2026年9月26日 技术热点总结

项目简介：Your Cheat Sheet For AI Engineering Interviews at Top AI Companies – Questions and Answers.

技术栈：Markdown、ai、ai-engineering、ai-engineering-interview、ai-interview、ai-interview-questions、company-wise-prep、company-wise-questions、interview-questions

核心介绍：> AI Engineering Interview Questions Company Wise – Your Cheat Sheet For AI Engineering Interviews at Top AI Companies > Real interview questions asked in AI Engineering interviews at 35 companies, organized company by company, with answers linked wherever we have them. > These interview questions and answers are helpful for r…

项目简介：A treemap for finding and removing what fills your disk, for Omarchy. Rust + GPUI.

技术栈：Rust

核心介绍：Find what is filling a disk, mark what should go, and remove it — with the volume’s free space in view the whole time. disktree is a treemap for Omarchy. It scans your home directory by default, draws every directory as a nested mosaic sized by what it really costs on disk, and lets you walk into it with the keyboard or the mouse. Mark as much as you like; nothing happens until you review the list and commit, and the permanent path always asks first. Built with GPUI through

*…

项目简介：Source code for the Claude Opus 5.5 music video for I’m Upping My P(doom)

技术栈：JavaScript

核心介绍：Source code for the Claude Opus 5.5 music video for *I’m Upping My P(doom)*. I’ve also made a new more reliable and general base for making Claude animations with Claude Opus 5.5. Highly recommend checking it out! The video took two generations, both in Claude Code: 1. First generation (legacy/): Claude Opus 5.5 (Medium) 2. Second generation (everything else): Claude Opus 5.5 ANIMATION_GUIDE.md was written by Opus to brief the subagents it ran in parallel.

项目数据：⭐ 8…

项目简介：Reference client daemon and verification worker for Kryvora Network nodes.

技术栈：Go、depin、distributed-systems、golang、infrastructure、node-runner、telemetry、worker-daemon

核心介绍：Official client daemon for Kryvora Network nodes. Kryvora is a decentralized physical infrastructure network designed for distributed verification and worker coordination. The kryvora-node daemon connects worker hardware to the Kryvora Hub, executes telemetry probes, verifies peer state, and reports proof of availability. +————————————————————-+ +——————…

Recently, Large Language Models (LLMs) have been increasingly able to solve advanced mathematical problems, including many that have been open for decades. This opens the door to expansion of mathematical knowledge at unprecedented scale. Yet, while LLMs may be able to conjecture and prove more and more theorems, it remains open whether this new mathematical knowledge is interesting or useful. We define intrinsic interestingness of a theorem as the ratio between the length of its proof and the length of its statement. We show that this correlates strongly with an extrinsic measure of the do…

In this paper, we propose RGBD20K, a novel dataset for facilitating the development of more robust and general RGB-D semantic segmentation by encompassing abundant categories and high-quality annotations. RGBD20K possesses several attractive properties: (1) Expanded Semantic Space. In particular, it covers 160 fine-grained categories, largely surpassing the category diversity of existing popular RGB-D benchmarks (e.g., NYUv2 with 40 classes and SUN RGB-D with 37 classes). With such enriched semantic coverage, we expect to promote the learning of more generalizable segmentation models. (2) L…

Recent years have witnessed major progress in joint audio-video generation. Existing models still suffer from limited per-modality fidelity, insufficient text-modality alignment and weak cross-modal synchronization. While reinforcement-learning post-training offers a promising remedy, directly adapting it to joint audio-video generation is challenging. Heterogeneous multimodal rewards entangle learning signals and complicate credit assignment. Joint optimization of two modality towers is computationally expensive given their divergent dynamics. Moreover, synchronization evaluation difficult…

While Large Language Models (LLMs) rely on highly non-linear components, in this work we demonstrate that they exhibit fundamental linearity: when inputs from distinct text streams are linearly combined, the model outputs a superposition of the individual next-token distributions. We term this the Superposition Linearity Hypothesis. We provide evidence that superposition is an intrinsic property of the Transformer architecture rather than an emergent consequence of training; in fact, we observe that it tends to diminish as pretraining progresses. However, we demonstrate that linearity can b…

Detectors of alignment failures screen deployed language models and score alignment benchmarks. Most are generative judges that spend a decoding pass on every criterion, and classifiers that read token probabilities, such as Llama Guard, still score one fixed label per call. Jev, a model trained with reinforcement learning for calibrated decisions (RLCD), answers many typed questions about one input with calibrated probabilities in a single call. Whether it detects alignment failures has not been measured. We present RLCDAlignBench, which benchmarks Jev on ten alignment failures: sycophancy…

Sparse AutoEncoders (SAEs) offer a promising way to inspect language model representations, but it is still unclear what kind of linguistic structure their latents expose. We use part-of-speech (PoS) categories as a controlled test case to study whether morpho-syntactic information is encoded by individual latents or by structured groups of features. We find that PoS distinctions are highly recoverable from SAE activations, but do not align with one-to-one latent / category mappings. This recoverability is not reducible to lexical memorisation, and Open and Closed PoS classes differ substan…