Transformers have achieved remarkable success across a wide range of applications, a feat often attributed to their scalability. Yet training them without skip (residual) connections remains notoriously difficult. While skips stabilize optimization, they also disrupt the hierarchical structure of representations, raising the long-standing question of whether transformers can be trained efficiently without them. In this work, we address this problem by analyzing the Jacobian of a skipless transformer block, showing why skips improve conditioning and revealing that their stabilization benefits can be recovered through a principled initialization strategy. Building on this insight, we introduce the first method that enables stable and efficient training of skipless transformers without altering the standard architecture. We validate our approach on Vision Transformers (ViTs) in both supervised and self-supervised settings, demonstrating that skipless ViTs trained with our initialization overcome the usual optimization barriers, learn richer hierarchical representations, and outperform strong baselines, that incorporate skip connections, on dense prediction benchmarks. These results show that skip connections are not a fundamental requirement for training ViTs and open new avenues for hierarchical representation learning in vision models.

The emergence of large pre-trained networks has revolutionized the AI field, unlocking new possibilities and achieving unprecedented performance. However, these models inherit a fundamental limitation from traditional Machine Learning approaches: their strong dependence on the \textit{i.i.d.} assumption hinders their adaptability to dynamic learning scenarios. We believe the next breakthrough in AI lies in enabling efficient adaptation to evolving environments -- such as the real world -- where new data and tasks arrive sequentially. This challenge defines the field of Continual Learning (CL), a Machine Learning paradigm focused on developing lifelong learning neural models. One alternative to efficiently adapt these large-scale models is known Parameter-Efficient Fine-Tuning (PEFT). These methods tackle the issue of adapting the model to a particular data or scenario by performing small and efficient modifications, achieving similar performance to full fine-tuning. However, these techniques still lack the ability to adjust the model to multiple tasks continually, as they suffer from the issue of Catastrophic Forgetting. In this survey, we first provide an overview of CL algorithms and PEFT methods before reviewing the state-of-the-art on Parameter-Efficient Continual Fine-Tuning (PECFT). We examine various approaches, discuss evaluation metrics, and explore potential future research directions. Our goal is to highlight the synergy between CL and Parameter-Efficient Fine-Tuning, guide researchers in this field, and pave the way for novel future research directions.

In the context of large language models (LLMs), current advanced reasoning methods have made impressive strides in various reasoning tasks. However, when it comes to logical reasoning tasks, major challenges remain in both efficacy and efficiency. This is rooted in the fact that these systems fail to fully leverage the inherent structure of logical tasks throughout the reasoning processes such as decomposition, search, and resolution. To address this, we propose a logic-complete reasoning framework, Aristotle, with three key components: Logical Decomposer, Logical Search Router, and Logical Resolver. In our framework, symbolic expressions and logical rules are comprehensively integrated into the entire reasoning process, significantly alleviating the bottlenecks of logical reasoning, i.e., reducing sub-task complexity, minimizing search errors, and resolving logical contradictions. The experimental results on several datasets demonstrate that Aristotle consistently outperforms state-of-the-art reasoning frameworks in both accuracy and efficiency, particularly excelling in complex logical reasoning scenarios. We will open-source all our code at this https URL.

Micro-expression recognition (MER) is a highly challenging task in affective computing. With the reduced-sized micro-expression (ME) input that contains key information based on key-frame indexes, key-frame-based methods have significantly improved the performance of MER. However, most of these methods focus on improving the performance with relatively accurate key-frame indexes, while ignoring the difficulty of obtaining accurate key-frame indexes and the objective existence of key-frame index errors, which impedes them from moving towards practical applications. In this paper, we propose CausalNet, a novel framework to achieve robust MER facing key-frame index errors while maintaining accurate recognition. To enhance robustness, CausalNet takes the representation of the entire ME sequence as the input. To address the information redundancy brought by the complete ME range input and maintain accurate recognition, first, the Causal Motion Position Learning Module (CMPLM) is proposed to help the model locate the muscle movement areas related to Action Units (AUs), thereby reducing the attention to other redundant areas. Second, the Causal Attention Block (CAB) is proposed to deeply learn the causal relationships between the muscle contraction and relaxation movements in MEs. Empirical experiments have demonstrated that on popular ME benchmarks, the CausalNet has achieved robust MER under different levels of key-frame index noise. Meanwhile, it has surpassed state-of-the-art (SOTA) methods on several standard MER benchmarks when using the provided annotated key-frames. Code is available at this https URL.

The Nancy Grace Roman Space Telescope (Roman) will conduct a Galactic Exoplanet Survey (RGES) to discover bound and free-floating exoplanets using gravitational microlensing. Roman should be sensitive to lenses with mass down to ~ 0.02 $M_{\oplus}$, or roughly the mass of Ganymede. Thus the detection of moons with masses similar to the giant moons in our Solar System is possible with Roman. Measuring the demographics of exomoons will provide constraints on both moon and planet formation. We conduct simulations of Roman microlensing events to determine the effects of exomoons on microlensing light curves, and whether these effects are detectable with Roman. We focus on giant planets from 30 $M_{\oplus}$ to 10 $M_{Jup}$ on orbits from 0.3 to 30 AU, and assume that each planet is orbited by a moon with moon-planet mass ratio from $10^{-4}$ to $10^{-2}$ and separations from 0.1 to 0.5 planet Hill radii. We find that Roman is sensitive to exomoons, although the number of expected detections is only of order one over the duration of the survey, unless exomoons are more common or massive than we assumed. We argue that changes in the survey strategy, in particular focusing on a few fields with higher cadence, may allow for the detection of more exomoons with Roman. Regardless, the ability to detect exomoons reinforces the need to develop robust methods for modeling triple lens microlensing events to fully utilize the capabilities of Roman.

Every year, most educational institutions seek and receive an enormous volume of text feedback from students on courses, teaching, and overall experience. Yet, turning this raw feedback into useful insights is far from straightforward. It has been a long-standing challenge to adopt automatic opinion mining solutions for such education review text data due to the content complexity and low-granularity reporting requirements. Aspect-based Sentiment Analysis (ABSA) offers a promising solution with its rich, sub-sentence-level opinion mining capabilities. However, existing ABSA research and resources are very heavily focused on the commercial domain. In education, they are scarce and hard to develop due to limited public datasets and strict data protection. A high-quality, annotated dataset is urgently needed to advance research in this under-resourced area. In this work, we present EduRABSA (Education Review ABSA), the first public, annotated ABSA education review dataset that covers three review subject types (course, teaching staff, university) in the English language and all main ABSA tasks, including the under-explored implicit aspect and implicit opinion extraction. We also share ASQE-DPT (Data Processing Tool), an offline, lightweight, installation-free manual data annotation tool that generates labelled datasets for comprehensive ABSA tasks from a single-task annotation. Together, these resources contribute to the ABSA community and education domain by removing the dataset barrier, supporting research transparency and reproducibility, and enabling the creation and sharing of further resources. The dataset, annotation tool, and scripts and statistics for dataset processing and sampling are available at this https URL.

Large Language Models have shown tremendous performance on a large variety of natural language processing tasks, ranging from text comprehension to common sense reasoning. However, the mechanisms responsible for this success remain opaque, and it is unclear whether LLMs can achieve human-like cognitive capabilities or whether these models are still fundamentally circumscribed. Abstract reasoning is a fundamental task for cognition, consisting of finding and applying a general pattern from few data. Evaluating deep neural architectures on this task could give insight into their potential limitations regarding reasoning and their broad generalisation abilities, yet this is currently an under-explored area. In this paper, we introduce a new benchmark for evaluating language models beyond memorization on abstract reasoning tasks. We perform extensive evaluations of state-of-the-art LLMs, showing that they currently achieve very limited performance in contrast with other natural language tasks, even when applying techniques that have been shown to improve performance on other NLP tasks. We argue that guiding LLM generation to follow causal paths could help improve the generalisation and reasoning abilities of LLMs.

We study data-driven decision problems where historical observations are generated by a time-evolving distribution whose consecutive shifts are bounded in Wasserstein distance. We address this nonstationarity using a distributionally robust optimization model with an ambiguity set that is a Wasserstein ball centered at a weighted empirical distribution, thereby allowing for the time decay of past data in a way which accounts for the drift of the data-generating distribution. Our main technical contribution is a concentration inequality for weighted empirical distributions that explicitly captures both the effective sample size (i.e., the equivalent number of equally weighted observations) and the distributional drift. Using our concentration inequality, we select observation weights that optimally balance the effective sample size against the extent of drift. The family of optimal weightings reveals an interplay between the order of the Wasserstein ambiguity ball and the time-decay profile of the optimal weights. Classical weighting schemes, such as time windowing and exponential smoothing, emerge as special cases of our framework, for which we derive principled choices of the parameters. Numerical experiments demonstrate the effectiveness of the proposed approach.

The proliferation of various data sources in urban and territorial environments has significantly facilitated the development of geospatial artificial intelligence (GeoAI) across a wide range of geospatial applications. However, geospatial data, which is inherently linked to geospatial objects, often exhibits data heterogeneity that necessitates specialized fusion and representation strategies while simultaneously being inherently sparse in labels for downstream tasks. Consequently, there is a growing demand for techniques that can effectively leverage geospatial data without heavy reliance on task-specific labels and model designs. This need aligns with the principles of self-supervised learning (SSL), which has garnered increasing attention for its ability to learn effective and generalizable representations directly from data without extensive labeled supervision. This paper presents a comprehensive and up-to-date survey of SSL techniques specifically applied to or developed for geospatial objects in three primary vector geometric types: Point, Polyline, and Polygon. We systematically categorize various SSL techniques into predictive and contrastive methods, and analyze their adaptation to different data types for representation learning across various downstream tasks. Furthermore, we examine the emerging trends in SSL for geospatial objects, particularly the gradual advancements towards geospatial foundation models. Finally, we discuss key challenges in current research and outline promising directions for future investigation. By offering a structured analysis of existing studies, this paper aims to inspire continued progress in integrating SSL with geospatial objects, and the development of geospatial foundation models in a longer term.