Live intelligence

Latest AI updates

2 updates
Research

Inherited Adapter Experts for Replay-Free Domain-Incremental Learning

The paper introduces Lamarck, a replay-free domain-incremental learning method that freezes a pre-trained ViT and trains per-domain adapter experts, using a training-free Mahalanobis router to select one expert per test image. It argues that the experts, not the router, limit accuracy, and strengthens them with adapters that correct every image token after each deep ViT block. Each expert is initialized from its predecessor at no extra cost because domains share their classes. Against twelve frozen-backbone baselines under one protocol, Lamarck reports first on CDDB-Hard by 5.2 points, second on DomainNet and OfficeHome, and the highest mean accuracy.

Domain-incremental learningVision TransformerAdapter experts
Research

dNaViT proposes discrete visual tokens as a general-purpose representation

Researchers introduce dNaViT, a Discrete Native-resolution Vision Transformer that produces discrete visual primitives analogous to language tokens, built on a frozen pretrained encoder. The method uses lightweight residual modulation, a dual-branch objective, and grouped product quantization to preserve semantic and pixel-level information without inflating sequence lengths. The paper reports that dNaViT matches or exceeds task-specific baselines across 12 visual understanding benchmarks, 4 generation benchmarks, and an image editing benchmark. It is positioned as a versatile discrete visual tokenizer supporting unified multimodal encoding and high-fidelity image decoding.

visual tokenizationmultimodal modelsvision transformer