Daily updates, analyzing cutting-edge AI research papers
The paper shows that, instead of increasing the number of training tasks, controlling the diversity of scene‑skill combinations via a large‑scale Skill Graph dramatically improves terminal‑agent performance, with Qwen3‑32B surpassing a 480B model on the Terminal‑Bench 2.0 benchmark.
The paper introduces Agentic Harness Engineering (AHE), showing that a 10‑round evolution improves Coding Agent pass@1 from 69.7% to 77.0% on Terminal‑Bench 2—outperforming Codex‑CLI—and that the evolved harness transfers zero‑shot to SWE‑bench and multiple model families, thanks to three observability pillars.
DeepSeek releases an open‑source multimodal LLM that introduces a visual‑primitive framework—elevating bounding boxes and points to token level—to close the reference gap, achieve extreme KV‑cache compression, and outperform GPT‑5.4, Claude‑Sonnet‑4.6 and Gemini‑3‑Flash on counting, spatial reasoning, maze navigation and path‑tracing benchmarks.
The article highlights the rapid rise of reinforcement learning across major 2026 conferences, curates 181 RL papers from eight top venues, and provides detailed summaries of innovative works such as MSRL and MedVR, offering free access to the papers and code.
The article reviews the paper "Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond", introducing a two‑axis framework (capability levels L1‑L3 and law domains) to map diverse world‑modeling systems, highlighting that most current systems stall at L1, that explicit law encoding is crucial for long‑term stability, and that L3 represents the ultimate, self‑evolving model.
The article introduces the TRS (Thinking with Reasoning Skills) framework, which distills historical LLM reasoning traces into reusable skill cards, enabling offline skill‑base construction and online retrieval that dramatically reduces token consumption (6‑59%) and often improves accuracy on math and coding tasks.
MiniCPM‑o 4.5 introduces the world’s first end‑to‑end full‑duplex multimodal 9‑billion‑parameter model, powered by the Omni‑Flow framework, running on a single consumer‑grade GPU with 12 GB memory, and delivers benchmark results that match or surpass Gemini 2.5 Flash while offering open‑source demos, APIs, and a Windows/macOS installer.
The article surveys recent LLM‑based agent memory research, presenting a unified framework that breaks memory systems into four components, detailing their design choices, experimental evaluation on LOCOMO and LONGMEMEVAL, key findings, and a new low‑token SOTA architecture.
The ICLR 2026 Outstanding Paper awards spotlight two studies—one proving Transformers are mathematically succinct and another showing that all major LLMs lose about 39% performance in multi‑turn conversations, exposing a reliability gap missed by single‑turn benchmarks.
The article shares four concrete principles distilled from Andrej Karpathy’s experience—captured in the 86.1k‑star "andrej‑karpathy‑skills" repository—to help developers steer large language models toward reliable, concise, and goal‑driven code changes, with installation tips for Claude Code and other AI assistants.
