Survival prediction of cancers is crucial for clinical practice, as it informs mortality risks and influences treatment plans. However, a static model trained on a single dataset fails to adapt to the dynamically evolving clinical environment and continuous data streams, limiting its practical utility. While continual learning (CL) offers a solution to learn dynamically from new datasets, existing CL methods primarily focus on unimodal inputs and suffer from severe catastrophic forgetting in survival prediction. In real-world scenarios, multimodal inputs often provide comprehensive and complementary information, such as whole slide images and genomics; and neglecting inter-modal correlations negatively impacts the performance. To address the two challenges of catastrophic forgetting and complex inter-modal interactions between gigapixel whole slide images and genomics, we propose ConSurv, the first multimodal continual learning (MMCL) method for survival analysis. ConSurv incorporates two key components: Multi-staged Mixture of Experts (MS-MoE) and Feature Constrained Replay (FCR). MS-MoE captures both task-shared and task-specific knowledge at different learning stages of the network, including two modality encoders and the modality fusion component, learning inter-modal relationships. FCR further enhances learned knowledge and mitigates forgetting by restricting feature deviation of previous data at different levels, including encoder-level features of two modalities and the fusion-level representations. Additionally, we introduce a new benchmark integrating four datasets, Multimodal Survival Analysis Incremental Learning (MSAIL), for comprehensive evaluation in the CL setting. Extensive experiments demonstrate that ConSurv outperforms competing methods across multiple metrics.

With the rising need for speech-based interaction models, end-to-end Spoken Language Models (SLMs) have emerged as a promising solution.While these models require comprehensive world knowledge for meaningful and reliable human interactions, existing questionanswering (QA) benchmarks fall short in evaluating SLMs' knowledge understanding due to their inability to support end-to-end speech evaluation and account for varied input audio conditions.To address these limitations, we present VoxEval, a novel SpeechQA benchmark that assesses SLMs' knowledge understanding through pure speech interactions.Our benchmark 1) uniquely maintains speech format for both inputs and outputs, 2) evaluates model robustness across diverse input audio conditions, and 3) pioneers the assessment of complex tasks like mathematical reasoning in spoken format.Systematic evaluation demonstrates that VoxEval presents significant challenges to current SLMs, revealing their sensitivity to varying audio conditions and highlighting the need to enhance reasoning capabilities in future development.We hope this benchmark could guide the advancement of more sophisticated and reliable SLMs. 1

Wenqian Cui, Dianzhi Yu, Xiaoqi Jiao, Ziqiao Meng, Guangyan Zhang, Qichao Wang, Steven Y. Guo, Irwin King. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Current research on long-form context in Large Language Models (LLMs) primarily focuses on the understanding of long-contexts, the Openended Long Text Generation (Open-LTG) remains insufficiently explored.Training a longcontext generation model requires curation of gold-standard reference data, which is typically nonexistent for informative Open-LTG tasks.However, previous methods only utilize general assessments as reward signals, which limits accuracy.To bridge this gap, we introduce ProxyReward, an innovative reinforcement learning (RL) based framework, which includes a dataset and a reward signal computation method.Firstly, ProxyReward Dataset generation is accomplished through simple prompts that enables the model to create automatically, obviating extensive labeled data or significant manual effort.Secondly, ProxyReward Signal offers a targeted evaluation of information comprehensiveness and accuracy for specific questions.The experimental results indicate that our method Prox-yReward surpasses even GPT-4-Turbo.It can significantly enhance performance by 20% on the Open-LTG task when training widely used open-source models, while also surpassing the LLM-as-a-Judge approach.Our work presents effective methods to enhance the ability of LLMs to address complex open-ended questions posed by humans.

Inspired by the impressive capabilities of GPT-4o, there is growing interest in enabling speech language models (SLMs) to engage in natural, fluid spoken interactions with humans. Recent advancements have led to the development of several SLMs that demonstrate promising results in this area. However, current approaches have yet to fully exploit dual-channel speech data, which inherently captures the structure and dynamics of human conversation. In this work, we systematically explore the use of dual-channel speech data in the context of modern large language models, and introduce a novel generative modeling paradigm—Next-Token-Pair Prediction (NTPP)—to enable speaker-independent dual-channel spoken dialogue learning using decoder-only architectures for the first time. We evaluate our approach on standard benchmarks, and empirical results show that our proposed method, NTPP, significantly improves the conversational abilities of SLMs in terms of turn-taking prediction, response coherence, and naturalness. Moreover, compared to existing methods, NTPP achieves substantially lower inference latency, highlighting its practical efficiency for real-time applications. Demo and code can be found at https://audio-3059.pages.dev.

Disease name normalization is an important task in the medical domain. It classifies disease names written in various formats into standardized names, serving as a fundamental component in smart healthcare systems for various disease-related functions. Nevertheless, the most significant obstacle to existing disease name normalization systems is the severe shortage of training data. Consequently, we present a novel data augmentation approach that includes a series of data augmentation techniques and some supporting modules to help mitigate the problem. Through extensive experimentation, we illustrate that our proposed approach exhibits significant performance improvements across various baseline models and training objectives, particularly in scenarios with limited training data1.

We introduce BioSentEval, a toolkit for evaluating the quality of medical sentence representations. BioSentEval contains a variety of tasks, medical term normalization, medical text classification a nd m edical s entences s emantic relationship determination. This set of tasks was selected based on CBLUE (Chinese Biomedical Language Understanding Evaluation), a Chinese biomedical information processing leaderboard led by CHIP (China Health Information Processing) Committee. Our toolkit provides scripts for pre-processing the datasets, as well as a simple interface for evaluating sentence representations. Given the popularity of SentEval, our aim is to generalize the idea from general domain to biomedical domain and to motivate further research works of biomedical sentence representations.

On object detection of remote sensing images, speed and accuracy are considered equally significant for time-sensitive tasks. Traditional object detection methods of remote sensing images use anchors to regress the bounding boxes and the position of objects which quite limit the speed of detection. In this paper, we make an attempt at employing an anchor-free method named CenterNet to address the problem of object detection for remote sensing images. This model computes the key point to regress the location, local offset and size without enumerating a nearly exhaustive list of potential object locations and size. Thus CenterNet is simpler, faster, more accurate and end-to-end differentiable than other anchor-based methods. We evaluate the performance of Cen-terNet with various backbones on remote sensing images dataset NWPU VHR-10. Among them, CenterNet based on DLA-34 backbone achieves the highest accuracy of 95.7% on the test set, which significantly outperforms most state-of-the-art methods while maintaining a real-time prediction speed of 31 FPS on NVIDIA GTX 1080. Analysis on experimental results demonstrates that this anchor-free network achieves a better balance between accuracy and speed for this task compared to other methods.