NOVA explores agents for evolving recommendation systems
Published on June 26, 2026, the summary presents NOVA, an agent harness that proposes changes to recommendation architectures under domain constraints.
Read articleKnowledge you can apply
Radar
Daily bookmarks: relevant news, facts and discoveries.
TAPA proposes token-aware positional encoding
The article introduces Token-Aware Phase Attention (TAPA), a positional encoding method with a trainable phase function, and analyzes RoPE's distance-dependent bias. It reports a comparison of performance in long contexts.
Read articleYaFF proposes zero-copy serialization for Protobuf
The YaFF C++ library offers a zero-copy wire format for the Protobuf ecosystem. Engineers evaluating serialization options can consult its implementation and documentation.
Read articleAutodata proposes agents to create training and evaluation data
Published on June 25, 2026, the text presents Autodata, a staged method in which AI agents act as data scientists to create and analyze training and evaluation data, with meta-optimization.
Read articleAutoDecompiler explores RL for iterative decompilation
AutoDecompiler is described as a reinforcement-learning-optimized model that uses feedback across multiple steps for decompilation, unlike single-step approaches.
Read articleRadar: glibc malloc internals
A June 25, 2026 publication points to the glibc wiki page on malloc internals and notes its usefulness for analyzing memory allocation behavior.
Read articleRadar: inside glibc malloc
An article published on June 25, 2026, introduces aspects of the glibc malloc allocator, including how the heap is obtained from the kernel and how memory is managed.
Read articleFramework compares RAG approaches for semi-structured data
A paper published on June 25, 2026 presents a framework for comparing regular, Graph, Modular, and Agentic RAG on semi-structured knowledge bases, using nine standardized scenarios.
Read articleFuncAttn applies functional attention to continuous data
Published on June 25, 2026, the post introduces FuncAttn, a method applying attention between function spaces, and reports results in PDE solving, 3D segmentation, and regression.
Read articleiLLaDA: an 8B masked diffusion model
Published on June 25, 2026, the entry describes iLLaDA, an 8-billion-parameter masked diffusion language model trained from scratch with fully bidirectional attention.
Read articleOrnith-1.0 presents open models for agentic coding
Published on June 25, 2026, Ornith-1.0 is a family of open models ranging from 9 billion to 397 billion parameters. Reinforcement learning is used to optimize both code solutions and the scaffolds that guide them.
Read articlePostTrainBench: GLM 5.2 tops the ranking
Published on June 25, 2026, the report gives GLM 5.2 (Max reasoning) a score of 34.29%, versus 34.08% for Opus 4.8 Max, and reports zero failures across 84 GLM 5.2 attempts.
Read articleReversecore_MCP automates reverse engineering
Listed on June 25, 2026, the entry describes an MCP server for AI agents with tools for binary, malware, and vulnerability analysis.
Read articleReverse Engineering and Malware Analysis Roadmap
A briefing published on June 25, 2026, describes a GitHub roadmap covering topics and tools for reverse engineering and malware analysis.
Read articleVideo introduces LoRA and fine-tuning methods
Published on June 25, 2026, the video introduces LoRA, LoRA+, QLoRA, VeRA, and DoRA as related parameter-efficient fine-tuning methods.
Read articleIterative code improves scientific forecasts
A Nature study assessed repeated code generation and testing for scientific forecasting across six fields. Of 55 combinations of existing methods, 24 outperformed both methods from which they originated.
Read articleNatureBench evaluates coding agents on 90 tasks
Published on June 24, 2026, the briefing describes a benchmark based on 90 tasks from papers in the Nature family. According to the post, the top agent exceeded the published reference result on 17.8% of tasks.
Read articlePhysical neural network of metasurfaces trained without digital models
A Nature Communications article describes a physical neural network with four programmable metasurfaces that learns through electromagnetic interactions, without relying on digital models.
Read articleRiot announces Vanguard On-Demand
On June 24, 2026, Riot announced an option to run Vanguard only during gameplay, conditional on specific device security features being enabled.
Read articleLong-horizon RL: PPO or GRPO?
A Zhihu post published on June 24, 2026 compares GRPO and PPO for reinforcement learning tasks with LLMs, highlighting credit-assignment challenges in long, noisy agentic rollouts.
Read articleRota Nacional
Bring privacy into your workflow.
30 days, no card, with a starting quota. After that, Pix credit from R$ 10,00.