Multi-task learning (MTL) is a paradigm that aims to improve the generalization of models by simultaneously learning multiple related tasks, leveraging shared representations and task-specific information to enhance performance on individual tasks. However, existing work has shown that MTL can potentially hinder generalization, with one key factor being spurious correlations between tasks. Owing to the knowledge-sharing property, the per-task predictors are more likely to develop reliance on spurious features. Most existing approaches address this issue through distributional robustness, aiming to maintain consistent performance across different distributions under unknown covariate shifts. However, this formulation lacks theoretical guarantees and can be sensitive to the construction of covariate shifts. In this work, we propose a novel perspective, where we seek to identify spurious correlations between tasks. Drawing inspirations from conventional formulations on spurious correlation, for each task, we propose to distinguish its spurious tasks using the difference in correlation coefficients between the empirical distribution and class-wise resampled distributions, thereby capturing the correlations between task labels w.r.t. each class. We prove theoretically the feasibility of the resampling strategy in characterizing spurious correlations between tasks. Furthermore, we propose a simple fine-tuning strategy, de-biased adversarial training, where the per-task predictors are adversarially trained to disregard information associated with their spurious tasks. Experimental results on six benchmark datasets show that our method effectively mitigates spurious correlations and outperforms state-of-the-art methods in improving generalization.

Multimodal models or Vision Language Models (VLMs) have reshaped the paradigm in machine learning, offering zero-shot capabilities that require no additional training when adapted to new classification tasks. However, despite their advancements, spurious correlations still exist in VLMs. Existing approaches to tackle this issue often require target label annotations, contradicting the principle of zero-shot classification, or they primarily focus on a single modality, risking misalignment between text and image modalities. Others rely on extensive domain knowledge or large language models (LLMs) to characterize spurious features, making the performance sensitive to the generated prompts and undermining zero-shot capability. In response, we propose a new solution that tackles spurious correlations in VLMs within the zero-shot setting. Our approach utilizes a translation operation that preserves the latent space distribution to address issues of spurious correlations. In particular, our method is grounded in and inspired by a theoretical analysis, which identifies that the optimal translation directions are along the spurious vector. As VLMs unify two modalities, we compute spurious vectors from the text prompts and guide the translation for image embeddings, aligning the requirements for the fusion of different modalities in VLMs. We conducted experiments on benchmark datasets, which have shown significant improvements in worst-group accuracy. Additionally, our visualizations of VLMs further demonstrate the effectiveness of this intervention.

In computational biology, accurately modeling biological systems is essential for understanding the underlying mechanisms of biological processes. One of the primary challenges in modeling lies in the inability to measure certain biological parameters directly. The task of identifying the parameters is known as the inverse problem and it entails estimating these unobservable parameters from available data. Existing modeling methods primarily focus on single-direction prediction. These methods include traditional Partial Differential Equation (PDE) modeling and machine learning approaches where given parameters are provided to predict the output and further compare the output with experimental evidence. However, these unidirectional methods struggle to effectively apply the experimental evidence directly to address the inverse problem. We contend that a single biological process should be modeled bidirectionally to simultaneously address the inverse problem. To this end, we propose leveraging the capabilities of the invertible neural network (INN) to establish a connection between the input parameter space and the model output space. We meticulously designed a bidirectional training technique that effectively applies real-world experimental data in guiding the INN in modeling the biological process. We tested our approach on a PDE-based Bone Morphogenic Protein (BMP) signaling network system in the zebrafish embryo and found the bidirectional modeling approach significantly enhances the alignment between simulation and experimental data. This method achieves a 94.65% reduction in Root Mean Square Error (RMSE) compared to the single-direction model when reconstructing experimental data using simulation with INN-identified parameters. Moreover, our method is rapid to implement, facilitating the precise identification of parameter ranges from experimental data. It makes parameter optimization feasible and provides a guideline for determining simulation parameter ranges, bypassing the traditional and laborious trialand-error method.

Although transformers demonstrate impressive capabilities in a variety of tasks, the fairness issue remains a significant concern when deploying these models. Existing works to address fairness issues in transformers require sensitive labels (such as age, gender, etc.), which can raise privacy concerns or violate legal regulations. An alternative way is through fairness without demographics. However, existing works that improve Rawlsian Max-Min fairness may impose overly restrictive constraints. Other methods that use auxiliary networks could be parameter inefficient. In this paper, we present a new approach to debiasing transformers by leveraging their inherent structure. By reconsidering the roles of important components (queries, keys, and values) in the attention mechanism, we introduce a simple yet effective debiasing strategy from two perspectives: 1) Grounded in theoretical analysis, we normalize and apply absolute value operations to queries and keys to minimize the bias in attention weight allocation; 2) We reduce the bias within values through local alignment via contrastive learning. Throughout the entire process, our approach does not require any sensitive labels. Furthermore, to enhance memory efficiency in the training phase, we propose a strategy that debias only the last encoder to improve fairness in pre-trained models. We conduct experiments in computer vision and natural language processing tasks and show that our method is comparable and even outperforms the state-of-the-art method with substantially lower energy consumption.

Although machine learning algorithms demonstrate impressive performance, their trustworthiness remains a critical issue, particularly concerning fairness when implemented in real-world applications. Many notions of group fairness aim to minimize disparities in performance across protected groups. However, it can inadvertently reduce performance in certain groups, leading to sub-optimal outcomes. In contrast, Min-max group fairness notion prioritizes the improvement for the worst-performing group, thereby advocating a utility-promoting approach to fairness. However, it has been proven that existing efforts to achieve Min-max fairness exhibit limited effectiveness. In response to this challenge, we leverage the recently proposed "Neural Collapse'' framework to re-examine Empirical Risk Minimization (ERM) training, specifically investigating the root causes of poor performance in minority groups. The layer-peeled model is employed to decompose a network into two parts: an encoder to learn latent representation, and a subsequent classifier, with a systematic characterization of their training behaviors being conducted. Our analysis reveals that while classifiers achieve maximum separation, the separability of representations is insufficient, particularly for minority groups. This indicates the sub-optimal performance in minority groups stems from less separable representations, rather than classifiers. To tackle this issue, we introduce a novel strategy that incorporates a frozen classifier to directly enhance representation. Furthermore, we introduce two easily implemented loss functions to guide the learning process. The experimental assessments carried out on real-world benchmark datasets spanning the domains of Computer Vision, Natural Language Processing, and Tabular data demonstrate that our approach outperforms existing state-of-the-art methods in promoting the Min-max fairness notion.