President Trump probably can't get rid of her yet, but FCC commissioner Anna Gomez still checks her email every day to see if he has. Until then, she wants to stand up for the First Amendment. If you've given much thought to the Federal Communications Commission in recent years, it probably had something to do with Brendan Carr . The group's chairman since 2025, Carr has been on an ongoing, public rampage against freedom of speech: he's gone after late-night hosts like Jimmy Kimmel, threatened to revoke broadcast licenses over Iran war coverage, and targeted networks for their DEI policies. Disturbing as Carr's rhetoric and actions have been, he does count at least one opponent within the agency: Commissioner Anna Gomez, currently the lone Democrat among three FCC commissioners, has been vocal about the damage she thinks the agency is doing to American press freedom--and has repeatedly urged the public and the press, namely major networks like ABC, CBS, and NBC, to fight back. In May, Commissioner Gomez penned a stunning public letter to Disney CEO Josh D'Amaro, wherein she warned that the company--which owns ABC--was being subjected to "a sustained, coordinated campaign of censorship and control, carried out through the weaponization of the FCC's authority as a federal regulator and aimed at pressuring a free and independent press." Gomez urged D'Amaro to fight the actions her own agency was taking, adding that "this is a fight worth having, and one that I am confident you will win." I wanted to talk to Commissioner Gomez about that bold letter, the risks she sees for the media and the American public under the Trump administration, and how she works alongside a chairman with whom she disagrees so fiercely. Gomez, whose FCC term ends this month, was generous enough to sit down and talk about all of it. You can read our conversation below, or listen to it on the podcast platform of your choice. KATIE DRUMMOND: Welcome, Commissioner Gomez. Thank you for being here. It's great to be here. I want to start, before we talk more about Disney and your letter and all the rest of it, with a very basic question for our listeners. What is your agency's basic role?

Guidance techniques are commonly used in diffusion and flow models to improve image quality and input consistency for conditional generative tasks such as classconditional and text-to-image generation. In particular, classifier-free guidance (CFG) is the most widely adopted guidance technique. It results, however, in trade-offs across quality, diversity and consistency: improving some at the expense of others. While recent work has shown that it is possible to disentangle thesefactors to some extent, such methods come with an overhead of requiring an additional (weaker) model, or require more forward passes per sampling step. In this paper, we propose Entropy Rectifying Guidance (ERG), a simple and effective guidance method based on inference-time changes in the attention mechanism of state-of-the-art diffusion transformer architectures, which allows for simultaneousimprovements over image quality, diversity and prompt consistency. ERG is more general than CFG and similar guidance techniques, as it extends to unconditional sampling. We show that ERG results in significant improvements in various tasks, including text-to-image, class-conditional and unconditional image generation. We also show that ERG can be seamlessly combined with other recent guidance methods such as CADS and APG, further improving generation results.

We consider the problem of diversity maximization from the perspective of individual fairness: given a set P of n points in a metric space, we aim to extract a subset S of size k from P so that (1) the diversity of S is maximized and (2) S is individually fair in the sense that every point in P has at least one of its nk-nearest neighbors as its "representative" in S. We propose (O(1),3)-bicriteria approximation algorithms for the individually fair variants of the three most common diversity maximization problems, namely, max-min diversification, max-sum diversification, and sum-min diversification. Specifically, the proposed algorithms provide a set of points where every point in the dataset finds a point within a distance at most 3 times its distance to its nk-nearest neighbor while achieving a diversity value at most O(1) times lower than the optimal solution. Numerical experiments on real-world and synthetic datasets demonstrate that the proposed algorithms generate solutions that are individually fairer than those produced by unconstrained algorithms and incur only modest losses in diversity.

Diversified recommendation has attracted increasing attention from both researchers and practitioners, which can effectively address the homogeneity of recommended items. Existing approaches predominantly aim to infer the diversity of user preferences from observed user feedback. Nonetheless, due to inherent data biases, the observed data may not fully reflect user interests, where underexplored preferences can be overwhelmed or remain unmanifested. Failing to capture these preferences can lead to suboptimal diversity in recommendations. To fill this gap, this work aims to study diversified recommendation from a data-bias perspective.

Diverse decoding of large language models is crucial for applications requiring multiple semantically distinct responses, yet existing methods primarily achieve lexical rather than semantic diversity. This limitation significantly constrains Bestof-N strategies, group-based reinforcement learning, and data synthesis. While temperature sampling and diverse beam search modify token distributions or apply n-gram penalties, they fail to ensure meaningful semantic differentiation. We introduce Semantic-guided Diverse Decoding (SemDiD), operating directly in embedding space that balances quality with diversity through three complementary mechanisms: orthogonal directional guidance, dynamic inter-group repulsion, and position-debiased probability assessment.

Indian languages exhibit high dialectal variation and are spoken by populations that remain digitally underserved. Existing speech corpora typically represent only standard dialects and lack domain and linguistic diversity.

Maritime object detection is essential for navigation safety, surveillance, and autonomous operations, yet constrained by two key challenges: the scarcity of annotated maritime data and poor generalization across various maritime attributes (e.g., object category, viewpoint, location, and imaging environment). To address these challenges, we propose Neptune-X, a data-centric generative-selection framework that enhances training effectiveness by leveraging synthetic data generation with task-aware sample selection. From the generation perspective, we develop X-to-Maritime, a multi-modality-conditioned generative model that synthesizes diverse and realistic maritime scenes. A key component is the Bidirectional ObjectWater Attention module, which captures boundary interactions between objects and their aquatic surroundings to improve visual fidelity. To further improve downstream tasking performance, we propose Attribute-correlated Active Sampling, which dynamically selects synthetic samples based on their task relevance. To support robust benchmarking, we construct the Maritime Generation Dataset, the first dataset tailored for generative maritime learning, encompassing a wide range of semantic conditions. Extensive experiments demonstrate that our approach sets a new benchmark in maritime scene synthesis, significantly improving detection accuracy, particularly in challenging and previously underrepresented settings.

Conditional image generative models hold considerable promise to produce infinite amounts of synthetic training data. Yet, recent progress in generation quality has come at the expense of generation diversity, limiting the utility of these models as a source of synthetic training data. Although guidance-based approaches have been introduced to improve the utility of generated data by focusing on quality or diversity, the (implicit or explicit) utility functions oftentimes disregard the potential distribution shift between synthetic and real data. In this work, we introduce Chamfer Guidance: a training-free guidance approach which leverages a handful of real exemplar images to characterize the quality and diversity of synthetic data. We show that by leveraging the proposed Chamfer Guidance, we can boost the diversity of the generations w.r.t. a dataset of real images while maintaining or improving the generation quality on ImageNet-1k and standard geo-diversity benchmarks. Our approach achieves state-of-the-art few-shot performance with as little as 2 exemplar real images, obtaining 96.4% in terms of precision, and 86.4% in terms of distributional coverage, which increase to 97.5% and 92.7%, respectively, when using 32 real images.

Knowledge Distillation (KD) aims to train a lightweight student model by transferring knowledge from a large, high-capacity teacher. Recent studies have shown that leveraging diverse teacher perspectives can significantly improve distillation performance; however, achieving such diversity typically requires multiple teacher networks, leading to high computational costs. In this work, we propose a novel cost-efficient knowledge augmentation method for KD that generates diverse multiviews by attaching multiple branches to a single teacher. To ensure meaningful semantic variation across multi-views, we introduce two angular diversity objectives: 1) constrained inter-angle diversify loss, which maximizes angles between augmented views while preserving proximity to the original teacher output, and 2) intra-angle diversify loss, which encourages an even distribution of views around the original output. The ensembled knowledge from these angularly diverse views, along with the original teacher, is distilled into the student. We further theoretically demonstrate that our objectives increase the diversity among ensemble members and thereby reduce the upper bound of the ensemble's expected loss, leading to more effective distillation. Experimental results show that our method surpasses an existing knowledge augmentation method across diverse configurations. Moreover, the proposed method is compatible with other KD frameworks in a plug-and-play fashion, providing consistent improvements in generalization performance.

Diffusion models have made significant advancements in recent years. However, their performance often deteriorates when trained or fine-tuned on imbalanced datasets. This degradation is largely due to the disproportionate representation of majority and minority data in image-text pairs. In this paper, we propose a general fine-tuning approach, dubbed PoGDiff, to address this challenge. Rather than directly minimizing the KL divergence between the predicted and ground-truth distributions, PoGDiff replaces the ground-truth distribution with a Product of Gaussians (PoG), which is constructed by combining the original ground-truth targets with the predicted distribution conditioned on a neighboring text embedding. Experiments on real-world datasets demonstrate that our method effectively addresses the imbalance problem in diffusion models, improving both generation accuracy and quality.