2026年9月6日 技术热点总结

项目简介：养成类游戏英语学习

技术栈：JavaScript

核心介绍：> 创建一只属于你的 Q 版 OC，给它捏脸、换装、布置小家——而赚积分的唯一方式，是认真学英语。 > 截图由 npm run screenshots 自动生成（无头 Chrome + 契约级后端 mock 跑真实前端）。 起个名字就能迎接一只空白小人。可以创建多只 OC，每只都有独立的脸蛋、衣柜和小家，随时在人物栏一键切换。 积分是唯一的货币，只能靠学习赚取，用来买装扮和家具装扮 OC 与小家。

项目数据：⭐ 822 Stars，🍴 8 Forks

技术栈：Lean

核心介绍：A complete, machine-checked proof of Fermat’s Last Theorem in Lean 4, built on Mathlib (Lean 4.33.1; Mathlib v4.33.0, pinned by commit in lakefile.lean). The argument is that of Frey, Serre, Ribet, Wiles and Taylor-Wiles. PROOF-PATH.md names each step and the Lean theorem that carries it, and the html/ folder presents the whole proof as web pages you can browse offline (see “Reading the proof in a browser” below). Research artifact. Not maintained and not accepting contributions.

项目数据：⭐ 683 Stars，🍴 50 Forks

项目简介：Open-source AI brand visibility and competitor reports

技术栈：TypeScript

核心介绍：简体中文 · Quick start · Releases · Packages · Compare tools More users now ask AI directly instead of clicking through a page of search results: > What tools should I use? > What products exist in this category? > What are the alternatives to this product? > Which one should I choose? Your website may already be indexed by search engines, but AI may still:

项目数据：⭐ 458 Stars，🍴 33 Forks

项目简介：One-click setup of the leaked DLSS 5 neural-rendering build for any DX11/DX12 game on RTX 20–50, with or without DLSS. ReShade + RenoDX add-on (or OptiScaler engine); DLSS5-Feeder + LumeniteFX for games without DLSS; dlss5-bridge for DX11. Rust, single exe.

技术栈：Rust

核心介绍：One button that sets up the leaked DLSS 5 neural-rendering build in any DirectX 11/12 game, with or without DLSS of its own. Single native Windows exe, no runtime. Everything it installs is downloaded from the projects that made it; the only third-party content inside the exe is three SIL-OFL fonts. D…

Robot learning increasingly depends on broad and diverse demonstrations, yet collecting robot data remains expensive and poorly suited to covering the long tail of real-world tasks. To address this bottleneck, we introduce RoboTok, an internet-scale data engine that, given a query human manipulation video, retrieves manipulation-relevant human demonstrations from web videos for training dexterous robot policies. Specifically, we learn a latent motion space from 3D hand trajectories expressed in estimated actor-centered reference frames. This representation enables manipulation behaviors to …

Speech brain-computer interfaces (speech BCIs) translate neural activity into language, offering a path towards restoring speech for people with paralysis and, more broadly, enabling new forms of natural human-computer interaction. Despite this promise, the field lacks a common measure of progress because systems use different datasets, recording methods, types of speech, and vocabularies, so their reported scores are rarely comparable. Underlying this measurement problem are two unresolved questions: (i) what distribution of words should a speech BCI enable a user to communicate, and (ii) …

Visual fluency in generated video does not imply physical reliability, and a scalar quality score alone is incapable of indicating the obligation a clip violates or the moment it fails. We present VeriPhy, an auditable physical-verification system in which a text-only planner compiles the prompt into typed physical obligations and a statically validated execution plan before any frame is observed. During execution, observations gate and scope only declared calls to frozen low-level experts (e.g., segmentation and tracking, counting, eleven typed physical measurements over the resulting trac…

Reinforcement learning with verifiable rewards (RLVR) substantially improves single-sample accuracy (pass@1) but causes the policy's solution space to contract, diminishing the returns of test-time scaling. In this work, we investigate where inside a reasoning trajectory this breadth is lost: does the policy fail to access a valid solution family, or does it fail to execute computation once initiated? To disentangle access from execution, we analyze the Countdown task, whose solution space can be exhaustively enumerated into discrete entrance families defined by the first operand and operat…

Reinforcement Learning from Verifiable Rewards works well when a task has a programmatic checker, but most long-horizon agent domains have none. We work in the outcome-blind setting, where ground-truth success signals are not available. Multi-criteria rubrics are a popular way to supply such a reward; they are scored once per trajectory, but a single scalar is a poor signal across tens of steps. We propose DRACO: Distributing Rubric-based Advantage for Credit Optimization. It generates rubrics dynamically during training to track the policy's evolving capability, scores those rubrics once p…

For scientific progress, we need benchmarks that test the limits of state-of-the-art models, and evaluation methods that inform us about failure cases. As models get stronger, standard benchmarks for machine translation are approaching saturation. Further, automatic translation metrics are unreliable, vulnerable to reward-hacking, and provide unactionable assessments. Even gold human evaluation is not problem-free, because it often lacks reproducibility, objectivity, and scalability. Overall, this prevents us from tracking objective progress in the field and identifying pathways for improve…