Searcharxiv⌕ Search

arXiv subjects

Gang Chen

Publications and source records attributed to Gang Chen.

At least 109 records · Page 6Linked to original sources
arXiv2025

Teaching Large Language Models to Maintain Contextual Faithfulness via Synthetic Tasks and Reinforcement Learning

Teaching large language models (LLMs) to be faithful in the provided context is crucial for building reliable information-seeking systems. Therefore, we propose a systematic framework, CANOE, to reduce faithfulness hallucinations of LLMs across different downstream tasks without human annotations. Specifically, we first synthesize short-form question-answering (QA) data with four diverse tasks to construct high-quality and easily verifiable training data without human annotation. Also, we propose Dual-GRPO, a rule-based reinforcement learning method that includes three tailored rule-based rewards derived from synthesized short-form QA data, while simultaneously optimizing both short-form and long-form response generation. Notably, Dual-GRPO eliminates the need to manually label preference data to train reward models and avoids over-optimizing short-form generation when relying only on the synthesized short-form QA data. Experimental results show that CANOE greatly improves the faithfulness of LLMs across 11 different tasks, even outperforming the most advanced LLMs, e.g., GPT-4o and OpenAI o1.

cs.CL↗
arXiv2025

DynamicRTL: RTL Representation Learning for Dynamic Circuit Behavior

There is a growing body of work on using Graph Neural Networks (GNNs) to learn representations of circuits, focusing primarily on their static characteristics. However, these models fail to capture circuit runtime behavior, which is crucial for tasks like circuit verification and optimization. To address this limitation, we introduce DR-GNN (DynamicRTL-GNN), a novel approach that learns RTL circuit representations by incorporating both static structures and multi-cycle execution behaviors. DR-GNN leverages an operator-level Control Data Flow Graph (CDFG) to represent Register Transfer Level (RTL) circuits, enabling the model to capture dynamic dependencies and runtime execution. To train and evaluate DR-GNN, we build the first comprehensive dynamic circuit dataset, comprising over 6,300 Verilog designs and 63,000 simulation traces. Our results demonstrate that DR-GNN outperforms existing models in branch hit prediction and toggle rate prediction. Furthermore, its learned representations transfer effectively to related dynamic circuit tasks, achieving strong performance in power estimation and assertion prediction.

cs.LG↗
arXiv2025

UI-Evol: Automatic Knowledge Evolving for Computer Use Agents

External knowledge has played a crucial role in the recent development of computer use agents. We identify a critical knowledge-execution gap: retrieved knowledge often fails to translate into effective real-world task execution. Our analysis shows even 90% correct knowledge yields only 41% execution success rate. To bridge this gap, we propose UI-Evol, a plug-and-play module for autonomous GUI knowledge evolution. UI-Evol consists of two stages: a Retrace Stage that extracts faithful objective action sequences from actual agent-environment interactions, and a Critique Stage that refines existing knowledge by comparing these sequences against external references. We conduct comprehensive experiments on the OSWorld benchmark with the state-of-the-art Agent S2. Our results demonstrate that UI-Evol not only significantly boosts task performance but also addresses a previously overlooked issue of high behavioral standard deviation in computer use agents, leading to superior performance on computer use tasks and substantially improved agent reliability.

cs.HC↗
arXiv2025

Generative AI for Healthcare: Fundamentals, Challenges, and Perspectives

Generative Artificial Intelligence (GenAI) is taking the world by storm. It promises transformative opportunities for advancing and disrupting existing practices, including healthcare. From large language models (LLMs) for clinical note synthesis and conversational assistance to multimodal systems that integrate medical imaging, electronic health records, and genomic data for decision support, GenAI is transforming the practice of medicine and the delivery of healthcare, such as diagnosis and personalized treatments, with great potential in reducing the cognitive burden on clinicians, thereby improving overall healthcare delivery. However, GenAI deployment in healthcare requires an in-depth understanding of healthcare tasks and what can and cannot be achieved. In this paper, we propose a data-centric paradigm in the design and deployment of GenAI systems for healthcare. Specifically, we reposition the data life cycle by making the medical data ecosystem as the foundational substrate for generative healthcare systems. This ecosystem is designed to sustainably support the integration, representation, and retrieval of diverse medical data and knowledge. With effective and efficient data processing pipelines, such as semantic vector search and contextual querying, it enables GenAI-powered operations for upstream model components and downstream clinical applications. Ultimately, it not only supplies foundation models with high-quality, multimodal data for large-scale pretraining and domain-specific fine-tuning, but also serves as a knowledge retrieval backend to support task-specific inference via the agentic layer. The ecosystem enables the deployment of GenAI for high-quality and effective healthcare delivery.

cs.AI↗
arXiv2025

HGraphScale: Hierarchical Graph Learning for Autoscaling Microservice Applications in Container-based Cloud Computing

Microservice architecture has become a dominant paradigm in application development due to its advantages of being lightweight, flexible, and resilient. Deploying microservice applications in the container-based cloud enables fine-grained elastic resource allocation. Autoscaling is an effective approach to dynamically adjust the resource provisioned to containers. However, the intricate microservice dependencies and the deployment scheme of the container-based cloud bring extra challenges of resource scaling. This article proposes a novel autoscaling approach named HGraphScale. In particular, HGraphScale captures microservice dependencies and the deployment scheme by a newly designed hierarchical graph neural network, and makes effective scaling actions for rapidly changing user requests workloads. Extensive experiments based on real-world traces of user requests are conducted to evaluate the effectiveness of HGraphScale. The experiment results show that the HGraphScale outperforms existing state-of-the-art autoscaling approaches by reducing at most 80.16\% of the average response time under a certain VM rental budget of application providers.

cs.DC↗
arXiv2025

Merge-of-Thought Distillation

Efficient reasoning distillation for long chain-of-thought (CoT) models is increasingly constrained by the assumption of a single oracle teacher, despite the practical availability of multiple candidate teachers and growing CoT corpora. We revisit teacher selection and observe that different students have different "best teachers," and even for the same student, the best teacher can vary across datasets. Therefore, to unify multiple teachers' reasoning abilities into a student to overcome conflicts among various teachers' supervision, we propose Merge-of-Thought Distillation (MoT), a lightweight framework that alternates between teacher-specific supervised fine-tuning branches and weight-space merging of the resulting student variants. On competition math benchmarks, using only about 200 CoT samples, applying MoT to a Qwen3-14B student surpasses strong models including Deepseek-R1, Qwen3-32B, and OpenAI-O1, demonstrating substantial gains. Besides, MoT consistently outperforms the best single-teacher distillation, improves general reasoning beyond mathematics while reducing catastrophic forgetting, and shows robustness to distribution-shifted and peer-level teachers. Finally, we have demonstrated MoT possesses consensus CoT by eliminating teacher-specific inductive biases and inter-teacher conflicts while repeatedly reinforcing the learning of consensus reasoning features. These results position MoT as a simple, effective route to efficiently distilling long CoT capabilities from diverse teachers into compact students.

cs.LG↗
arXiv2025

The gravitational Compton amplitude from flat and curved spacetimes at second post-Minkowskian order

We utilize various computational techniques in flat and curved backgrounds to calculate the classical gravitational Compton amplitude up to the second post-Minkowskian order. Our novel result supports the use of worldline quantum field theory in non-trivial background spacetimes to obtain new theoretical insights that can both enhance computational efficiency and provide useful cross-checks of results, particularly in the context of classical binary black hole mergers.

hep-th↗
arXiv2025

Data-Augmented Machine Learning for Predicting Biomass-Derived Hard Carbon Anode Performance in Sodium-Ion Batteries

Biomass-derived hard carbon has become the most promising anode material for sodium-ion batteries (SIBs) due to its high capacity and excellent cycling stability. However, the effects of synthesis parameters and structural features on hard carbon's (HC) electrochemical performance are still unclear, requiring time-consuming and resource-intensive experimental investigations. Machine learning (ML) offers a promising solution by training on large datasets to predict hard carbon performance more efficiently, saving time and resources. In this study, four ML models were used to predict the capacity and initial Coulombic efficiency (ICE) of HC. Data augmentation based on the TabPFN technique was employed to improve model robustness under limited data conditions, and the relationships between features and electrochemical performance were examined. Notably, the XGBoost model achieved an R2 of 0.854 and an RMSE of 23.290 mAh g-1 for capacity prediction, and an R2 of 0.868 and an RMSE of 3.813% for ICE prediction. Shapley Additive Explanations (SHAP) and Partial Dependence Plot (PDP) analyses identified carbonization temperature (Temperature_2) as the most important factor influencing both capacity and ICE. Furthermore, we used bamboo as the precursor to synthesize four hard carbons based on the predictive approach. The electrochemical performance of these samples closely matched our predictions. By leveraging machine-learning approach, this study provides an efficient framework for accelerating the screening process of biomass-derived hard carbon candidates.

physics.chem-ph↗
arXiv2025

New Physics Search at the CEPC: a General Perspective

The Circular Electron-Positron Collider (CEPC), a proposed next-generation Higgs factory, provides new opportunities to explore physics beyond the Standard Model (SM). With its clean electron-positron collision environment and the ability to collect large samples of Higgs, W, and Z bosons, the CEPC enables precision measurements and searches for new physics. This white paper outlines the CEPC's discovery potential, including studies of exotic decays of the Higgs, Z, and top quarks, dark matter and dark sector phenomena, long-lived particles, supersymmetry, and neutrino-related signatures. Advanced detector technologies and reconstruction techniques, such as one-to-one correspondence reconstruction and jet origin identification, significantly improve sensitivity to rare and weakly interacting processes. The CEPC is particularly well suited to probe the electroweak phase transition and test models of electroweak baryogenesis and dark sector interactions. In addition, global fit analyses highlight the CEPC's complementary role in constraining a wide range of new physics scenarios. These features position the CEPC as a powerful tool for exploring the next frontier in fundamental particle physics in the post-Higgs discovery era.

hep-ex↗
arXiv2025

Evaporative Refrigeration Effect in Evaporation and Condensation between Two Parallel Plates

It is well-known that evaporation can lead to cooling. However, little is known that evaporation can actually create a refrigeration effect, i.e., the vapor phase temperature can drop below the temperature of the cooling wall. This possibility was recently pointed out via modeling based on an approximate quasi-continuum approach. This work examines this effect rigorously by studying evaporation and condensation between two parallel plates by coupling the solution of the Boltzmann transport equation in the vapor phase with the continuum treatments in both liquid films. Numerical results show that the vapor phase temperature at the evaporating side can be much lower than the coldest wall temperature at the condensing surface, reaffirming the evaporative refrigeration effect. The present work further reveals that this effect is caused by two mechanisms. While the dominant mechanism is the asymmetry in the molecular distribution between the outgoing and the incoming molecules at the interface, additional cooling occurs within the Knudsen layer due to the sudden expansion, similar to the Joule-Thomson effect, although with subtle differences in that the interfacial expansion is not an isenthalpic process. The impacts of key parameters, including liquid thickness, Knudsen number, and accommodation coefficient, are investigated. The numerical simulation shows that with a thicker vapor, a thinner liquid film, and a larger accommodation coefficient, leads to stronger evaporative refrigeration effect. This work will motivate future experiments to further confirm this prediction and explore its potential applications in air-conditioning, refrigeration, and membrane distillation.

cond-mat.stat-mech↗
arXiv2025

GATEAU: Selecting Influential Samples for Long Context Alignment

Aligning large language models to handle instructions with extremely long contexts has yet to be fully investigated. Previous studies have attempted to scale up the available data volume by synthesizing long instruction-following samples, as constructing such a dataset tends to be challenging for annotators. However, a lack of a well-defined strategy for ensuring data quality may introduce low-quality samples and restrict the model's performance. Thus, we propose GATEAU, a novel framework to address the unique challenge of long context alignment by identifying the influential samples enriched with long-range dependency relations. Specifically, GATEAU measures the long-range dependencies from two essential aspects: the difficulty of generating target responses due to the long-range dependencies, and the difficulty of understanding long inputs due to such dependencies. Comprehensive experiments indicate that GATEAU effectively identifies influential samples, and the model trained on these selected samples exhibits better instruction-following and long-context understanding capabilities.

cs.CL↗
arXiv2025

GFS: A Preemption-aware Scheduling Framework for GPU Clusters with Predictive Spot Instance Management

The surge in large language models (LLMs) has fundamentally reshaped the landscape of GPU usage patterns, creating an urgent need for more efficient management strategies. While cloud providers employ spot instances to reduce costs for low-priority (LP) tasks, existing schedulers still grapple with high eviction rates and lengthy queuing times. To address these limitations, we present GFS, a novel preemptive scheduling framework that enhances service-level objective (SLO) compliance for high-priority (HP) tasks while minimizing preemptions to LP tasks. Firstly, GFS utilizes a lightweight forecasting model that predicts GPU demand among different tenants, enabling proactive resource management. Secondly, GFS employs a dynamic allocation mechanism to adjust the spot quota for LP tasks with guaranteed durations. Lastly, GFS incorporates a preemptive scheduling policy that prioritizes HP tasks while minimizing the impact on LP tasks. We demonstrate the effectiveness of GFS through both real-world implementation and simulations. The results show that GFS reduces eviction rates by 33.0\%, and cuts queuing delays by 44.1\% for LP tasks. Furthermore, GFS enhances the GPU allocation rate by up to 22.8\% in real production clusters. In a production cluster of more than 10,000 GPUs, GFS yields roughly \$459,715 in monthly benefits.

cs.DC↗
arXiv2025

Very High-order Compact Gas-kinetic Scheme With Discontinuity Feedback Factor

This paper presents a robust and efficient very high-order scheme for compressible flow simulation, addressing critical limitations of existing high-order methods. The proposed scheme combines the compact gas-kinetic scheme (CGKS) with an adaptive stencil extension reconstruction with discontinuity feedback factor (ASE-DFF), achieving significant improvements in both robustness and computational efficiency. Traditional weighted essentially non-oscillatory (WENO) schemes suffer from reduced robustness at higher order and require costly smoothness indicators for large stencils. Meanwhile, compact methods based on Discontinuous Galerkin (DG) and Flux Reconstruction (FR) struggle with poor time-marching efficiency. In contrast, the ASE-DFF-CGKS introduces two key innovations: (1) a unified framework enabling arbitrarily high-order compact gas-kinetic scheme without sacrificing large CFL number, and (2) a discontinuity feedback factor that eliminates the need for expensive smoothness indicator calculations while essentially keeping first-order robustness near discontinuities. The scheme's advantages are demonstrated through benchmark simulations. It maintains a CFL number above 0.5 for up to 9th-order case, unlike conventional compact methods that restrict a CFL less than 0.05. Also it delivers high-resolution results for flow involving strong shock and rarefaction wave. This work provides a practically impactful solution for high-fidelity compressible flow simulation, balancing computational efficiency, high-order accuracy and robustness in challenging flow regimes.

physics.comp-ph↗
arXiv2025

CRouting: Reducing Expensive Distance Calls in Graph-Based Approximate Nearest Neighbor Search

Approximate nearest neighbor search (ANNS) is a crucial problem in information retrieval and AI applications. Recently, there has been a surge of interest in graph-based ANNS algorithms due to their superior efficiency and accuracy. However, the repeated computation of distances in high-dimensional spaces constitutes the primary time cost of graph-based methods. To accelerate the search, we propose a novel routing strategy named CRouting, which bypasses unnecessary distance computations by exploiting the angle distributions of high-dimensional vectors. CRouting is designed as a plugin to optimize existing graph-based search with minimal code modifications. Our experiments show that CRouting reduces the number of distance computations by up to 41.5% and boosts queries per second by up to 1.48$\times$ on two predominant graph indexes, HNSW and NSG. Code is publicly available at https://github.com/ISCS-ZJU/CRouting.

cs.DB↗
arXiv2025

Robust ID-Specific Face Restoration via Alignment Learning

The latest developments in Face Restoration have yielded significant advancements in visual quality through the utilization of diverse diffusion priors. Nevertheless, the uncertainty of face identity introduced by identity-obscure inputs and stochastic generative processes remains unresolved. To address this challenge, we present Robust ID-Specific Face Restoration (RIDFR), a novel ID-specific face restoration framework based on diffusion models. Specifically, RIDFR leverages a pre-trained diffusion model in conjunction with two parallel conditioning modules. The Content Injection Module inputs the severely degraded image, while the Identity Injection Module integrates the specific identity from a given image. Subsequently, RIDFR incorporates Alignment Learning, which aligns the restoration results from multiple references with the same identity in order to suppress the interference of ID-irrelevant face semantics (e.g. pose, expression, make-up, hair style). Experiments demonstrate that our framework outperforms the state-of-the-art methods, reconstructing high-quality ID-specific results with high identity fidelity and demonstrating strong robustness.

cs.CV↗
arXiv2025

SpecVLM: Enhancing Speculative Decoding of Video LLMs via Verifier-Guided Token Pruning

Video large language models (Vid-LLMs) have shown strong capabilities in understanding video content. However, their reliance on dense video token representations introduces substantial memory and computational overhead in both prefilling and decoding. To mitigate the information loss of recent video token reduction methods and accelerate the decoding stage of Vid-LLMs losslessly, we introduce SpecVLM, a training-free speculative decoding (SD) framework tailored for Vid-LLMs that incorporates staged video token pruning. Building on our novel finding that the draft model's speculation exhibits low sensitivity to video token pruning, SpecVLM prunes up to 90% of video tokens to enable efficient speculation without sacrificing accuracy. To achieve this, we performs a two-stage pruning process: Stage I selects highly informative tokens guided by attention signals from the verifier (target model), while Stage II prunes remaining redundant ones in a spatially uniform manner. Extensive experiments on four video understanding benchmarks demonstrate the effectiveness and robustness of SpecVLM, which achieves up to 2.68$\times$ decoding speedup for LLaVA-OneVision-72B and 2.11$\times$ speedup for Qwen2.5-VL-32B. Code is available at https://github.com/zju-jiyicheng/SpecVLM.

cs.CV↗
arXiv2025

Recommendations for Best Practices for Data Preservation and Open Science in HEP

These recommendations are the result of reflections by scientists and experts who are, or have been, involved in the preservation of high-energy physics data. The work has been done under the umbrella of the Data Lifecycle panel of the International Committee of Future Accelerators (ICFA), drawing on the expertise of a wide range of stakeholders. A key indicator of success in the data preservation efforts is the long-term usability of the data. Experience shows that achieving this requires providing a rich set of information in various forms, which can only be effectively collected and preserved during the period of active data use. The recommendations are intended to be actionable by the indicated actors and specific to the particle physics domain. They cover a wide range of actions, many of which are interdependent. These dependencies are indicated within the recommendations and can be used as a road map to guide implementation efforts. These recommendations are best accessed and viewed through the web application, see https://icfa-data-best-practices.app.cern.ch/

hep-ex↗