Paper Reviews by AI

CURA: Verbal process supervision improves coding agents.

Bottleneck Sampling: Accelerate diffusion models without retraining by cleverly using low-resolution priors for efficient inference!

MetaSpatial: RL for 3D Spatial Reasoning in VLMs

LookAhead Tuning: Safer LLMs via Partial Answer Previews by preserving initial token distributions.

LSRNA: Super-resolution in latent space enhances image generation with diffusion models, achieving faster speeds and improved detail.

LLMs’ reasoning is decoded via sparse autoencoders, revealing key features that, when steered, enhance performance. First mechanistic account of reasoning in LLMs!

FRESA: fast feedforward 3D personalized avatar creation from few images.

FDConv: Adaptable convolution via frequency domain learning, enhancing performance without heavy parameter cost.

FFN Fusion: Parallelizing sequential computation in large language models for significant speedups!

Aligning image generation subtasks: Equivariant modeling boosts efficiency and generalization by leveraging natural visual signal invariance.

Diffusion-4K: Synthesizing ultra-high-resolution images with a new benchmark dataset and wavelet-based fine-tuning that makes 4K image creation more detailed and accessible!

COMP: Continually pre-training Vision Foundation Models for better vision and language alignment and arbitrary size inputs.

CFG-Zero*: A better Classifier-Free Guidance to improve the image quality and text alignment in Flow Matching models.

Hummingbird: An efficient text-to-video model that balances quality and computational efficiency via pruning and visual feedback learning.

AlphaSpace enables robotic actions via semantic tokenization and symbolic reasoning, enhancing spatial intelligence in LLMs.

AETHER: a unified framework enabling geometry-aware reasoning in world models, achieving zero-shot generalization from synthetic to real-world data.

Vision-R1: Improves LVLMs via vision-guided reinforcement learning, eliminating the need for human feedback and specialized reward models.

PathoHR: Boost breast cancer survival prediction with high-resolution pathology images!

OmnimatteZero: Real-time omnimatte using pre-trained video diffusion, no training needed!

LLMs falter on culturally adapted math problems, revealing a critical cultural bias.