Li Zhang

A single adversarial token can bypass most safety filters.

Physical controls like damping and size can precisely guide audio foley generation, achieving near-perfect correlation with real-world sound targets without fine-tuning.

Current interpretability objectives can misrank circuits, favoring less faithful ones that mimic model behavior imperfectly, even when identical circuits are available.

Entity copying isn't a simple direct readout: late attention layers can execute verbatim retrieval entirely on their own, but the circuit catastrophically fails unless surrounding context tokens actively attend to the target entity first.

LLMs fail at chemical planning not from deficient reasoning capacity, but from an acute representation alignment bottleneck that disappears when structured through intermediate molecule and reaction abstractions.

Standard mechanistic circuits can mimic over 97% of a model's correct outputs while missing up to 89% of its errors, exposing that common ablation validations reward functional mimicry rather than true causal mechanisms.

This paper presents UltraTex, an efficient end-to-end framework for high-resolution multi-view diffusion-based 3D texturing and introduces Background Token Dropping, which removes background tokens before the DiT backbone, and Block-Sparse Attention, which reduces attention computation over the retained foreground sequence.