Agents' Last Exam

Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional domains. We argue that this gap is largely an evaluation problem: widely used benchmarks lack sustained performance measurement on real and economically valuable workflows. This paper introduces Agents' Last Exam (ALE), a benchmark designed to evaluate AI agents on long-horizon, economically valuable, real-world tasks with verifiable outcomes. Developed in collaboration with 250+ industry experts, ALE covers non-physical industries defined with reference to O*NET / SOC 2018 (the U.S. federal occupational taxonomy). It is organized around a task taxonomy with 55 subfields grouped into 13 industry clusters covering 1K+ tasks. Current results show that the hardest tier remains far from saturated: across mainstream harness and backbone configurations, the average full pass rate is 2.6%. ALE is designed as a living benchmark: its task pool grows continuously as new workflows and industries are onboarded. More broadly, ALE is intended not merely as another leaderboard, but as an instrument for closing the gap between benchmark success and GDP-relevant impact.

BreastGPT: A Multimodal Large Language Model for the Full Spectrum of Breast Cancer Clinical Routine

Breast cancer remains a leading cause of cancer-related mortality among women. Its clinical management requires multimodal reasoning across a clinical workflow that spans \textit{screening}, \textit{diagnosis} and \textit{treatment planning}, where each stage involves distinct imaging modalities, task objectives, and reasoning patterns. However, constrained by data scarcity and model versatility, existing medical MLLMs are typically evaluated on isolated modalities or narrow task families, limiting their ability to support workflow-level clinical reasoning. In this work, we first introduce \textbf{BreastStage}, a workflow-aligned breast imaging instruction corpus comprising 1.86M instruction-following pairs curated from 17 sub-datasets across 5 imaging modalities and 136 task templates. Its held-out split, \textbf{BreastStage-Bench}, provides a comprehensive benchmark for evaluating multimodal reasoning across the breast cancer care continuum. Building on this corpus, we propose \textbf{BreastGPT}, a unified MLLM equipped with a dual-branch visual encoder and concept-preserving token compression to bridge the scale gap between standard radiology and gigapixel pathology. On BreastStage-Bench, BreastGPT achieves 75.66\% closed-ended accuracy and 89.92\% open-ended score, outperforming both general-purpose and medical-specific MLLMs across clinical stages and task formats. These results suggest that workflow-aligned data and cross-scale visual modeling are critical for clinically grounded medical MLLMs. All data, code, and model checkpoints are released at https://yangyy-liu.github.io/BreastGPT.io.

Self-Supervised Learning for Medical Image Data with Anatomy-Oriented Imaging Planes

Self-supervised learning has emerged as a powerful tool for pretraining deep networks on unlabeled data, prior to transfer learning of target tasks with limited annotation. The relevance between the pretraining pretext and target tasks is crucial to the success of transfer learning. Various pretext tasks have been proposed to utilize properties of medical image data (e.g., three dimensionality), which are more relevant to medical image analysis than generic ones for natural images. However, previous work rarely paid attention to data with anatomy-oriented imaging planes, e.g., standard cardiac magnetic resonance imaging views. As these imaging planes are defined according to the anatomy of the imaged organ, pretext tasks effectively exploiting this information can pretrain the networks to gain knowledge on the organ of interest. In this work, we propose two complementary pretext tasks for this group of medical image data based on the spatial relationship of the imaging planes. The first is to learn the relative orientation between the imaging planes and implemented as regressing their intersecting lines. The second exploits parallel imaging planes to regress their relative slice locations within a stack. Both pretext tasks are conceptually straightforward and easy to implement, and can be combined in multitask learning for better representation learning. Thorough experiments on two anatomical structures (heart and knee) and representative target tasks (semantic segmentation and classification) demonstrate that the proposed pretext tasks are effective in pretraining deep networks for remarkably boosted performance on the target tasks, and superior to other recent approaches.

A Concept-based Interpretable Model for the Diagnosis of Choroid Neoplasias using Multimodal Data

Diagnosing rare diseases presents a common challenge in clinical practice, necessitating the expertise of specialists for accurate identification. The advent of machine learning offers a promising solution, while the development of such technologies is hindered by the scarcity of data on rare conditions and the demand for models that are both interpretable and trustworthy in a clinical context. Interpretable AI, with its capacity for human-readable outputs, can facilitate validation by clinicians and contribute to medical education. In the current work, we focus on choroid neoplasias, the most prevalent form of eye cancer in adults, albeit rare with 5.1 per million. We built the so-far largest dataset consisting of 750 patients, incorporating three distinct imaging modalities collected from 2004 to 2022. Our work introduces a concept-based interpretable model that distinguishes between three types of choroidal tumors, integrating insights from domain experts via radiological reports. Remarkably, this model not only achieves an F1 score of 0.91, rivaling that of black-box models, but also boosts the diagnostic accuracy of junior doctors by 42%. This study highlights the significant potential of interpretable machine learning in improving the diagnosis of rare diseases, laying a groundwork for future breakthroughs in medical AI that could tackle a wider array of complex health scenarios.

Co-Learning Semantic-aware Unsupervised Segmentation for Pathological Image Registration

The registration of pathological images plays an important role in medical applications. Despite its significance, most researchers in this field primarily focus on the registration of normal tissue into normal tissue. The negative impact of focal tissue, such as the loss of spatial correspondence information and the abnormal distortion of tissue, are rarely considered. In this paper, we propose GIRNet, a novel unsupervised approach for pathological image registration by incorporating segmentation and inpainting through the principles of Generation, Inpainting, and Registration (GIR). The registration, segmentation, and inpainting modules are trained simultaneously in a co-learning manner so that the segmentation of the focal area and the registration of inpainted pairs can improve collaboratively. Overall, the registration of pathological images is achieved in a completely unsupervised learning framework. Experimental results on multiple datasets, including Magnetic Resonance Imaging (MRI) of T1 sequences, demonstrate the efficacy of our proposed method. Our results show that our method can accurately achieve the registration of pathological images and identify lesions even in challenging imaging modalities. Our unsupervised approach offers a promising solution for the efficient and cost-effective registration of pathological images. Our code is available at https://github.com/brain-intelligence-lab/GIRNet.

Converting Artificial Neural Networks to Spiking Neural Networks via Parameter Calibration

Spiking Neural Network (SNN), originating from the neural behavior in biology, has been recognized as one of the next-generation neural networks. Conventionally, SNNs can be obtained by converting from pre-trained Artificial Neural Networks (ANNs) by replacing the non-linear activation with spiking neurons without changing the parameters. In this work, we argue that simply copying and pasting the weights of ANN to SNN inevitably results in activation mismatch, especially for ANNs that are trained with batch normalization (BN) layers. To tackle the activation mismatch issue, we first provide a theoretical analysis by decomposing local conversion error to clipping error and flooring error, and then quantitatively measure how this error propagates throughout the layers using the second-order analysis. Motivated by the theoretical results, we propose a set of layer-wise parameter calibration algorithms, which adjusts the parameters to minimize the activation mismatch. Extensive experiments for the proposed algorithms are performed on modern architectures and large-scale tasks including ImageNet classification and MS COCO detection. We demonstrate that our method can handle the SNN conversion with batch normalization layers and effectively preserve the high accuracy even in 32 time steps. For example, our calibration algorithms can increase up to 65% accuracy when converting VGG-16 with BN layers.

When Sparsity Meets Dynamic Convolution

Dynamic convolution achieves a substantial performance boost for efficient CNNs at a cost of increased convolutional weights. Contrastively, mask-based unstructured pruning obtains a lightweight network by removing redundancy in the heavy network at risk of performance drop. In this paper, we propose a new framework to coherently integrate these two paths so that they can complement each other compensate for the disadvantages. We first design a binary mask derived from a learnable threshold to prune static kernels, significantly reducing the parameters and computational cost but achieving higher performance in Imagenet-1K(0.6\% increase in top-1 accuracy with 0.67G fewer FLOPs). Based on this learnable mask, we further propose a novel dynamic sparse network incorporating the dynamic routine mechanism, which exerts much higher accuracy than baselines ($2.63\%$ increase in top-1 accuracy for MobileNetV1 with $90\%$ sparsity). As a result, our method demonstrates a more efficient dynamic convolution with sparsity.

Temporal Efficient Training of Spiking Neural Network via Gradient Re-weighting

Recently, brain-inspired spiking neuron networks (SNNs) have attracted widespread research interest because of their event-driven and energy-efficient characteristics. Still, it is difficult to efficiently train deep SNNs due to the non-differentiability of its activation function, which disables the typically used gradient descent approaches for traditional artificial neural networks (ANNs). Although the adoption of surrogate gradient (SG) formally allows for the back-propagation of losses, the discrete spiking mechanism actually differentiates the loss landscape of SNNs from that of ANNs, failing the surrogate gradient methods to achieve comparable accuracy as for ANNs. In this paper, we first analyze why the current direct training approach with surrogate gradient results in SNNs with poor generalizability. Then we introduce the temporal efficient training (TET) approach to compensate for the loss of momentum in the gradient descent with SG so that the training process can converge into flatter minima with better generalizability. Meanwhile, we demonstrate that TET improves the temporal scalability of SNN and induces a temporal inheritable training for acceleration. Our method consistently outperforms the SOTA on all reported mainstream datasets, including CIFAR-10/100 and ImageNet. Remarkably on DVS-CIFAR10, we obtained 83$\%$ top-1 accuracy, over 10$\%$ improvement compared to existing state of the art. Codes are available at \url{https://github.com/Gus-Lab/temporal_efficient_training}.

Multi-modal Attention Network for Stock Movements Prediction

Stock prices move as piece-wise trending fluctuation rather than a purely random walk. Traditionally, the prediction of future stock movements is based on the historical trading record. Nowadays, with the development of social media, many active participants in the market choose to publicize their strategies, which provides a window to glimpse over the whole market's attitude towards future movements by extracting the semantics behind social media. However, social media contains conflicting information and cannot replace historical records completely. In this work, we propose a multi-modality attention network to reduce conflicts and integrate semantic and numeric features to predict future stock movements comprehensively. Specifically, we first extract semantic information from social media and estimate their credibility based on posters' identity and public reputation. Then we incorporate the semantic from online posts and numeric features from historical records to make the trading strategy. Experimental results show that our approach outperforms previous methods by a significant margin in both prediction accuracy (61.20\%) and trading profits (9.13\%). It demonstrates that our method improves the performance of stock movements prediction and informs future research on multi-modality fusion towards stock prediction.

A Unified Framework for Generalized Low-Shot Medical Image Segmentation with Scarce Data

Medical image segmentation has achieved remarkable advancements using deep neural networks (DNNs). However, DNNs often need big amounts of data and annotations for training, both of which can be difficult and costly to obtain. In this work, we propose a unified framework for generalized low-shot (one- and few-shot) medical image segmentation based on distance metric learning (DML). Unlike most existing methods which only deal with the lack of annotations while assuming abundance of data, our framework works with extreme scarcity of both, which is ideal for rare diseases. Via DML, the framework learns a multimodal mixture representation for each category, and performs dense predictions based on cosine distances between the pixels' deep embeddings and the category representations. The multimodal representations effectively utilize the inter-subject similarities and intraclass variations to overcome overfitting due to extremely limited data. In addition, we propose adaptive mixing coefficients for the multimodal mixture distributions to adaptively emphasize the modes better suited to the current input. The representations are implicitly embedded as weights of the fc layer, such that the cosine distances can be computed efficiently via forward propagation. In our experiments on brain MRI and abdominal CT datasets, the proposed framework achieves superior performances for low-shot segmentation towards standard DNN-based (3D U-Net) and classical registration-based (ANTs) methods, e.g., achieving mean Dice coefficients of 81%/69% for brain tissue/abdominal multiorgan segmentation using a single training sample, as compared to 52%/31% and 72%/35% by the U-Net and ANTs, respectively.