NEST is a nested transformer for sequences of multisets that uses masked set modeling to learn improved set-level representations from hierarchical event streams like EHRs.
On layer normalization in the transformer architecture
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5verdicts
UNVERDICTED 5representative citing papers
DAR replaces residual addition in DiTs with learnable, timestep-adaptive aggregation of sublayer outputs, yielding 2.11 FID improvement on SiT-XL/2 and 8.75x faster convergence on ImageNet 256x256.
A 1B-parameter hierarchical recurrent model pretrained on 40B instruction-response tokens achieves 60.7% MMLU and strong results on ARC-C, DROP, GSM8K, and MATH while using 100-900x fewer tokens than standard baselines.
Drafter models in speculative decoding suffer progressive attention drift caused by monotonically growing hidden-state magnitudes along the residual path; post-norm plus per-state RMSNorm reduces this drift and improves acceptance length up to 2x on perturbed templates and 1.18x on long-context data
VQ-VAE concept learning enables controllable recombination of crystal motifs to generate structures with reported gains in validity-stability-uniqueness-novelty metrics on MP-20 and Alex-MP-20.
citing papers explorer
-
NEST: Nested Event Stream Transformer for Sequences of Multisets
NEST is a nested transformer for sequences of multisets that uses masked set modeling to learn improved set-level representations from hierarchical event streams like EHRs.
-
Rethinking Cross-Layer Information Routing in Diffusion Transformers
DAR replaces residual addition in DiTs with learnable, timestep-adaptive aggregation of sublayer outputs, yielding 2.11 FID improvement on SiT-XL/2 and 8.75x faster convergence on ImageNet 256x256.
-
HRM-Text: Efficient Pretraining Beyond Scaling
A 1B-parameter hierarchical recurrent model pretrained on 40B instruction-response tokens achieves 60.7% MMLU and strong results on ARC-C, DROP, GSM8K, and MATH while using 100-900x fewer tokens than standard baselines.
-
Attention Drift: What Autoregressive Speculative Decoding Models Learn
Drafter models in speculative decoding suffer progressive attention drift caused by monotonically growing hidden-state magnitudes along the residual path; post-norm plus per-state RMSNorm reduces this drift and improves acceptance length up to 2x on perturbed templates and 1.18x on long-context data
-
Composable Crystals: Controllable Materials Discovery via Concept Learning
VQ-VAE concept learning enables controllable recombination of crystal motifs to generate structures with reported gains in validity-stability-uniqueness-novelty metrics on MP-20 and Alex-MP-20.