissions achieved state-of-the-art (SOTA) performance on the Raindrop Clarity dataset. The project can be found at https://lixinustc.github.io/CVPR-NTIRE2025-RainDrop-Competition.github.io/.

cs.CV↗

arXiv2025

DOMAC: Differentiable Optimization for High-Speed Multipliers and Multiply-Accumulators

Multipliers and multiply-accumulators (MACs) are fundamental building blocks for compute-intensive applications such as artificial intelligence. With the diminishing returns of Moore's Law, optimizing multiplier performance now necessitates process-aware architectural innovations rather than relying solely on technology scaling. In this paper, we introduce DOMAC, a novel approach that employs differentiable optimization for designing multipliers and MACs at specific technology nodes. DOMAC establishes an analogy between optimizing multi-staged parallel compressor trees and training deep neural networks. Building on this insight, DOMAC reformulates the discrete optimization challenge into a continuous problem by incorporating differentiable timing and area objectives. This formulation enables us to utilize existing deep learning toolkit for highly efficient implementation of the differentiable solver. Experimental results demonstrate that DOMAC achieves significant enhancements in both performance and area efficiency compared to state-of-the-art baselines and commercial IPs in multiplier and MAC designs.

cs.AR↗

arXiv2025

DiffuSE: Cross-Layer Design Space Exploration of DNN Accelerator via Diffusion-Driven Optimization

The proliferation of deep learning accelerators calls for efficient and cost-effective hardware design solutions, where parameterized modular hardware generator and electronic design automation (EDA) tools play crucial roles in improving productivity and final Quality-of-Results (QoR). To strike a good balance across multiple QoR of interest (e.g., performance, power, and area), the designers need to navigate a vast design space, encompassing tunable parameters for both hardware generator and EDA synthesis tools. However, the significant time for EDA tool invocations and complex interplay among numerous design parameters make this task extremely challenging, even for experienced designers. To address these challenges, we introduce DiffuSE, a diffusion-driven design space exploration framework for cross-layer optimization of DNN accelerators. DiffuSE leverages conditional diffusion models to capture the inverse, one-to-many mapping from QoR objectives to parameter combinations, allowing for targeted exploration within promising regions of the design space. By carefully selecting the conditioning QoR values, the framework facilitates an effective trade-off among multiple QoR metrics in a sample-efficient manner. Experimental results under 7nm technology demonstrate the superiority of the proposed framework compared to previous arts.

cs.AR↗

arXiv2025

MegaTTS 3: Sparse Alignment Enhanced Latent Diffusion Transformer for Zero-Shot Speech Synthesis

While recent zero-shot text-to-speech (TTS) models have significantly improved speech quality and expressiveness, mainstream systems still suffer from issues related to speech-text alignment modeling: 1) models without explicit speech-text alignment modeling exhibit less robustness, especially for hard sentences in practical applications; 2) predefined alignment-based models suffer from naturalness constraints of forced alignments. This paper introduces \textit{MegaTTS 3}, a TTS system featuring an innovative sparse alignment algorithm that guides the latent diffusion transformer (DiT). Specifically, we provide sparse alignment boundaries to MegaTTS 3 to reduce the difficulty of alignment without limiting the search space, thereby achieving high naturalness. Moreover, we employ a multi-condition classifier-free guidance strategy for accent intensity adjustment and adopt the piecewise rectified flow technique to accelerate the generation process. Experiments demonstrate that MegaTTS 3 achieves state-of-the-art zero-shot TTS speech quality and supports highly flexible control over accent intensity. Notably, our system can generate high-quality one-minute speech with only 8 sampling steps. Audio samples are available at https://sditdemo.github.io/sditdemo/.

eess.AS↗

arXiv2025

Extinction Distributions in Nearby Star-resolved Galaxies. I. M31

An extinction distribution of the Andromeda Galaxy (M31) is constructed with member stars as tracers by fitting multiband photometric data from UKIRT/WFCAM, PS1, and Gaia DR3. The resulting extinction distribution covers approximately 10 deg$^2$ of M31 with a resolution of approximately 50 arcsec, providing the largest coverage to date based on stellar observations. The derived average extinction, $A_V = 1.17$ mag, agrees well with previous studies. To account for foreground extinction, an extinction map of the Milky Way toward M31 with a resolution of $\sim$ 1.7 arcmin is also constructed, yielding an average extinction of $A_V \approx 0.185$ mag. The results offer a valuable tool for extinction correction in future observations, such as those from the China Space Station Telescope, and provide insights for improving dust models based on the spatial distribution of dust in galaxies like M31.

astro-ph.GA↗

arXiv2025

Learning Nash Equilibrial Hamiltonian for Two-Player Collision-Avoiding Interactions

We consider the problem of learning Nash equilibrial policies for two-player risk-sensitive collision-avoiding interactions. Solving the Hamilton-Jacobi-Isaacs equations of such general-sum differential games in real time is an open challenge due to the discontinuity of equilibrium values on the state space. A common solution is to learn a neural network that approximates the equilibrium Hamiltonian for given system states and actions. The learning, however, is usually supervised and requires a large amount of sample equilibrium policies from different initial states in order to mitigate the risks of collisions. This paper claims two contributions towards more data-efficient learning of equilibrium policies: First, instead of computing Hamiltonian through a value network, we show that the equilibrium co-states have simple structures when collision avoidance dominates the agents' loss functions and system dynamics is linear, and therefore are more data-efficient to learn. Second, we introduce theory-driven active learning to guide data sampling, where the acquisition function measures the compliance of the predicted co-states to Pontryagin's Maximum Principle. On an uncontrolled intersection case, the proposed method leads to more generalizable approximation of the equilibrium policies, and in turn, lower collision probabilities, than the state-of-the-art under the same data acquisition budget.

cs.RO↗

arXiv2025

Beyond Overfitting: Doubly Adaptive Dropout for Generalizable AU Detection

Facial Action Units (AUs) are essential for conveying psychological states and emotional expressions. While automatic AU detection systems leveraging deep learning have progressed, they often overfit to specific datasets and individual features, limiting their cross-domain applicability. To overcome these limitations, we propose a doubly adaptive dropout approach for cross-domain AU detection, which enhances the robustness of convolutional feature maps and spatial tokens against domain shifts. This approach includes a Channel Drop Unit (CD-Unit) and a Token Drop Unit (TD-Unit), which work together to reduce domain-specific noise at both the channel and token levels. The CD-Unit preserves domain-agnostic local patterns in feature maps, while the TD-Unit helps the model identify AU relationships generalizable across domains. An auxiliary domain classifier, integrated at each layer, guides the selective omission of domain-sensitive features. To prevent excessive feature dropout, a progressive training strategy is used, allowing for selective exclusion of sensitive features at any model layer. Our method consistently outperforms existing techniques in cross-domain AU detection, as demonstrated by extensive experimental evaluations. Visualizations of attention maps also highlight clear and meaningful patterns related to both individual and combined AUs, further validating the approach's effectiveness.

cs.CV↗

arXiv2025

Identification of Star Clusters in M31 from PAndAS Images Based on Deep Learning

The identification of star clusters holds significant importance in studying galaxy formation and evolution history. However, the task of swiftly and accurately identifying star clusters from vast amounts of photometric images presents an immense challenge. To address these difficulties, we employ deep learning models for image classification to identify young disk star clusters in M31 from the Pan-Andromeda Archaeological Survey (PAndAS) images. For training, validation, and testing, we utilize the Panchromatic Hubble Andromeda Treasury (PHAT) survey catalogs. We evaluate the performance of various deep learning models, using different classification thresholds and limiting magnitudes. Our findings indicate that the ResNet-50 model exhibits the highest overall accuracy. Moreover, using brighter limiting magnitudes and increasing the classification thresholds can effectively enhance the accuracy and precision of cluster identification. Through our experiments, we found that the model achieves optimal performance when the limiting magnitude is set to brighter than 21 mag. Based on this, we constructed a training dataset with magnitudes less than 21 mag and trained a second ResNet-50 model. This model achieved a purity of 89.30%, a recall of 73.55%, and an F1 score of 80.66% when the classification threshold was set to 0.669. Applying the second model to all sources in the PAndAS fields within a projected radius of 30 kpc from the center of M31, we identified 2,228 new unique star cluster candidates. We conducted visual inspections to validate the results produced by our automated methods, and we ultimately obtained 1,057 star cluster candidates, of which 745 are newly identified.

astro-ph.GA↗

arXiv2025

Parametric Value Approximation for General-sum Differential Games with State Constraints

General-sum differential games can approximate values solved by Hamilton-Jacobi-Isaacs (HJI) equations for efficient inference when information is incomplete. However, solving such games through conventional methods encounters the curse of dimensionality (CoD). Physics-informed neural networks (PINNs) offer a scalable approach to alleviate the CoD and approximate values, but there exist convergence issues for value approximations through vanilla PINNs when state constraints lead to values with large Lipschitz constants, particularly in safety-critical applications. In addition to addressing CoD, it is necessary to learn a generalizable value across a parametric space of games, rather than training multiple ones for each specific player-type configuration. To overcome these challenges, we propose a Hybrid Neural Operator (HNO), which is an operator that can map parameter functions for games to value functions. HNO leverages informative supervised data and samples PDE-driven data across entire spatial-temporal space for model refinement. We evaluate HNO on 9D and 13D scenarios with nonlinear dynamics and state constraints, comparing it against a Supervised Neural Operator (a variant of DeepONet). Under the same computational budget and training data, HNO outperforms SNO for safety performance. This work provides a step toward scalable and generalizable value function approximation, enabling real-time inference for complex human-robot or multi-agent interactions.

cs.RO↗

arXiv2025

New Red Supergiant Stars in the other side of our Galaxy

Red supergiant stars (RSGs) are massive stars in a late stage of evolution, crucial for understanding stellar life cycles and Galactic structure. However, RSGs on the far side of our Galaxy have been underexplored due to observational challenges. In this study, we introduce a novel method and present a new catalogue comprising 474 RSGs situated on the far side of the Milky Way, sourced from the OGLE-III catalogue of Variable Stars (OIII-CVS). The identification of these RSGs was made possible by analyzing the granulation parameters extracted from OGLE I-band time-series data and the stellar parameters from Gaia DR3. Additionally, we estimate the distances to these RSGs using an empirical relation between their characteristic amplitude, absolute magnitude, and intrinsic color, achieving a distance uncertainty of 13%. These newly identified RSGs are distributed at Galactocentric distances between 0 and 30kpc, and reside roughly 1 to 4kpc above and below the Galactic plane. This distribution provides new insights into the structure of the Milky Way, particularly at its outer boundaries. Our results reveal that the vertical distribution of these RSGs is consistent with the flare structure of the Galactic disk, confirming that the far side of the Milky Way exhibits a similar flaring pattern to the near side. This catalogue offers a valuable resource for future detailed studies of RSGs and contributes to a more comprehensive understanding of the Galactic structure and stellar evolution.

astro-ph.GA↗