Symmetry in variational inference forces approximate minimizers to recover target statistics under misspecification, unifying prior results and yielding new directional guarantees.
A time-sensitive testing-by-betting framework favors early rejection via time-weighted rewards, yielding Bellman-optimal e-processes and an exponential-decay-optimal criterion recovering classical growth-rate optimality at large scales.
TokenSwap benchmarks and reduces MLLMs' modality gap by interleaving visual tokens with text, finding reasoning models have smaller gaps and training with TokenSwap mitigates it.
Linear probes on LLM residual streams identify a shared preference vector tracking pairwise choices across personas, with cross-persona transfer and causal steering.