2026年9月18日 技术热点总结

技术栈：Python

核心介绍：Give it one goal. TypeSafe’s Jev picks an operation and an element. A small LLM writes text only when the operation is TYPE_TEXT. Watch the MP4 · Measurements · Read the loop Every observation produces a new element table: [1] button Change ticket type · Round trip [2] combobox Where from? · San Francisco [3] combobox Where to? · empty [4] textbox Departure · empty

项目数据：⭐ 1,984 Stars，🍴 108 Forks

项目简介：Can we run something like Jev on a 3090 at home?

技术栈：Python

核心介绍：Most agent decisions are small: *route this*, *retry that*, *does the evidence support X?* A chat model can answer them, but it spends time generating text that software immediately parses back into an if statement. Jev is TypeSafe’s closed service for runtime-defined semantic decisions. This project reproduces that interface pattern with open models; it does not reproduce Jev’s undisclosed model or training.

项目数据：⭐ 749 Stars，🍴 51 Forks

技术栈：Python

核心介绍：Train a small model that chooses among a changing list of text options. A Jev-like model takes a piece of text and a list of N text options. It returns one probability for each option. It does this in one pass instead of writing an answer word by word. Jev is TypeSafe’s commercial model for this kind of task. TypeSafe has not published its design. This repository is an independent starter model with the same input and output shape.

项目数据：⭐ 654 Stars，🍴 60 Forks

项目简介：⚡ Blitz Strike — a universal MCP penetration-testing toolbelt. Structured methodology: reconnaissance & attack-surface mapping, source-to-sink analysis, and live validation. 57 escalation chains, 130-tool catalog, intelligence data layer. One server, every agent.

技术栈：TypeScript、active-directory、blue-team、bug-bounty、cve、defensive-security、exploit、mcp、mcp-server

核心介绍：Blitz Strike is a structured penetration-testing methodology — reconnaissance, source analysis, and validation — delivered as a universal MCP server. It enumerates the attack surface (BLITZ), traces source-to-s…

项目简介：The offical repo for “Dream-RSI: Recursive Self-Improvement through Evolving Worlds”

核心介绍：through Evolving Worlds alt=”An agent must dream to recursively self-improve. History is the world it dreams in.”> Tong Zheng1,2, Xidong Wu1✉, Zheng Zhang1✉, Zhankui He3, Chaoyi Zhang1, Benjamin Coleman3, Ruoqiao Wei1, Haolin Liu4, Xue Wang1, Yue Zhuan1, Wang-Cheng Kang3, Renkai Xiang1, Heng Huang2, Xinwu Cheng1, Yunsong Guo1 1Google · 2University of Maryland, College Park · 3Google DeepMind · 4University of Virginia ✉ Corresponding authors

项目数据：⭐ 543 Stars，🍴 50 Forks

A walking robotic hand must use the same fingers to move its body, support its weight, and interact with the environment. We show how an anthropomorphic hand can learn these skills while retaining its finger design and position controller. Onboard power and computation make the platform self-contained. Our reinforcement learning approach accounts for the hand's unequal fingers, with training in a simulator calibrated from hardware measurements. In simulation, the hand moves faster with our reward formulation than with tuned rewards originally designed for quadrupeds. On hardware, task-speci…

Current Mixture-of-Agents (MoA) paradigms generally treat query routing and agent fine-tuning as separate processes, limiting their ability to respond to evolving agent capabilities. This disconnect prevents routing strategies from adapting to evolving agent capabilities during post-training and prevents agents from achieving synergistic data-driven specialization. To resolve this, we introduce CERA-MoA (Co-Evolving Router with continually learning Agents for Mixture-of-Agents), an iterative reinforcement learning framework where the dynamic router and independent agent policies co-evolve. …

When agentic sessions run to a million tokens with many sessions resident at once, the KV cache and the index that ranks it live in host memory, and the scan that ranks all n keys for a top-k step becomes the traffic that bounds decoding. We present Fathom, a key scan in which each query decides how many bits of each key channel to read. The 4-bit K cache is stored channel-major as bit planes, so a prefix of t planes is exactly the channel's t-bit quantizer, and the query spends its bit budget by reverse water-filling over the variance-weighted importance of its channels. At one million tok…

Enabling robots to adapt to unfamiliar environments as readily as humans remains a moonshot goal of embodied AI. No finite collection of demonstrations can cover every task and situation a robot will encounter, making the ability to learn from context at deployment essential for generalization. Such in-context learning (ICL), however, remains largely beyond the reach of existing robotic policies. The broad agentic capabilities of commercial vision-language models (VLMs), such as GPT-6 Astra, raise a compelling question: can these models learn from demonstrations, examples, and interaction f…

BraTS-GoAT evaluates tumor segmentation across heterogeneous populations. We trained a conventional 3D nnU-Net on 1,351 labeled cases using five-fold cross-validation and 1,000 epochs per fold. The final predictor averaged all folds and applied test-time mirroring. On pooled official validation, global DSC values were 0.7805, 0.8288, and 0.8854 for enhancing tumor (ET), tumor core (TC), and whole tumor (WT). Under matched fold-0 inference, mean regional Dice decreased from 0.9058 on source out-of-fold (OOF) cases to 0.8310 on pooled validation (difference–0.0747). Mirroring gave small sing…

Mixture-of-experts (MoE) inference on consumer hardware is bounded by weight memory: a 35B-class model is 19.5GB at 4-bit, and sparsity shrinks the compute per token, not the bytes that must be held. Naive offloading to SSD does not help on its own, because layer N+1's experts must be chosen before layer N's output exists, so the reads cannot start early enough to hide behind compute. We present Edge0, a streaming MoE inference engine that closes the gap with a prerouter: a per-layer head predicts the next layer's routing one token ahead, and the prediction is consumed as the routing itself…