A BSTRACT Direct Preference Optimization (DPO) has gained attention as an efficient alternative to reinforcement learning from human feedback (RLHF) for aligning large language models (LLMs) with human preferences. Despite its advantages, DPO suffers from a length bias, generating responses longer than those from the reference model. Existing solutions like SimPO and SamPO address this issue but uniformly treat the contribution of rewards across sequences, overlooking temporal dynamics. To this end, we propose an enhanced preference optimization method that incorporates a temporal decay factor controlled by a gamma parameter. This dynamic weighting mechanism adjusts the influence of each reward based on its position in the sequence, prioritizing earlier tokens that are more critical for alignment. By adaptively focusing on more relevant feedback, our approach mitigates overfitting to less pertinent data and remains responsive to evolving human preferences. Experimental results on several benchmarks show that our approach consistently outperforms vanilla DPO by 5.9-8.8 points on AlpacaEval 2 and 3.3-9.7 points on Arena-Hard across different model architectures and sizes. Furthermore, additional experiments on mathematical and reasoning benchmarks (MMLU, GSM8K, and MA TH) confirm that our method enhances performance without compromising general capabilities. Our codebase would be available at https://github.com/LotuSrc/D2PO . 1 I NTRODUCTION Direct Preference Optimization (DPO) (Rafailov et al., 2023) has recently emerged as a highly efficient alternative for aligning large language models (LLMs) with human preferences (Askell et al., 2021; Ouyang et al., 2022). Unlike reinforcement learning from human feedback (RLHF), which involves training a reward model followed by iterative policy updates, DPO reframes the problem as a binary classification task directly over human preference data. Compared to supervised fine-tuning, DPO enables the model not only to learn what is good but also to be aware of what is bad. This formulation allows DPO to optimize preference alignment in a single-stage training process, bypassing the complexities of reinforcement learning, such as policy sampling or extensive hyperparameter tuning.

Despite the success of Transformer-based models in the time-series prediction (TSP) tasks, the existing Transformer architecture still face limitations and the literature lacks comprehensive explorations into alternative architectures. To address these challenges, we propose AutoFormer-TS, a novel framework that leverages a comprehensive search space for Transformer architectures tailored to TSP tasks. Our framework introduces a differentiable neural architecture search (DNAS) method, AB-DARTS, which improves upon existing DNAS approaches by enhancing the identification of optimal operations within the architecture. AutoFormer-TS systematically explores alternative attention mechanisms, activation functions, and encoding operations, moving beyond the traditional Transformer design. Extensive experiments demonstrate that AutoFormer-TS consistently outperforms state-of-the-art baselines across various TSP benchmarks, achieving superior forecasting accuracy while maintaining reasonable training efficiency.

Marine heatwaves (MHWs) are extreme ocean-temperature events with significant impacts on marine ecosystems and related industries. Accurate forecasts (one to six months ahead) of MHWs would aid in mitigating these impacts. However, forecasting MHWs presents a challenging imbalanced regression task due to the rarity of extreme temperature anomalies in comparison to more frequent moderate conditions. In this study, we examine monthly MHW forecasts for 12 locations around New Zealand. We use a fully-connected neural network and compare standard and specialized regression loss functions, including the mean squared error (MSE), the mean absolute error (MAE), the Huber, the weighted MSE, the focal-R, the balanced MSE, and a proposed scaling-weighted MSE. Results show that (i) short lead times (one month) are considerably more predictable than three- and six-month leads, (ii) models trained with the standard MSE or MAE losses excel at forecasting average conditions but struggle to capture extremes, and (iii) specialized loss functions such as the balanced MSE and our scaling-weighted MSE substantially improve forecasting of MHW and suspected MHW events. These findings underscore the importance of tailored loss functions for imbalanced regression, particularly in forecasting rare but impactful events such as MHWs.

These magnificent rays are at risk of disappearing due to targeted fishing, being caught as bycatch, and climate change. Scientists at the research collaboration Mobula Conservation are teaming up with artisanal and industrial fishermen to protect them. Also known as "Devil Rays," mobulas are elasmobranchs: a subclass of fish--including sharks, skates, and sawfish--that are distinguished by having skeletons primarily made from cartilage. More than a third of the species in this group are threatened with extinction. Of the nine species of mobulas, seven are endangered and two are vulnerable according to the International Union for Conservation of Nature.

We explore how private synthetic text can be generated by suitably prompting a large language model (LLM). This addresses a challenge for organizations like hospitals, which hold sensitive text data like patient medical records, and wish to share it in order to train machine learning models for medical tasks, while preserving patient privacy. Methods that rely on training or finetuning a model may be out of reach, either due to API limits of third-party LLMs, or due to ethical and legal prohibitions on sharing the private data with the LLM itself. We propose Differentially Private Keyphrase Prompt Seeding (DP-KPS), a method that generates a private synthetic text corpus from a sensitive input corpus, by accessing an LLM only through privatized prompts. It is based on seeding the prompts with private samples from a distribution over phrase embeddings, thus capturing the input corpus while achieving requisite output diversity and maintaining differential privacy. We evaluate DP-KPS on downstream ML text classification tasks, and show that the corpora it generates preserve much of the predictive power of the original ones. Our findings offer hope that institutions can reap ML insights by privately sharing data with simple prompts and little compute.

Accurate and efficient climate simulations are crucial for understanding Earth's evolving climate. However, current general circulation models (GCMs) face challenges in capturing unresolved physical processes, such as cloud and convection. A common solution is to adopt cloud resolving models, that provide more accurate results than the standard subgrid parametrisation schemes typically used in GCMs. However, cloud resolving models, also referred to as super paramtetrizations, remain computationally prohibitive. Hybrid modeling, which integrates deep learning with equation-based GCMs, offers a promising alternative but often struggles with long-term stability and accuracy issues. In this work, we find that water vapor oversaturation during condensation is a key factor compromising the stability of hybrid models. To address this, we introduce CondensNet, a novel neural network architecture that embeds a self-adaptive physical constraint to correct unphysical condensation processes. CondensNet effectively mitigates water vapor oversaturation, enhancing simulation stability while maintaining accuracy and improving computational efficiency compared to super parameterization schemes. We integrate CondensNet into a GCM to form PCNN-GCM (Physics-Constrained Neural Network GCM), a hybrid deep learning framework designed for long-term stable climate simulations in real-world conditions, including ocean and land. PCNN-GCM represents a significant milestone in hybrid climate modeling, as it shows a novel way to incorporate physical constraints adaptively, paving the way for accurate, lightweight, and stable long-term climate simulations.

Tensor decomposition is a fundamental tool for analyzing multi-dimensional data by learning low-rank factors to represent high-order interactions. While recent works on temporal tensor decomposition have made significant progress by incorporating continuous timestamps in latent factors, they still struggle with general tensor data with continuous indexes not only in the temporal mode but also in other modes, such as spatial coordinates in climate data. Additionally, the problem of determining the tensor rank remains largely unexplored in temporal tensor models. To address these limitations, we propose \underline{G}eneralized temporal tensor decomposition with \underline{R}ank-r\underline{E}vealing laten\underline{T}-ODE (GRET). Our approach encodes continuous spatial indexes as learnable Fourier features and employs neural ODEs in latent space to learn the temporal trajectories of factors. To automatically reveal the rank of temporal tensors, we introduce a rank-revealing Gaussian-Gamma prior over the factor trajectories. We develop an efficient variational inference scheme with an analytical evidence lower bound, enabling sampling-free optimization. Through extensive experiments on both synthetic and real-world datasets, we demonstrate that GRET not only reveals the underlying ranks of temporal tensors but also significantly outperforms existing methods in prediction performance and robustness against noise.

Context. As software systems become more integrated into society's infrastructure, the responsibility of software professionals to ensure compliance with various non-functional requirements increases. These requirements include security, safety, privacy, and, increasingly, non-discrimination. Motivation. Fairness in pricing algorithms grants equitable access to basic services without discriminating on the basis of protected attributes. Method. We replicate a previous empirical study that used black box testing to audit pricing algorithms used by Italian car insurance companies, accessible through a popular online system. With respect to the previous study, we enlarged the number of tests and the number of demographic variables under analysis. Results. Our work confirms and extends previous findings, highlighting the problematic permanence of discrimination across time: demographic variables significantly impact pricing to this day, with birthplace remaining the main discriminatory factor against individuals not born in Italian cities. We also found that driver profiles can determine the number of quotes available to the user, denying equal opportunities to all. Conclusion. The study underscores the importance of testing for non-discrimination in software systems that affect people's everyday lives. Performing algorithmic audits over time makes it possible to evaluate the evolution of such algorithms. It also demonstrates the role that empirical software engineering can play in making software systems more accountable.

Autoregressive attention-based time series forecasting (TSF) has drawn increasing interest, with mechanisms like linear attention sometimes outperforming vanilla attention. However, deeper Transformer architectures frequently misalign with autoregressive objectives, obscuring the underlying VAR structure embedded within linear attention and hindering their ability to capture the data generative processes in TSF. In this work, we first show that a single linear attention layer can be interpreted as a dynamic vector autoregressive (VAR) structure. We then explain that existing multi-layer Transformers have structural mismatches with the autoregressive forecasting objective, which impair interpretability and generalization ability. To address this, we show that by rearranging the MLP, attention, and input-output flow, multi-layer linear attention can also be aligned as a VAR model. Then, we propose Structural Aligned Mixture of VAR (SAMoVAR), a linear Transformer variant that integrates interpretable dynamic VAR weights for multivariate TSF. By aligning the Transformer architecture with autoregressive objectives, SAMoVAR delivers improved performance, interpretability, and computational efficiency, comparing to SOTA TSF models.

Accurate and efficient analysis of multivariate spatio-temporal data is critical in climatology, geology, and sociology applications. Existing models usually assume simple inter-dependence among variables, space, and time, and are computationally expensive. We propose a unified low rank tensor learning framework for multivariate spatio-temporal analysis, which can conveniently incorporate different properties in spatio-temporal data, such as spatial clustering and shared structure among variables. We demonstrate how the general framework can be applied to cokriging and forecasting tasks, and develop an efficient greedy algorithm to solve the resulting optimization problem with convergence guarantee. We conduct experiments on both synthetic datasets and real application datasets to demonstrate that our method is not only significantly faster than existing methods but also achieves lower estimation error.