News
AI, cybersecurity, machine learning, scraping & programming news.
76 results on this page · clear filters
Text-to-Image Generative Modeling Without Language Pretraining
The dominant paradigm for text-to-image (T2I) generation has long relied on internet-scale language pretraining. Models such as those in the LlamaGen and JiT-T2I families incorporate large language model components pretrained on v...
Reinforcement Learning with Visually Attributable On-Policy Self-Distillation
Multimodal reinforcement learning with verifiable rewards (RLVR) provides reliable task-level supervision, yet offers limited guidance for localizing errors in visual reasoning. On-policy self-distillation (OPSD) complements this...
Rethinking the design space of Vanilla Pixel Diffusion Transformers
Pixel-space diffusion models have re-emerged as an alternative to latent diffusion, eliminating the need for a separately trained autoencoder and its associated bottleneck. A prior result from JiT demonstrated that a vanilla diffu...
CocoRNA: Collective RNA Design with Cooperative Multi-agent Reinforcement Learning
Designing RNA sequences that reliably fold into specific secondary structures is essential for understanding their biological functions but remains a challenging computational problem. RNA, or ribonucleic acid, stores genetic info...
Steering beyond Language: Improving Multilingual Training-Free Text Embeddings via Paired-Residual Subspace Removal
Deriving text embeddings directly from frozen large language models has emerged as a promising direction, yet existing methods focus almost exclusively on English, leaving multilingual settings largely unexplored. This work presen...
FuTransformer: FP8 prefill and INT4 decode for Llama-style inference on Hopper
Sparse autoencoders (SAEs) explain foundation model representations by decomposing activations into sparse combinations of learned dictionary atoms. However, the learned dictionaries can vary across training runs, limiting the rel...
SoftCache: Adaptive Step Size Caching for Diffusion Transformers
Diffusion and flow models generate a sample over iterative steps, each a full pass through a large network, and that pass dominates generation cost. Training-free caching accelerates them by reusing activations from recent evaluat...
PathFM: Fast Generative Forecasting of Multivariate Trajectories
Time series foundation models (TSFMs) are a hot topic for general-purpose forecasting, but most leading models predict marginal quantiles, leaving the dependence structure of future trajectories unspecified. This distinction matte...
Rethinking Routing Exploration in Modern Mixture-of-Experts
Mixture-of-experts (MoE) models have become a cornerstone approach for scaling neural networks beyond the limits of conventional dense architectures. By routing each input token to a subset of specialized computation paths, MoE sy...
Lean Graph Search for Formal Theorem Proving
Formal theorem proving sits at the intersection of logic, mathematics, and machine learning. In systems like Lean 4, provers must construct machine-checkable proofs by drawing on large formalized libraries of mathematical results....