探索全球前沿学术脉络

Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds

What is the geometry of a visual percept? The most widely used protocols for decomposing neural network representations into interpretable parts treat concepts as isolated directions, yet recent work shows that concepts are often realized as geometric structures in low dimensional regions of activation space. We turn to the literature of Structured sparsity to close this gap, and show that block sparsity, which groups directions into blocks, is the prior matched to a generative model in which a representation is a sparse sum of low-dimensional manifolds: the modern, learned form of a classical idea in visual neuroscience, where a visual feature is carried by a coordinated group of neurons rather than a single tuned one. We implement three variants of block-sparse featurizers (BSFs) and, through a minimum-description-length analysis, show that all three describe activations more compactly than direction-based featurizers, with the recovered concepts typically two- to four-dimensional. We then use BSFs to (i) recontextualize prior work, showing that curve detectors in InceptionV1 actually read from a single continuous curve manifold, (ii) discover novel manifolds including shadows and lighting in DINOv3, and (iii) support interpretable control of image generation in diffusion models (SDXL) via manifold steering.

Anatomy of Post-Training: Using Interpretability to Characterize Data and Shape the Learning Signal

arXiv:2606.12360v1 Announce Type: new Abstract: Language-model post-training is the main stage at which model behavior is shaped, yet it still largely involves optimization of scalar rewards that summarize diverse desiderata. This abstraction gives practitioners little visibility into what their data actually teaches models, allowing spurious correlations to be learned by a model and inducing undesirable behaviors such as over-stylization and sycophancy. To address this problem, we ask: can we inspect a preference dataset before optimization and decide, at the level of concepts, which behaviors a model should be allowed to learn? Motivated by this, we introduce a data-centric post-training pipeline that uses interpretability protocols to develop statistical hypotheses for the latent concepts separating preferred from dispreferred generations, making them explicit for fine-grained user feedback. Building on this view, we unify several interpretability-based training protocols as ways of shaping rewards via feature or data interventions. Empirically, we show that our pipeline diagnoses undesirable signals in existing preference data, mitigates off-target learning, and can also help amplify or shape desired properties such as safeguards and model personality. More broadly, our results suggest that interpretability can turn post-training from optimizing opaque proxy rewards into a process of auditing and sculpting the learning signal itself.

Ribosomes are covered by a coat of flexible protein fragments

Ribosomal proteins contain flexible terminal regions that are averaged out during electron density reconstructions, rendering them absent from experimental models derived by X-ray crystallography or cryogenic electron microscopy. These flexible protein fragments (FPFs) collectively form an invisible coat on the ribosome surface whose presence has been systematically overlooked. Here we analysed FPFs from 36 ribosomes spanning bacteria, eukaryotes, and mitochondria. We found that mitoribosomes harbour the most numerous and longest FPFs. Structural predictions confirmed that FPFs are predominantly disordered across all ribosome classes. Comparison of FPF amino acid composition against proteome-wide background frequencies revealed strong and domain-specific compositional biases. The balance between arginine and lysine content tracks the cardiolipin content of the membrane each ribosome class contacts. The arginine enrichment in mitoribosomal FPFs may additionally reflect selection arising from the RNA-rich environment of mitochondrial RNA granules, membraneless condensates where mitoribosomes are assembled. FPFs are uniformly depleted in aromatic residues, arguing against protein-driven liquid–liquid phase separation propensity. Our findings suggest that the flexibly tethered coat is a highly functional intrinsic part of all ribosomes.

Differential COVID-19 Outcomes Across Lysosomal Disorders

Background Lysosomal disorders (LDs) are a heterogeneous group of rare inherited disorders characterized by multi-system involvement and high comorbidity burden, which raises concerns about severe COVID-19 outcomes. Conversely, because SARS-CoV-2 relies on endolysosomal pathways for cellular entry and replication, certain LDs may exert a protective effect against viral pathogenesis. Prior clinical evidence investigating LDs and severe SARS-CoV-2 infection has been limited by small sample sizes and inconsistent findings. Therefore, to resolve these conflicting biological hypotheses and estimate population-level outcomes, we conducted a large-scale retrospective cohort study using nationwide U.S. harmonized electronic health record data from the National Clinical Cohort Collaborative (N3C). This design utilized longitudinal records starting January 1, 2018, to evaluate COVID-19 infections captured between January 1, 2020, and July 11, 2024. Results The study included 16,380 individuals, comprising 5,460 patients with lysosomal disorders and 10,920 matched controls. Patients with LDs had significantly higher odds of COVID-19 hospitalization compared with controls (OR = 1.86, 95% CI: 1.70-2.04). Elevated odds were observed across the evaluated categories, but varied substantially. Notably, neurodegenerative LDs such as neuronal ceroid lipofuscinosis (OR = 9.32) and metachromatic leukodystrophy (OR = 2.33) remained associated with hospitalization after adjustment for comorbidities. Contrarily, the elevated odds for Fabry disease and Gaucher disease were no longer significant after adjustment. Mortality among hospitalized patients with LDs was comparable to that of matched controls (one-year survival: 82.1% vs 82.0%), suggesting that LD status does not independently worsen survival once hospitalization occurs. Conclusions Patients with LDs were at an increased odds of COVID-19 hospitalization, driven by a combination of elevated comorbidity burden and disorder-specific effects, which vary significantly across LD categories. This study clarifies that excess risk is concentrated in the transition to hospitalization. These patients may thus require personalized clinical care to mitigate the negative consequences of COVID-19.