Ask or search anything...
Disclosing Submillimeter Galaxy Formation: Mergers or Secular Evolution?
University of Science and Technology of ChinaUtilizing multi-band JWST observations, this research reveals that high-redshift submillimeter galaxies primarily form through secular evolution and internal processes rather than major mergers, uncovering a significant population of central stellar structures that do not conform to established local galaxy classifications.
View blog
QDepth-VLA: Quantized Depth Prediction as Auxiliary Supervision for Vision-Language-Action Models
Chinese Academy of SciencesResearchers at the Chinese Academy of Sciences developed QDepth-VLA, a framework that enhances Vision-Language-Action (VLA) models with robust 3D geometric understanding through quantized depth prediction as auxiliary supervision. This approach improves performance on fine-grained robotic manipulation tasks, achieving up to 29.7% higher success rates on complex simulated tasks and 20.0% gains in real-world pick-and-place scenarios compared to existing baselines.
View blog
Pentest-R1: Towards Autonomous Penetration Testing Reasoning Optimized via Two-Stage Reinforcement Learning
Automating penetration testing is crucial for enhancing cybersecurity, yet current Large Language Models (LLMs) face significant limitations in this domain, including poor error handling, inefficient reasoning, and an inability to perform complex end-to-end tasks autonomously. To address these challenges, we introduce Pentest-R1, a novel framework designed to optimize LLM reasoning capabilities for this task through a two-stage reinforcement learning pipeline. We first construct a dataset of over 500 real-world, multi-step walkthroughs, which Pentest-R1 leverages for offline reinforcement learning (RL) to instill foundational attack logic. Subsequently, the LLM is fine-tuned via online RL in an interactive Capture The Flag (CTF) environment, where it learns directly from environmental feedback to develop robust error self-correction and adaptive strategies. Our extensive experiments on the Cybench and AutoPenBench benchmarks demonstrate the framework's effectiveness. On AutoPenBench, Pentest-R1 achieves a 24.2\% success rate, surpassing most state-of-the-art models and ranking second only to Gemini 2.5 Flash. On Cybench, it attains a 15.0\% success rate in unguided tasks, establishing a new state-of-the-art for open-source LLMs and matching the performance of top proprietary models. Ablation studies confirm that the synergy of both training stages is critical to its success.
CodeR: Issue Resolving with Multi-Agent and Task Graphs
Peking UniversityDeveloped by Huawei Co., Ltd., CODER introduces a multi-agent framework guided by pre-defined task graphs to automate GitHub issue resolution. The system achieved a 28.33% resolved rate on SWE-bench lite, establishing a new state-of-the-art for the benchmark.
View blog
STSA: Federated Class-Incremental Learning via Spatial-Temporal Statistics Aggregation
Federated Class-Incremental Learning (FCIL) enables Class-Incremental
Learning (CIL) from distributed data. Existing FCIL methods typically integrate
old knowledge preservation into local client training. However, these methods
cannot avoid spatial-temporal client drift caused by data heterogeneity and
often incur significant computational and communication overhead, limiting
practical deployment. To address these challenges simultaneously, we propose a
novel approach, Spatial-Temporal Statistics Aggregation (STSA), which provides
a unified framework to aggregate feature statistics both spatially (across
clients) and temporally (across stages). The aggregated feature statistics are
unaffected by data heterogeneity and can be used to update the classifier in
closed form at each stage. Additionally, we introduce STSA-E, a
communication-efficient variant with theoretical guarantees, achieving similar
performance to STSA-E with much lower communication overhead. Extensive
experiments on three widely used FCIL datasets, with varying degrees of data
heterogeneity, show that our method outperforms state-of-the-art FCIL methods
in terms of performance, flexibility, and both communication and computation
efficiency.
Beyond Single LLMs: Enhanced Code Generation via Multi-Stage Performance-Guided LLM Orchestration
01 Oct 2025
University of GeorgiaWhile Large Language Models (LLMs) have become the predominant paradigm for automated code generation, current single-model approaches fundamentally ignore the heterogeneous computational strengths that different models exhibit across programming languages, algorithmic domains, and development stages. This paper challenges the single-model convention by introducing a multi-stage, performance-guided orchestration framework that dynamically routes coding tasks to the most suitable LLMs within a structured generate-fix-refine workflow. Our approach is grounded in a comprehensive empirical study of 17 state-of-the-art LLMs across five programming languages (Python, Java, C++, Go, and Rust) using HumanEval-X benchmark. The study, which evaluates both functional correctness and runtime performance metrics (execution time, mean/max memory utilization, and CPU efficiency), reveals pronounced performance heterogeneity by language, development stage, and problem category. Guided by these empirical insights, we present PerfOrch, an LLM agent that orchestrates top-performing LLMs for each task context through stage-wise validation and rollback mechanisms. Without requiring model fine-tuning, PerfOrch achieves substantial improvements over strong single-model baselines: average correctness rates of 96.22% and 91.37% on HumanEval-X and EffiBench-X respectively, surpassing GPT-4o's 78.66% and 49.11%. Beyond correctness gains, the framework delivers consistent performance optimizations, improving execution time for 58.76% of problems with median speedups ranging from 17.67% to 27.66% across languages on two benchmarks. The framework's plug-and-play architecture ensures practical scalability, allowing new LLMs to be profiled and integrated seamlessly, thereby offering a paradigm for production-grade automated software engineering that adapts to the rapidly evolving generative AI landscape.
Fine-Tuning Jailbreaks under Highly Constrained Black-Box Settings: A Three-Pronged Approach
With the rapid advancement of large language models (LLMs), ensuring their safe use becomes increasingly critical. Fine-tuning is a widely used method for adapting models to downstream tasks, yet it is vulnerable to jailbreak attacks. However, most existing studies focus on overly simplified attack scenarios, limiting their practical relevance to real-world defense settings. To make this risk concrete, we present a three-pronged jailbreak attack and evaluate it against provider defenses under a dataset-only black-box fine-tuning interface. In this setting, the attacker can only submit fine-tuning data to the provider, while the provider may deploy defenses across stages: (1) pre-upload data filtering, (2) training-time defensive fine-tuning, and (3) post-training safety audit. Our attack combines safety-styled prefix/suffix wrappers, benign lexical encodings (underscoring) of sensitive tokens, and a backdoor mechanism, enabling the model to learn harmful behaviors while individual datapoints appear innocuous. Extensive experiments demonstrate the effectiveness of our approach. In real-world deployment, our method successfully jailbreaks GPT-4.1 and GPT-4o on the OpenAI platform with attack success rates above 97% for both models. Our code is available at this https URL.
Let AI Read First: Enhancing Reading Abilities for Individuals with Dyslexia through Artificial Intelligence
Zhejiang UniversityResearchers developed LARF (Let AI Read First), an AI-powered system leveraging GPT-4 to annotate important information in texts with visual cues, improving reading performance and subjective experience for individuals with dyslexia, particularly those with more severe conditions. The system enhanced objective detail retrieval and comprehension while preserving original content.
View blog
CodeV: Issue Resolving with Visual Data
23 Dec 2024
Shandong UniversityLarge Language Models (LLMs) have advanced rapidly in recent years, with their applications in software engineering expanding to more complex repository-level tasks. GitHub issue resolving is a key challenge among these tasks. While recent approaches have made progress on this task, they focus on textual data within issues, neglecting visual data. However, this visual data is crucial for resolving issues as it conveys additional knowledge that text alone cannot. We propose CodeV, the first approach to leveraging visual data to enhance the issue-resolving capabilities of LLMs. CodeV resolves each issue by following a two-phase process: data processing and patch generation. To evaluate CodeV, we construct a benchmark for visual issue resolving, namely Visual SWE-bench. Through extensive experiments, we demonstrate the effectiveness of CodeV, as well as provide valuable insights into leveraging visual data to resolve GitHub issues.
Ultracompact high-Q whispering gallery mode microresonator in a non-closed waveguide path
15 Oct 2025
Integrated photonic circuits are foundational for versatile applications, where high-performance traveling-wave optical resonators are critical. Conventional whispering-gallery mode microresonators (WGMRs) confine light in closed-loop waveguide paths, thus inevitably occupy large footprints. Here, we report an ultracompact high loaded Q silicon photonic WGMR in an open curved path instead. By leveraging spatial mode multiplexing, low-loss mode converter-based photonic routers enable reentrant photon recycling in a single non-closed waveguide. The fabricated device achieves a measured loaded Q-factor of 1.78*10^5 at 1554.3 nm with a 1.05 nm free spectral range in a ultracompact footprint of 0.00137 mm^2-6*smaller than standard WGMRs while delivering 100*higher Q-factor than photonic crystal counterparts. This work pioneers dense integration of high-performance WGMR arrays through open-path mode recirculation.
Are Your LLM-based Text-to-SQL Models Secure? Exploring SQL Injection via Backdoor Attacks
08 Sep 2025
Cornell UniversityToxicSQL introduces a framework for investigating and exploiting SQL injection vulnerabilities in LLM-based Text-to-SQL models through backdoor attacks. The work demonstrates that these models can be trained with low poisoning rates to generate malicious, executable SQL queries while retaining normal performance on benign inputs, thereby exposing critical security flaws in database interaction systems.
View blog
PanGu-Coder2: Boosting Large Language Models for Code with Ranking Feedback
27 Jul 2023
Huawei Cloud Co., Ltd. researchers developed PanGu-Coder2, a Code LLM fine-tuned with the RRTF framework, achieving 61.64% pass@1 on HumanEval and outperforming prior open-source models as well as several larger commercial models.
View blog
A Comprehensive Survey on Magnetic Resonance Image Reconstruction
10 Mar 2025
Magnetic resonance imaging (MRI) reconstruction is a fundamental task aimed
at recovering high-quality images from undersampled or low-quality MRI data.
This process enhances diagnostic accuracy and optimizes clinical applications.
In recent years, deep learning-based MRI reconstruction has made significant
progress. Advancements include single-modality feature extraction using
different network architectures, the integration of multimodal information, and
the adoption of unsupervised or semi-supervised learning strategies. However,
despite extensive research, MRI reconstruction remains a challenging problem
that has yet to be fully resolved. This survey provides a systematic review of
MRI reconstruction methods, covering key aspects such as data acquisition and
preprocessing, publicly available datasets, single and multi-modal
reconstruction models, training strategies, and evaluation metrics based on
image reconstruction and downstream tasks. Additionally, we analyze the major
challenges in this field and explore potential future directions.
SWE-bench-java: A GitHub Issue Resolving Benchmark for Java
26 Aug 2024
This paper introduces SWE-BENCH-JAVA, a new benchmark designed to evaluate large language models on their ability to resolve real-world GitHub issues in Java repositories. The benchmark comprises 91 manually verified issue instances from popular Java projects, demonstrating that current LLMs achieve relatively low success rates on these complex tasks, with DeepSeek models generally outperforming others.
View blog
QuMATL: Query-based Multi-annotator Tendency Learning
15 Jul 2025
Osaka UniversityDifferent annotators often assign different labels to the same sample due to backgrounds or preferences, and such labeling patterns are referred to as tendency. In multi-annotator scenarios, we introduce a novel task called Multi-annotator Tendency Learning (MATL), which aims to capture each annotator tendency. Unlike traditional tasks that prioritize consensus-oriented learning, which averages out annotator differences and leads to tendency information loss, MATL emphasizes learning each annotator tendency, better preserves tendency information. To this end, we propose an efficient baseline method, Query-based Multi-annotator Tendency Learning (QuMATL), which uses lightweight query to represent each annotator for tendency modeling. It saves the costs of building separate conventional models for each annotator, leverages shared learnable queries to capture inter-annotator correlations as an additional hidden supervisory signal to enhance modeling performance. Meanwhile, we provide a new metric, Difference of Inter-annotator Consistency (DIC), to evaluate how effectively models preserve annotators tendency information. Additionally, we contribute two large-scale datasets, STREET and AMER, providing averages of 4300 and 3118 per-annotator labels, respectively. Extensive experiments verified the effectiveness of our QuMATL.
TopoBind: Multi-Modal Prediction of Antibody-Antigen Binding Free Energy via Sequence Embeddings and Structural Topology
27 Aug 2025
The University of Hong Kong
HKUSTPredicting the binding free energy between antibodies and antigens is a key challenge in structure-aware biomolecular modeling, with direct implications for antibody design. Most existing methods either rely solely on sequence embeddings or struggle to capture complex structural relationships, thus limiting predictive performance. In this work, we present a novel framework that integrates sequence-based representations from pre-trained protein language models (ESM-2) with a set of topological features. Specifically, we extract contact map metrics reflecting residue-level connectivity, interface geometry descriptors characterizing cross-chain interactions, distance map statistics quantifying spatial organization, and persistent homology invariants that systematically capture the emergence and persistence of multi-scale topological structures - such as connected components, cycles, and cavities - within individual proteins and across the antibody-antigen interface. By leveraging a cross-attention mechanism to fuse these diverse modalities, our model effectively encodes both global and local structural organization, thereby substantially enhancing the prediction of binding free energy. Extensive experiments demonstrate that our model consistently outperforms sequence-only and conventional structural models, achieving state-of-the-art accuracy in binding free energy prediction.
StableMoFusion: Towards Robust and Efficient Diffusion-based Motion Generation Framework
09 Dec 2024
Thanks to the powerful generative capacity of diffusion models, recent years have witnessed rapid progress in human motion generation. Existing diffusion-based methods employ disparate network architectures and training strategies. The effect of the design of each component is still unclear. In addition, the iterative denoising process consumes considerable computational overhead, which is prohibitive for real-time scenarios such as virtual characters and humanoid robots. For this reason, we first conduct a comprehensive investigation into network architectures, training strategies, and inference processs. Based on the profound analysis, we tailor each component for efficient high-quality human motion generation. Despite the promising performance, the tailored model still suffers from foot skating which is an ubiquitous issue in diffusion-based solutions. To eliminate footskate, we identify foot-ground contact and correct foot motions along the denoising process. By organically combining these well-designed components together, we present StableMoFusion, a robust and efficient framework for human motion generation. Extensive experimental results show that our StableMoFusion performs favorably against current state-of-the-art methods. Project page: this https URL
Global Rice Multi-Class Segmentation Dataset (RiceSEG): A Comprehensive and Diverse High-Resolution RGB-Annotated Images for the Development and Benchmarking of Rice Segmentation Algorithms
02 Apr 2025
the University of Tokyo
Kyoto UniversityDeveloping computer vision-based rice phenotyping techniques is crucial for
precision field management and accelerating breeding, thereby continuously
advancing rice production. Among phenotyping tasks, distinguishing image
components is a key prerequisite for characterizing plant growth and
development at the organ scale, enabling deeper insights into eco-physiological
processes. However, due to the fine structure of rice organs and complex
illumination within the canopy, this task remains highly challenging,
underscoring the need for a high-quality training dataset. Such datasets are
scarce, both due to a lack of large, representative collections of rice field
images and the time-intensive nature of annotation. To address this gap, we
established the first comprehensive multi-class rice semantic segmentation
dataset, RiceSEG. We gathered nearly 50,000 high-resolution, ground-based
images from five major rice-growing countries (China, Japan, India, the
Philippines, and Tanzania), encompassing over 6,000 genotypes across all growth
stages. From these original images, 3,078 representative samples were selected
and annotated with six classes (background, green vegetation, senescent
vegetation, panicle, weeds, and duckweed) to form the RiceSEG dataset. Notably,
the sub-dataset from China spans all major genotypes and rice-growing
environments from the northeast to the south. Both state-of-the-art
convolutional neural networks and transformer-based semantic segmentation
models were used as baselines. While these models perform reasonably well in
segmenting background and green vegetation, they face difficulties during the
reproductive stage, when canopy structures are more complex and multiple
classes are involved. These findings highlight the importance of our dataset
for developing specialized segmentation models for rice and other crops.
Deep Frequency Derivative Learning for Non-stationary Time Series Forecasting
29 Jun 2024
University of OxfordWhile most time series are non-stationary, it is inevitable for models to face the distribution shift issue in time series forecasting. Existing solutions manipulate statistical measures (usually mean and std.) to adjust time series distribution. However, these operations can be theoretically seen as the transformation towards zero frequency component of the spectrum which cannot reveal full distribution information and would further lead to information utilization bottleneck in normalization, thus hindering forecasting performance. To address this problem, we propose to utilize the whole frequency spectrum to transform time series to make full use of data distribution from the frequency perspective. We present a deep frequency derivative learning framework, DERITS, for non-stationary time series forecasting. Specifically, DERITS is built upon a novel reversible transformation, namely Frequency Derivative Transformation (FDT) that makes signals derived in the frequency domain to acquire more stationary frequency representations. Then, we propose the Order-adaptive Fourier Convolution Network to conduct adaptive frequency filtering and learning. Furthermore, we organize DERITS as a parallel-stacked architecture for the multi-order derivation and fusion for forecasting. Finally, we conduct extensive experiments on several datasets which show the consistent superiority in both time series forecasting and shift alleviation.
CDRNP: Cross-Domain Recommendation to Cold-Start Users via Neural Process
12 Jun 2024
Tingwen Liu
Wenyuan Zhang
Cross-domain recommendation (CDR) has been proven as a promising way to tackle the user cold-start problem, which aims to make recommendations for users in the target domain by transferring the user preference derived from the source domain. Traditional CDR studies follow the embedding and mapping (EMCDR) paradigm, which transfers user representations from the source to target domain by learning a user-shared mapping function, neglecting the user-specific preference. Recent CDR studies attempt to learn user-specific mapping functions in meta-learning paradigm, which regards each user's CDR as an individual task, but neglects the preference correlations among users, limiting the beneficial information for user representations. Moreover, both of the paradigms neglect the explicit user-item interactions from both domains during the mapping process. To address the above issues, this paper proposes a novel CDR framework with neural process (NP), termed as CDRNP. Particularly, it develops the meta-learning paradigm to leverage user-specific preference, and further introduces a stochastic process by NP to capture the preference correlations among the overlapping and cold-start users, thus generating more powerful mapping functions by mapping the user-specific preference and common preference correlations to a predictive probability distribution. In addition, we also introduce a preference remainer to enhance the common preference from the overlapping users, and finally devises an adaptive conditional decoder with preference modulation to make prediction for cold-start users with items in the target domain. Experimental results demonstrate that CDRNP outperforms previous SOTA methods in three real-world CDR scenarios.
There are no more papers matching your filters at the moment.