中国科学院大学

High-dimensional neuroimaging analyses for clinical diagnosis are often constrained by compromises in spatiotemporal fidelity and the limited adaptability of large-scale, general-purpose models. To address these challenges, we introduce Dynamic Curriculum Learning for Spatiotemporal Encoding (DCL-SE), an end-to-end framework centered on data-driven spatiotemporal encoding (DaSE). We leverage Approximate Rank Pooling (ARP) to efficiently encode three-dimensional volumetric brain data into information-rich, two-dimensional dynamic representations, and then decode this representation through a DGM-implemented architectural curriculum, where the fixed hierarchical decoder progressively refines features from global anatomical structures to fine pathological details without introducing a threshold-controlled stage-switch scheduler. Evaluated across six publicly available datasets, including Alzheimer’s disease and brain tumor classification, cerebral artery segmentation, and brain age prediction, DCL-SE consistently outperforms existing methods in accuracy, robustness, and interpretability. These findings underscore the critical importance of compact, task-specific architectures in the era of large-scale pretrained networks.

Point-supervised Temporal Action Localization (PTAL) adopts a lightly frame-annotated paradigm (i.e., labeling only a single frame per action instance) to train a model to effectively locate action instances within untrimmed videos. Most existing approaches design the task head of models with only a point-supervised snippet-level classification, without explicit modeling of understanding temporal relationships among frames of an action. However, understanding the temporal relationships of frames is crucial because it can help a model understand how an action is defined and therefore benefits the localization of complete action instances. To this end, in this paper, we design a multi-task learning framework that fully utilizes point supervision to boost the model’s temporal understanding capability for action localization. Specifically, we design three self-supervised temporal understanding tasks: (i) Action Completion, (ii) Action Order Understanding, and (iii) Action Regularity Understanding. These tasks help a model understand the temporal consistency of actions across videos. To the best of our knowledge, this is the first attempt to explicitly explore temporal consistency for point-supervision action localization. Extensive experimental results on four benchmark datasets demonstrate the effectiveness of the proposed method compared to several state-of-the-art approaches.

Reliability in natural language generation (NLG) is often compromised by semantically plausible but unsupported continuations. We present Two-Stage Distributionally Robust Decoding (TS-DRD), an inference-time framework that combines Wasserstein-robust logit correction with uncertainty-conditioned affine steering. The robust correction is analyzed at the level of the surrogate objective used inside the transport layer. A first-order variational characterization explains why the correction smooths semantically vulnerable token configurations, although it provides no theorem-level guarantee on factual correctness or hallucination rate. Factuality-oriented gains are therefore reported as empirical findings on public benchmarks. We evaluate TS-DRD on TruthfulQA (N=817), HaluEval (N=10,000), CNN/DailyMail (N=11,490), and PopQA. Empirically, TS-DRD improves TruthfulQA MC2 accuracy from 53.8% to 58.2% over static robust baselines and reduces the HaluEval summarization hallucination rate from 28.8% to 24.2%. Disaggregated diagnostics further characterize the source of these gains: a baseline-correct vs. baseline-hallucinating bucket split shows that gains come predominantly from correcting hallucinated cases rather than from uniformly shortening responses; a held-out fidelity check between our single-pass uncertainty predictor and the offline Monte Carlo (MC) Dropout statistics it approximates yields Pearson r=0.86; and per-method generation-length and entity-density measurements rule out trivial conservative shortening. These results support TS-DRD as a practical approach for reliability-oriented decoding in information processing tasks.

Symbolic computation, powered by modern computer algebra systems, has important applications in mathematical reasoning through exact deep computations. The efficiency of symbolic computation is largely constrained by such deep computations in high dimension. This creates a fundamental barrier on labelled data acquisition if leveraging supervised deep learning to accelerate symbolic computation. Cylindrical algebraic decomposition (CAD) is a pillar symbolic computation method for reasoning with first-order logic formulas over reals with many applications in formal verification and automatic theorem proving. Variable orderings have a huge impact on its efficiency. Impeded by the difficulty to acquire abundant labelled data, existing learning-based approaches are only competitive with the best expert-based heuristics. In this work, we address this problem by designing a series of intimately connected tasks for which a large amount of annotated data can be easily obtained. We pre-train a Transformer model with these data and then fine-tune it on the datasets for CAD ordering. Experiments on publicly available CAD ordering datasets show that on average the orderings predicted by the new model are significantly better than those suggested by the best heuristic methods.

This study investigates the behavioral proxies for investor group thinking modes and examines their contemporaneous associations with asset pricing, as reflected in stock returns under different market and information conditions. Using a dataset of 10,451,366 daily user posts and 24,284,543 comments from Eastmoney Guba, China's leading financial social media platform, we construct behavioral proxies for fast and slow thinking modes. Our methodological approach constructs the fast thinking proxy through sentiment analysis via supervised fine-tuned large language models, while the slow thinking proxy builds on a multi-attribute decision method rooted in prospect theory. By employing regression models and robustness tests, our key findings show that both thinking modes are positively associated with stock returns. Our results show a synergistic relationship between the two modes, with fast thinking playing a dominant role. Interestingly, fast thinking is related to higher returns by increasing idiosyncratic risk, whereas slow thinking is associated with higher returns by promoting heterogeneous information exchange. Further, we explore the roles of specific conditions, including bull and bear markets and environments with conflicting sentiment between posts and news, in the relationship between thinking modes and stock returns. These findings provide new insights into investor information processing and group decision-making and highlight the applicability of dual-system theory in the context of financial social media.