arXiv subjects

Tengfei Li

Publications and source records attributed to Tengfei Li.

18 recordsLinked to original sources

arXiv2026

DiagEval: Trajectory-Conditioned Diagnosis for Reliable Software Evaluation with GUI Agents

Evaluating LLM-generated interactive software requires execution in addition to static analysis. The key difficulty is that correctness is a graph-level reachable property over latent UI state-transition graphs, whereas a GUI evaluator observes only a single execution trajectory. A failed rollout therefore rules out only one realized path, leaving failure attribution ambiguous between evaluator-side execution error and genuine software defect. We present DiagEval, a trajectory-conditioned diagnostic evaluation protocol for post-failure GUI-agent evaluation of interactive software. Rather than blindly retrying from scratch, DiagEval reuses the failed trajectory to choose targeted diagnostic probes and aggregates their outcomes into an internal attribution signal. The latent-graph view motivates the diagnostic problem; DiagEval does not reconstruct the graph or estimate calibrated posterior probabilities. We evaluate DiagEval on WebDevJudge-Unit and RealDevBench across multiple GUI-agent evaluators and LLM backbones. On false-negative cases, DiagEval recovers 45.6-62.1% of failures that were initially misattributed to software defects, outperforming retry-based baselines with 34.4-160.6% relative gains. On the full evaluation sets, this recovery improves accuracy from 69.9% to 78.3% on WebDevJudge-Unit and from 65.0% to 81.6% on RealDevBench. These results suggest that reliable GUI-agent evaluation requires not only stronger execution, but also active failure diagnosis to disambiguate evaluator-side errors from genuine software defects. Our code is available at https://github.com/scutGit/DiagEval.

cs.SE↗

arXiv2026

LATTICE: Evaluating Decision Support Utility of Crypto Agents

We introduce LATTICE, a benchmark for evaluating the decision support utility of crypto agents in realistic user-facing scenarios. Prior crypto agent benchmarks mainly focus on reasoning-based or outcome-based evaluation, but do not assess agents' ability to assist user decision-making. LATTICE addresses this gap by: (1) defining six evaluation dimensions that capture key decision support properties; (2) proposing 16 task types that span the end-to-end crypto copilot workflow; and (3) using LLM judges to automatically score agent outputs based on these dimensions and tasks. Crucially, the dimensions and tasks are designed to be evaluable at scale using LLM judges, without relying on ground truth from expert annotators or external data sources. In lieu of these dependencies, LATTICE's LLM judge rubrics can be continually audited and updated given new dimensions, tasks, criteria, and human feedback, thus promoting reliable and extensible evaluation. While other benchmarks often compare foundation models sharing a generic agent framework, we use LATTICE to assess production-level agents used in actual crypto copilot products, reflecting the importance of orchestration and UI/UX design in determining agent quality. In this paper, we evaluate six real-world crypto copilots on 1,200 diverse queries and report breakdowns across dimensions, tasks, and query categories. Our experiments show that most of the tested copilots achieve comparable aggregate scores, but differ more significantly on dimension-level and task-level performance. This pattern suggests meaningful trade-offs in decision support quality: users with different priorities may be better served by different copilots than the aggregate rankings alone would indicate. To support reproducible research, we open-source all LATTICE code and data used in this paper.

cs.CR↗

arXiv2026

Causal Inference in Biomedical Imaging via Functional Linear Structural Equation Models

Understanding the causal effects of organ-specific features from medical imaging on clinical outcomes is essential for biomedical research and patient care. We propose a novel Functional Linear Structural Equation Model (FLSEM) to capture the relationships among clinical outcomes, functional imaging exposures, and scalar covariates like genetics, sex, and age. Traditional methods struggle with the infinite-dimensional nature of exposures and complex covariates. Our FLSEM overcomes these challenges by establishing identifiable conditions using scalar instrumental variables. We develop the Functional Group Support Detection and Root Finding (FGS-DAR) algorithm for efficient variable selection, supported by rigorous theoretical guarantees, including selection consistency and accurate parameter estimation. We further propose a test statistic to test the nullity of the functional coefficient, establishing its null limit distribution. Our approach is validated through extensive simulations and applied to UK Biobank data, demonstrating robust performance in detecting causal relationships from medical imaging.

stat.ME↗

arXiv2025

Benchmarking Generalizable Bimanual Manipulation: RoboTwin Dual-Arm Collaboration Challenge at CVPR 2025 MEIS Workshop

Embodied Artificial Intelligence (Embodied AI) is an emerging frontier in robotics, driven by the need for autonomous systems that can perceive, reason, and act in complex physical environments. While single-arm systems have shown strong task performance, collaborative dual-arm systems are essential for handling more intricate tasks involving rigid, deformable, and tactile-sensitive objects. To advance this goal, we launched the RoboTwin Dual-Arm Collaboration Challenge at the 2nd MEIS Workshop, CVPR 2025. Built on the RoboTwin Simulation platform (1.0 and 2.0) and the AgileX COBOT-Magic Robot platform, the competition consisted of three stages: Simulation Round 1, Simulation Round 2, and a final Real-World Round. Participants totally tackled 17 dual-arm manipulation tasks, covering rigid, deformable, and tactile-based scenarios. The challenge attracted 64 global teams and over 400 participants, producing top-performing solutions like SEM and AnchorDP3 and generating valuable insights into generalizable bimanual policy learning. This report outlines the competition setup, task design, evaluation methodology, key findings and future direction, aiming to support future research on robust and generalizable bimanual manipulation policies. The Challenge Webpage is available at https://robotwin-benchmark.github.io/cvpr-2025-challenge/.

cs.RO↗

arXiv2025

MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention

We introduce MiniMax-M1, the world's first open-weight, large-scale hybrid-attention reasoning model. MiniMax-M1 is powered by a hybrid Mixture-of-Experts (MoE) architecture combined with a lightning attention mechanism. The model is developed based on our previous MiniMax-Text-01 model, which contains a total of 456 billion parameters with 45.9 billion parameters activated per token. The M1 model natively supports a context length of 1 million tokens, 8x the context size of DeepSeek R1. Furthermore, the lightning attention mechanism in MiniMax-M1 enables efficient scaling of test-time compute. These properties make M1 particularly suitable for complex tasks that require processing long inputs and thinking extensively. MiniMax-M1 is trained using large-scale reinforcement learning (RL) on diverse problems including sandbox-based, real-world software engineering environments. In addition to M1's inherent efficiency advantage for RL training, we propose CISPO, a novel RL algorithm to further enhance RL efficiency. CISPO clips importance sampling weights rather than token updates, outperforming other competitive RL variants. Combining hybrid-attention and CISPO enables MiniMax-M1's full RL training on 512 H800 GPUs to complete in only three weeks, with a rental cost of just $534,700. We release two versions of MiniMax-M1 models with 40K and 80K thinking budgets respectively, where the 40K model represents an intermediate phase of the 80K training. Experiments on standard benchmarks show that our models are comparable or superior to strong open-weight models such as the original DeepSeek-R1 and Qwen3-235B, with particular strengths in complex software engineering, tool utilization, and long-context tasks. We publicly release MiniMax-M1 at https://github.com/MiniMax-AI/MiniMax-M1.

cs.CL↗

arXiv2025

Lane Departure Accident Prevention in Foggy Conditions: A Prior-Guided Dynamic Feature Fusion Transformer Framework for Real-Time Lane Detection

Lane departure accident prevention plays a critical role in enhancing road safety, and lane detection is a core technology to achieve this goal, especially under complex weather conditions. While existing lane detection algorithms perform well under favorable weather conditions, their effectiveness significantly degrades in foggy environments, which increases the risk of traffic accidents. In response to this challenge, we propose PDT-Net, a robust Prior-Guided Dynamic Feature Fusion Transformer framework designed for real-time lane detection in foggy conditions. This framework integrates three key modules: a Global Feature Fusion Module (GFFM) to capture the relationship between local and global features in foggy images, a Dynamic Feature Fusion Module (DFFM) to model the structural and positional relationships of lane instances, and a Prior-Guided Edge Enhancement Module (PEM) to recover lost edge details in foggy environments. Furthermore, we introduce the FoggyLane dataset, a real-world dataset that specifically targets lane detection in foggy conditions, along with two synthesized datasets, FoggyCULane and FoggyTusimple, to address the lack of fog-specific data for lane detection. Extensive experiments show that PDT-Net achieves state-of-the-art performance with F1-scores of 95.04% on FoggyLane, 79.85% on FoggyCULane, and 96.95% on FoggyTusimple. Moreover, with TensorRT acceleration, our method achieves a processing speed of 38.4 FPS on the NVIDIA Jetson AGX Orin, confirming its real-time capability and robustness in challenging foggy environments. By improving the precision of lane detection, our framework can contribute to active safety warning systems, helping to prevent accidents in foggy conditions.

cs.CV↗

arXiv2025

ABCDWaveNet: Advancing Robust Road Ponding Detection in Fog through Dynamic Frequency-Spatial Synergy

Road ponding presents a significant threat to vehicle safety, particularly in adverse fog conditions, where reliable detection remains a persistent challenge for Advanced Driver Assistance Systems (ADAS). To address this, we propose ABCDWaveNet, a novel deep learning framework leveraging Dynamic Frequency-Spatial Synergy for robust ponding detection in fog. The core of ABCDWaveNet achieves this synergy by integrating dynamic convolution for adaptive feature extraction across varying visibilities with a wavelet-based module for synergistic frequency-spatial feature enhancement, significantly improving robustness against fog interference. Building on this foundation, ABCDWaveNet captures multi-scale structural and contextual information, subsequently employing an Adaptive Attention Coupling Gate (AACG) to adaptively fuse global and local features for enhanced accuracy. To facilitate realistic evaluations under combined adverse conditions, we introduce the Foggy Low-Light Puddle dataset. Extensive experiments demonstrate that ABCDWaveNet establishes new state-of-the-art performance, achieving significant Intersection over Union (IoU) gains of 3.51%, 1.75%, and 1.03% on the Foggy-Puddle, Puddle-1000, and our Foggy Low-Light Puddle datasets, respectively. Furthermore, its processing speed of 25.48 FPS on an NVIDIA Jetson AGX Orin confirms its suitability for ADAS deployment. These findings underscore the effectiveness of the proposed Dynamic Frequency-Spatial Synergy within ABCDWaveNet, offering valuable insights for developing proactive road safety solutions capable of operating reliably in challenging weather conditions.

cs.CV↗

arXiv2024

EMDFNet: Efficient Multi-scale and Diverse Feature Network for Traffic Sign Detection

The detection of small objects, particularly traffic signs, is a critical subtask within object detection and autonomous driving. Despite the notable achievements in previous research, two primary challenges persist. Firstly, the main issue is the singleness of feature extraction. Secondly, the detection process fails to effectively integrate with objects of varying sizes or scales. These issues are also prevalent in generic object detection. Motivated by these challenges, in this paper, we propose a novel object detection network named Efficient Multi-scale and Diverse Feature Network (EMDFNet) for traffic sign detection that integrates an Augmented Shortcut Module and an Efficient Hybrid Encoder to address the aforementioned issues simultaneously. Specifically, the Augmented Shortcut Module utilizes multiple branches to integrate various spatial semantic information and channel semantic information, thereby enhancing feature diversity. The Efficient Hybrid Encoder utilizes global feature fusion and local feature interaction based on various features to generate distinctive classification features by integrating feature information in an adaptable manner. Extensive experiments on the Tsinghua-Tencent 100K (TT100K) benchmark and the German Traffic Sign Detection Benchmark (GTSDB) demonstrate that our EMDFNet outperforms other state-of-the-art detectors in performance while retaining the real-time processing capabilities of single-stage models. This substantiates the effectiveness of EMDFNet in detecting small traffic signs.

cs.CV↗

arXiv2024

A Multivariate Equivalence Test Based on Mahalanobis Distance with a Data-Driven Margin

Multivariate equivalence testing is needed in a variety of scenarios for drug development. For example, drug products obtained from natural sources may contain many components for which the individual effects and/or their interactions on clinical efficacy and safety cannot be completely characterized. Such lack of sufficient characterization poses a challenge for both generic drug developers to demonstrate and regulatory authorities to determine the sameness of a proposed generic product to its reference product. Another case is to ensure batch-to-batch consistency of naturally derived products containing a vast number of components, such as botanical products. The equivalence or sameness between products containing many components that cannot be individually evaluated needs to be studied in a holistic manner. Multivariate equivalence test based on Mahalanobis distance may be suitable to evaluate many variables holistically. Existing studies based on such method assumed either a predetermined constant margin, for which a consensus is difficult to achieve, or a margin derived from the data, where, however, the randomness is ignored during the testing. In this study, we propose a multivariate equivalence test based on Mahalanobis distance with a data-drive margin with the randomness in the margin considered. Several possible implementations are compared with existing approaches via extensive simulation studies.

stat.ME↗

arXiv2024

A Classifier-Free Incremental Learning Framework for Scalable Medical Image Segmentation

Current methods for developing foundation models in medical image segmentation rely on two primary assumptions: a fixed set of classes and the immediate availability of a substantial and diverse training dataset. However, this can be impractical due to the evolving nature of imaging technology and patient demographics, as well as labor-intensive data curation, limiting their practical applicability and scalability. To address these challenges, we introduce a novel segmentation paradigm enabling the segmentation of a variable number of classes within a single classifier-free network, featuring an architecture independent of class number. This network is trained using contrastive learning and produces discriminative feature representations that facilitate straightforward interpretation. Additionally, we integrate this strategy into a knowledge distillation-based incremental learning framework, facilitating the gradual assimilation of new information from non-stationary data streams while avoiding catastrophic forgetting. Our approach provides a unified solution for tackling both class- and domain-incremental learning scenarios. We demonstrate the flexibility of our method in handling varying class numbers within a unified network and its capacity for incremental learning. Experimental results on an incompletely annotated, multi-modal, multi-source dataset for medical image segmentation underscore its superiority over state-of-the-art alternative approaches.

cs.CV↗

arXiv2022

Statistical learning methods for neuroimaging data analysis with applications

The aim of this paper is to provide a comprehensive review of statistical challenges in neuroimaging data analysis from neuroimaging techniques to large-scale neuroimaging studies to statistical learning methods. We briefly review eight popular neuroimaging techniques and their potential applications in neuroscience research and clinical translation. We delineate the four common themes of neuroimaging data and review major image processing analysis methods for processing neuroimaging data at the individual level. We briefly review four large-scale neuroimaging-related studies and a consortium on imaging genomics and discuss four common themes of neuroimaging data analysis at the population level. We review nine major population-based statistical analysis methods and their associated statistical challenges and present recent progress in statistical methodology to address these challenges.

stat.AP↗

arXiv2022

Evaluating Deep Vs. Wide & Deep Learners As Contextual Bandits For Personalized Email Promo Recommendations

Personalization enables businesses to learn customer preferences from past interactions and thus to target individual customers with more relevant content. We consider the problem of predicting the optimal promotional offer for a given customer out of several options as a contextual bandit problem. Identifying information for the customer and/or the campaign can be used to deduce unknown customer/campaign features that improve optimal offer prediction. Using a generated synthetic email promo dataset, we demonstrate similar prediction accuracies for (a) a wide and deep network that takes identifying information (or other categorical features) as input to the wide part and (b) a deep-only neural network that includes embeddings of categorical features in the input. Improvements in accuracy from including categorical features depends on the variability of the unknown numerical features for each category. We also show that selecting options using upper confidence bound or Thompson sampling, approximated via Monte Carlo dropout layers in the wide and deep models, slightly improves model performance.

cs.LG↗

arXiv2021

The Alzheimer's Disease Prediction Of Longitudinal Evolution (TADPOLE) Challenge: Results after 1 Year Follow-up

We present the findings of "The Alzheimer's Disease Prediction Of Longitudinal Evolution" (TADPOLE) Challenge, which compared the performance of 92 algorithms from 33 international teams at predicting the future trajectory of 219 individuals at risk of Alzheimer's disease. Challenge participants were required to make a prediction, for each month of a 5-year future time period, of three key outcomes: clinical diagnosis, Alzheimer's Disease Assessment Scale Cognitive Subdomain (ADAS-Cog13), and total volume of the ventricles. The methods used by challenge participants included multivariate linear regression, machine learning methods such as support vector machines and deep neural networks, as well as disease progression models. No single submission was best at predicting all three outcomes. For clinical diagnosis and ventricle volume prediction, the best algorithms strongly outperform simple baselines in predictive ability. However, for ADAS-Cog13 no single submitted prediction method was significantly better than random guesswork. Two ensemble methods based on taking the mean and median over all predictions, obtained top scores on almost all tasks. Better than average performance at diagnosis prediction was generally associated with the additional inclusion of features from cerebrospinal fluid (CSF) samples and diffusion tensor imaging (DTI). On the other hand, better performance at ventricle volume prediction was associated with inclusion of summary statistics, such as the slope or maxima/minima of biomarkers. TADPOLE's unique results suggest that current prediction algorithms provide sufficient accuracy to exploit biomarkers related to clinical diagnosis and ventricle volume, for cohort refinement in clinical trials for Alzheimer's disease. However, results call into question the usage of cognitive test scores for patient selection and as a primary endpoint in clinical trials.

q-bio.PE↗

arXiv2020

A New Many-Objective Evolutionary Algorithm Based on Determinantal Point Processes

To handle different types of Many-Objective Optimization Problems (MaOPs), Many-Objective Evolutionary Algorithms (MaOEAs) need to simultaneously maintain convergence and population diversity in the high-dimensional objective space. In order to balance the relationship between diversity and convergence, we introduce a Kernel Matrix and probability model called Determinantal Point Processes (DPPs). Our Many-Objective Evolutionary Algorithm with Determinantal Point Processes (MaOEADPPs) is presented and compared with several state-of-the-art algorithms on various types of MaOPs \textcolor{blue}{with different numbers of objectives}. The experimental results demonstrate that MaOEADPPs is competitive.

cs.NE↗

arXiv2019

Electrolytic Conversion of Bicarbonate into CO in a Flow Cell

Electrochemical CO2 reduction offers a method to use renewable electricity to convert CO2 into CO and other carbon-based chemical building blocks. While nearly all studies rely on a CO2 feed, we show herein that aqueous bicarbonate solutions can also be electrochemically converted into CO gas at meaningful rates in a flow cell. We achieved this result in a flow cell containing a bipolar membrane (BPM) and a silver nanoparticle catalyst on a porous carbon support. Electrolysis upon a N2-saturated 3.0-M potassium bicarbonate electrolyte solution yields CO with a faradaic efficiency (F.E.CO) of 81% at 25 mA cm-2 and 37% at 100 mA cm-2. This output is comparable to the analogous experiment where the electrolyte is saturated with gaseous CO2 (faradaic efficiency for CO is 78% at 25 mA cm-2 and 35% at 100 mA cm-2). The H+ flux from the BPM is critical to this chemistry in that it reacts with the bicarbonate feed to generate CO2, which is then reduced to CO at the gas diffusion electrode. These results are important in that they show that the addition of gaseous CO2 to bicarbonate electrolytes is not necessary in order to obtain reduced carbon products with a flow cell architecture. This process offers a means of using electrolysis to bypass the thermally-intensive step of extracting CO2 from bicarbonate solutions generated in carbon capture schemes.

physics.chem-ph↗

arXiv2017

Filtering Signal Processes in Molecular Multimedia Memristors

To obtain precisely controllable, robust as well as reproduceable memristor for efficient neuromorphic computing still very challenging. Molecular tailoring aims at obtaining the much more flexibly tuning plasticity has recently generated significant interest as new paradigms toward the realization of novel memristor-based synapses. Herein, inspired by the deliberate oxygen transport carried by the hemoglobin in our blood circulation, we report a novel molecular-regulated electronic/ionic semiconducting (MEIS) platform ITO/MTPP/Al2O3-x/Al with a series of metallophorphyrins (MTPPs) to delicately regulate the ionic migration for robust molecular multimedia memristors. The stable pinched hysteresis resulted from the coordination-regulated ionic migration was verified by different device structures, operation modes, as well as the characterizations of scanning transmission electron microscopy with energy-dispersive X-ray spectroscopy (STEM-EDX) and X-ray photoelectron spectroscopy (XPS). Metal coordination-dependent device parameters such as potential and depression as well as retention curves further support the correlation between the coordination and stimulating flux-dependent memristive behaviors. In the 5,10,15,20-tetraphenyl-21H,23H-porphyrin zinc(II) (ZnTPP) synapse, we implement versatile emulations, mainly including transition from short-term memory (STM) to long-term-memory (LTM), learning experience and activity-dependent learning-memory process in integrated neuromorphic configurations based on the biological Hebbian rules, and develop the fresh Spike-Amplitude-Dependent-Plasticity (SADP) with the applications of signal filtering and habituation and sensitization which are beyond the prevalent Hebbian rules.

cond-mat.mes-hall↗

arXiv2016

Hyperbolic Metamaterials and Coupled Surface Plasmon Polaritons: comparative analysis

We investigate the optical properties of sub-wavelength layered metal/dielectric structures, also known as hyperbolic metamaterials (HMMs), using exact analytical Kronig Penney (KP) model. We show that hyperbolic isofrequency surfaces exist for all combinations of layer permittivities and thicknesses, and the largest Purcell enhancements (PE) of spontaneous radiation are achieved away from the nominally hyperbolic region. Detailed comparison of field distributions, dispersion curves, and Purcell factors (PF) between the HMMs and Surface Plasmon Polaritons (SPPs) guided modes in metal/dielectric waveguides demonstrates that HMMs are nothing but weakly coupled gap or slab SPPs modes. Broadband PE is not specific to the HMMs and can be easily attained in single thin metallic layers. Furthermore, large wavevectors and PE are always combined with high loss, short propagation distances and large impedances; hence PE in HMMs is essentially a direct coupling of the energy into the free electron motion in the metal, or quenching of radiative lifetime. PE in HMMs is not related to the hyperbolicity per se but is simply the consequence of the strong dispersion of permittivity in the metals or polar dielectrics, as our conclusions are relevant also for the infrared HMMs occurring in nature. When it comes to enhancement of radiating processes and field concentrations, HMMs are not superior to far simpler plasmonic structures.

physics.optics↗

arXiv2013

Likelihood Adaptively Modified Penalties

A new family of penalty functions, adaptive to likelihood, is introduced for model selection in general regression models. It arises naturally through assuming certain types of prior distribution on the regression parameters. To study stability properties of the penalized maximum likelihood estimator, two types of asymptotic stability are defined. Theoretical properties, including the parameter estimation consistency, model selection consistency, and asymptotic stability, are established under suitable regularity conditions. An efficient coordinate-descent algorithm is proposed. Simulation results and real data analysis show that the proposed method has competitive performance in comparison with existing ones.

stat.ME↗