Selection features papers on search and generative embeddings
Published on October 5, 2025, volume 124 of a weekly paper selection covers generative embeddings with test-time scaling and efficient retrievers for visual documents.
Read articleKnowledge you can apply
Radar
Daily bookmarks: relevant news, facts and discoveries.
RLAD tests short hints to guide LLM reasoning
A study trains a hint generator and a hint-conditioned solver with reinforcement learning. The report says it achieved 44% higher math accuracy on AIME 2025 than RL with long chains.
Read articleGrokking in modular addition: a three-stage explanation
A post published on October 4, 2025 describes a paper that presents grokking in modular addition as three stages and proposes an example threshold for generalization.
Read articleSlides link kernel regression to attention
Published on October 4, 2025, a ten-slide summary examines the relationship between kernel regression and the attention mechanism.
Read articleMaxsun and Abee announce a four-GPU workstation
Announcement dated October 4, 2025: Maxsun and Abee presented a workstation with four Intel Arc Pro B60 Dual 48G GPUs, for a total of 192 GB of VRAM.
Read articleDragon Hatchling explores local neural memory in language models
Published on October 3, 2025, the post describes Dragon Hatchling, a model whose connections store short-term memory through Hebbian learning, and its GPU-optimized version.
Read articleRLAD explores procedural hints for reasoning in LLMs
Published on October 3, 2025, this record presents RLAD, a two-player reinforcement learning framework for discovering natural-language hints that structure reasoning exploration.
Read articleckit.h offers utilities for C projects
Published on October 2, 2025, ckit.h is described as a small kit of essentials for C projects that engineers can assess as a source of reusable utilities.
Read articleSDL3 can fill the browser window in Emscripten
A publication dated October 2, 2025, shows an option for SDL3 games compiled with Emscripten to fill the browser window.
Read articleInside NVIDIA GPUs: matmul kernels
An article published October 1, 2025, covering GPU architecture and kernel techniques for high-performance matrix multiplication.
Read articleDragon Hatchling explores locally interacting neural particles
The Dragon Hatchling architecture combines locally interacting neural particles in a biologically inspired, scale-free network, with an emphasis on interpretability.
Read articleGPU kernels, graphs and profiling
A blog post covers kernels, graphs, profiling and related topics, potentially offering engineers an overview of GPU programming.
Read articleToolUniverse connects LLMs to research tools
Published on September 30, 2025, the post presents a shared interface for discovering and calling local and remote research tools via MCP. The library includes more than 600 tools and a drug-discovery demonstration.
Read articleGuide covers high-performance CUDA matmul kernels
Published on September 29, 2025, the material covers GPU memory hierarchy, the CUDA programming model, and PTX/SASS inspection to guide compiler output for matmul kernels.
Read articleLoRA recipes for reinforcement learning fine-tuning
A Thinky Machines post dated September 29, 2025 highlights LoRA configuration choices for reinforcement learning fine-tuning, including 10× higher learning rates and LoRA in every layer.
Read articleEnergy-based transformers refine next-token prediction
Published on September 27, 2025, the report describes an approach that iteratively reduces the energy assigned to next-token candidates. In tests with 44-million-parameter models on RedPajama-Data-v2, it outperformed same-sized vanilla transformers on three of four benchmarks.
Read articleFlashAttention-4: kernel optimization
Published on September 26, 2025, Modal’s post describes reverse engineering FlashAttention-4 and attributes a 20% performance gain to changes in computation and execution.
Read articleRLPT applies reinforcement learning to pre-training data
A September 25, 2025 publication presents RLPT, which uses next-segment rewards to train models on pre-training data without additional human annotations. It reports benchmark gains for Qwen3-4B-Base.
Read articleStanford course surveys self-improving agents
Published on September 25, 2025, the briefing introduces Stanford’s CS329A course and names three works on self-improving AI agents.
Read articleLinux boot chapter updated
Update dated September 24, 2025: the fourth linux-insides chapter on Linux kernel boot has been revised to cover modern versions.
Read articleRota Nacional
Bring privacy into your workflow.
30 days, no card, with a starting quota. After that, Pix credit from R$ 10,00.