Online Multi-Class Selection with Group Fairness Guarantee

This study explores a class of distributionally robust games where agents can choose their risk aversion levels in response to distributional shifts in uncertainty. By applying heterogeneous Wasserstein ball constraints through a Lagrangian formulation, the research formulates the distributionally robust Nash equilibrium problem. Under specific assumptions, this problem is shown to be equivalent to a finite-dimensional variational inequality problem. An approximate Nash equilibrium seeking algorithm is designed, demonstrating convergence of average regret through numerical simulations.

The article discusses the challenges of trajectory prediction, particularly in learning and predicting latencies, which are the temporal delays in agents' responses to trajectory-changing events. It highlights that different agents may have varying latency preferences, which can affect the accuracy of predictions. The authors propose a new reverberation transform and a corresponding model called Reverberation (Rev) that aims to simulate and predict these latency preferences, achieving competitive accuracy in trajectory forecasting.

Vision-Language Models (VLMs) are emerging models that integrate visual content with natural language. Current methods typically fuse data either early in the encoding process or late through pooled embeddings. This paper introduces a lightweight fusion module utilizing cross-only, bidirectional attention layers to align hidden states from both modalities, enhancing understanding while keeping encoders non-causal. The proposed method aims to improve the performance of VLMs by leveraging the inherent structure of visual and textual data.

The paper titled 'Bias-Restrained Prefix Representation Finetuning for Mathematical Reasoning' introduces a new method called Bias-REstrained Prefix Representation FineTuning (BREP ReFT). This approach aims to enhance the mathematical reasoning capabilities of models by addressing the limitations of existing Representation finetuning (ReFT) methods, which struggle with mathematical tasks. The study demonstrates that BREP ReFT outperforms both standard ReFT and weight-based Parameter-Efficient finetuning (PEFT) methods through extensive experiments.

The study investigates the ability of transformer models to predict long steps in the Collatz sequence, a complex arithmetic function that maps odd integers to their successors. The accuracy of the models varies significantly depending on the base used for encoding, achieving up to 99.7% accuracy for bases 24 and 32, while dropping to 37% and 25% for bases 11 and 3. Despite these variations, all models exhibit a common learning pattern, accurately predicting inputs with similar residuals modulo 2^p.

Higher-order Neural Additive Models (HONAMs) have been introduced as an advancement over Neural Additive Models (NAMs), which are known for their predictive performance and interpretability. HONAMs address the limitation of NAMs by effectively capturing feature interactions of arbitrary orders, enhancing predictive accuracy while maintaining interpretability, crucial for high-stakes applications. The source code for HONAM is publicly available on GitHub.