A 53K-parameter model generates 95% valid SMILES on ZINC-250K, outperforming larger models, by resolving chemical constraints in fixed order: brackets first, rings second, valence last.
Relaxed recursive transformers: Effective parameter sharing with layer-wise LoRA
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 6roles
background 1polarities
background 1representative citing papers
A recurrent-depth architecture enables language models to improve reasoning performance by iterating computation in latent space, achieving gains equivalent to much larger models on benchmarks.
CEM recasts Transformer layers as energy minimization steps, enabling constrained parameterizations like weight sharing and low-rank interactions that match standard baselines in 100M-scale language modeling.
Parcae stabilizes looped LLMs via spectral norm constraints on injection parameters, enabling power-law scaling for training FLOPs and saturating exponential scaling at test time that improves quality over fixed-depth baselines under fixed parameter budgets.
Circuit duplication on frozen DINOv3 embeddings raises macro F1 to 0.875 on AQUA20, within 1.4 points of supervised ConvNeXt, with class-specific circuits helping 75% of species.
Hyperloop Transformers outperform standard and mHC Transformers with roughly 50% fewer parameters by looping a middle block of layers and applying hyper-connections only after each loop.
citing papers explorer
-
SMolLM: Small Language Models Learn Small Molecular Grammar
A 53K-parameter model generates 95% valid SMILES on ZINC-250K, outperforming larger models, by resolving chemical constraints in fixed order: brackets first, rings second, valence last.
-
Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
A recurrent-depth architecture enables language models to improve reasoning performance by iterating computation in latent space, achieving gains equivalent to much larger models on benchmarks.
-
Revisiting Transformer Layer Parameterization Through Causal Energy Minimization
CEM recasts Transformer layers as energy minimization steps, enabling constrained parameterizations like weight sharing and low-rank interactions that match standard baselines in 100M-scale language modeling.
-
Parcae: Scaling Laws For Stable Looped Language Models
Parcae stabilizes looped LLMs via spectral norm constraints on injection parameters, enabling power-law scaling for training FLOPs and saturating exponential scaling at test time that improves quality over fixed-depth baselines under fixed parameter budgets.
-
Inference-Path Optimization via Circuit Duplication in Frozen Visual Transformers for Marine Species Classification
Circuit duplication on frozen DINOv3 embeddings raises macro F1 to 0.875 on AQUA20, within 1.4 points of supervised ConvNeXt, with class-specific circuits helping 75% of species.
-
Hyperloop Transformers
Hyperloop Transformers outperform standard and mHC Transformers with roughly 50% fewer parameters by looping a middle block of layers and applying hyper-connections only after each loop.