RAG: study evaluates retrieval techniques with a reranker
Published on June 30, 2026, the study examines RAG retrieval techniques with a strong cross-encoder reranker and tests whether reported gains hold across varied collections.
Read articleKnowledge you can apply
Radar
Daily bookmarks: relevant news, facts and discoveries.
Reducing the Train–Inference Gap in RL for LLMs
An article examines how separate engines can assign inconsistent probabilities to the same trajectories during LLM post-training, and describes TIS and rejecting checkpoint updates.
Read articleStanford course covers GPUs, CUDA, and parallel programming
Published on June 30, 2026, the listing presents a Stanford course with 19 lessons and 24 hours on GPU and CUDA architecture, performance optimization, and multicore architectures.
Read articleTwo-stage distillation for multitask agentic LLMs
Published on June 30, 2026, the summary describes consolidating reinforcement-learning specialists into a multitask model through off-policy initialization and on-policy refinement.
Read articleDockerless checks patches without running code
Published on June 30, 2026, the brief presents Dockerless, a verifier that assesses patches generated by coding agents without executing them.
Read articleEvoRec proposes agents to optimize recommender systems
Published on June 30, 2026, the summary presents EvoRec, a multi-agent framework intended to automate recommender-system optimization and reduce manual iteration.
Read articleHydraHead combines attention by head
Published on June 30, 2026, the post describes an architecture combining Full Attention and Linear Attention at the head level for long-context models.
Read articleIW-OPD reweights tokens in on-policy distillation
A June 30, 2026 publication presents IW-OPD, a method that gives more weight to early tokens. The post reports faster convergence and a 6.9-point advantage over standard OPD on AIME-2025.
Read articleKbSD proposes self-distillation to calibrate agentic search
Published on June 30, 2026, the summary introduces KbSD, a self-distillation framework using dense token-level supervision for agentic search decisions. It reports no evaluation results.
Read articleLearning rate may scale nonlinearly
A paper dated June 30, 2026 evaluates learning-rate scaling in GPT-2-style models and reports an upward trend for the optimal rate.
Read articleTradecraft Garden publishes research on evasion and linking
A June 30, 2026 post describes Crystal Palace, a linker with code transformation and randomization, YARA rule generation, and support for reusable tradecraft modules.
Read articleBranch prediction and software performance
Lesson 17 in the Digital Design and Computer Architecture series covers advanced branch prediction, a processor feature that can affect software performance.
Read articleCode agents: limits of reward signals
Published on June 30, 2026, the paper examines test approval, language-model judges, and execution traces as reward signals for code agents.
Read articleDiscoBench evaluates deep-search agents
Published on June 29, 2026, DiscoBench is described as a benchmark that tests LLM search agents on vague, underspecified, or factually incorrect requests. The description provides no performance results.
Read articleFrom nvcc to CUDA kernel execution
An article published on June 29, 2026 follows a vector-addition kernel from compilation with nvcc to execution by GPU warps.
Read articleUsing programs to explain attention heads
An article published on June 29, 2026 proposes approximating deep-network components with executable programs, focusing on attention heads in Transformers. It calculates attention matrices on randomly selected training examples and uses a pretrained language model.
Read articlePaper Assistant and scaling scientific review
An article treats the scalability of scientific review as a systems challenge and proposes four levels of collaboration between AI and people, while discussing Google Paper Assistant as an initial tool.
Read articlePaper Assistant and AI-assisted scientific review
Published on June 29, 2026, the item presents a paper proposing a taxonomy for AI-assisted scientific review and describes a system focused on detecting errors and reviewing manuscripts.
Read articlePaper explains attention heads with Python programs
An automated interpretability technique describes Transformer attention heads as Python programs; replacing about 40% of patterns in Llama-3B had almost no effect on task performance.
Read articleRed Queen Gödel Machine coevolves agents and evaluators
A self-improvement method evolves an agent and its evaluator, keeping evaluators fixed during each epoch and replacing them only after stronger performance on held-out ground truth.
Read articleRota Nacional
Bring privacy into your workflow.
30 days, no card, with a starting quota. After that, Pix credit from R$ 10,00.