Hugging Face Daily Papers5858JEPAs jointly embed predictive architectures in multimodal alignment observations, converging more efficiently to shared representations than next-word prediction, proving that it is the positive view design rather than scale itself that drives cross-modal alignment.
Hugging Face Blog4040MonsterHouse is an interactive projection built with NVIDIA Nemotron Voice Agent, NVIDIA Nemotron ASR, NVIDIA Nemotron 3.5 Bolt E4B 4GB, Magpie TTS Zeroshot, and Audio2Face-3D, which can communicate with visitors through voice and imitate the voice of a grumpy old pirate.
The Verge · AISame story · 2 outlets6565Google's Project Suncatcher satellite successfully launched its on-board Tensor Processing Units (TPUs), enabling the on-board Gemma 4 model to create haikus about space based on prompts.
How the accounts differ:TechCrunch states the satellite creates poetry; The Verge specifies haiku about space.
MarkTechPost6060Anthropic's Claude Security runs within Claude Code sessions without requiring an additional API key, outputting Markdown reports and SARIF files.
TechCrunch · AISame story · 2 outlets6060Google Suncatcher satellite, carrying four Tensor Processing Units (TPUs), has been launched by SpaceX and is now operational, successfully sending prompts to Google DeepMind's Gemma 4 large language model to create poems.
How the accounts differ:TechCrunch states the satellite creates poetry; The Verge specifies haiku about space.
AWS Machine Learning Blog3838FaceWall AI's operator generation system ForgeKernel generated the DeepSeek-V4.1 FlashMLA Prefill operator from scratch on an NVIDIA B200, with a final time of 472 microseconds, 9.8% faster than FlashMLA's 518 microseconds. The entire process took 22.5 hours, 42 iterations, and 32 versions, using 1 B200 and 22 GPU hours.
Why it matters: The original provided live stream iteration data and Ascend 3 benchmark results can be used to judge the path of operator generation shifting from manual tuning to intelligent agent self-evolution.
Zhiyuan released the embodied base model GO-3 Preview and the full-modal interactive large model WITA-OMNI at the Innovation Day. The former topped the RoboDojo, RoboTwin, and RoboColiseum rankings, while the latter led the Daily-Omni, TimeLens-Bench, EmbSpatialBench, and MMAU rankings.
Why it matters: Zhiyuan provides dual model rankings, tri-intelligent hardware, and an open-source OS plan at once, allowing comparison with its production line test data to judge the implementation level.
Factory News3434Factory acquires Typedef, providing deeper data context capabilities for enterprise software engineering. Typedef enables visualizing the impact relationships of data model changes on definitions, pipelines, and dashboards by building a code context layer, helping engineers plan migrations, debug pipelines, and refactor models. Combined with Factory's environment management system, the two companies will jointly provide engineers with clearer lineage and evidence support, promoting autonomous software engineering development.
Hugging Face Daily Papers6060Youtu-Parsing-Omni encodes documents, images, audio, and video into spatiotemporal structured records, achieving optimal parsing performance through self-evolution strategies and distillation. Single-modal encoders share weights and integrate cross-modal information, a single architecture unifies outputs for ten parsing tasks. The 5B parameter version achieved top scores of 96.96 and 75.08 on OmniDocBench v1.6 and OmniParsingBench, with code and weights open-sourced.
AMD releases AMD Software: Adrenalin Edition 26.10.2 Optional driver, fixes 10 issues in 'Immortal Wishes', no new game support or performance optimizations, targets RX 9000 / 7000 series graphics cards and Radeon RX graphics for Windows 10 and Windows 11 systems.
Hugging Face Daily Papers3131SpecSlot improves pixel space generation Transformer by using local spectral slots and limiting global attention without increasing computational cost, enhancing image details. SpecSlot expands each spatial anchor to multiple slots, assigning each slot a different DCT coefficient band, which separates anchor density and spectral refinement; only the low-frequency slots of each anchor participate in global attention, while all slots perform local interactions.
Google DeepMind5858PlanAssist tool can reduce the time planning officials spend assessing housing applications by 58%, has been piloted in Barnet, Camden, and Dorset in the UK, and is now being rolled out to 12 more local planning authorities. The tool was developed by Google DeepMind in collaboration with the UK government, Google Cloud, and Faculty, based on the Gemini architecture. It accelerates the process by automatically verifying documents, checking policy constraints, and drafting reports, while retaining the final decision-making authority of officials.
TechCrunch · AI4242Xiaomi open-sources lightweight document understanding model Xiaomi-OCR-0, with 0.8B parameter size, weights and inference code have been uploaded to Hugging Face and GitHub, a single model covers three tasks of document parsing, key information extraction, and OCR visual question answering.
Why it matters: 0.8B parameters rank second on the overall list in OmniDocBench v1.6, and weights, MCP services, and Skill toolkits are also opened, which can be used to judge the boundary of small models for document parsing.
The DecoderFeaturedSame story · 2 outlets8888Epoch AI researchers discovered that the Chinese open-source weight model GLM 5.3 can exploit the window between the release of security patches and their deployment to attack Firefox, covering 6 versions released from September 17 to October 26. The API cost for each successful attack is approximately 1000 US dollars.
How the accounts differ:The Decoder mentions the cost of 1000 dollars and that the vulnerability was fixed in Firefox 159.0, while Epoch AI does not mention these details.
Ant Group inclusionAI · Hugging Face new models3838inclusionAI releases ZwZ2-9B GGUF format weights, belonging to the Zooming without Zooming v2 project, including language models (FP16, Q8_0, Q4_K_M) and visual encoders and projectors (mmproj, FP16, Q8_0, Q4_K_M), requiring pairing with llama.cpp Qwen3.5 multimodal support for use, supporting CLI inference and Web Chat deployment.
Ant Group inclusionAI · Hugging Face new models2222inclusionAI/ZwZ2-27B-GGUF model weights released, including language models and visual components with FP16, Q8_0, Q4_K_M precision, supporting llama.cpp multimodal inference.
MarkTechPost5959Samaya AI releases the Sterling model, fine-tuned from the 30B mixed expert model Nemotron 3.5 Lightning, with 3B active parameters. In the FrontierFinance benchmark test, Sterling scored 58.9%, higher than Claude Fable 5.1's 55.9%, and the cost per question is only $0.61 (Fable 5.1 is $6.89).
Ant Group inclusionAI · Hugging Face new modelsThe inclusionAI/ZwZ2-4B-GGUF model weights have been released. This model is part of Zooming without Zooming v2 and is divided into a language model (FP16, Q8_0, Q4_K_M) and a visual encoder and projector (mmproj, FP16, Q8_0, Q4_K_M), offering multiple precision combinations. It needs to be used with llama.cpp (supporting Qwen3.5 multimodal).
NVIDIA and MIT team open source video enhancement solution Open-H3-Regenerate-2K, based on MiniMax H3 model to regenerate low-resolution video to 2K (2560×1440) or 4K (3840×2160) video.
Latent Space6565AIE CODE 2026 will explore three themes: redundant-free expansion, system-one model, and raising ambition, focusing on AI programming agent development, attracting 1000 top smart agent engineers globally. The conference will showcase the practical application of the System One model and focus on software rewrite cases from browser to Adobe Photoshop Suite.
Hugging Face Daily Papers5858L3G-World uses compact latent 3D Gaussian memory and video diffusion Transformers to alternately perform autoregressive updates, generating spatially consistent interactive worlds in real time. The model grows memory with the scene range rather than the flow length by generating Gaussians in new areas and refining existing Gaussians, each storing VAE latent features that can be directly rendered into DiT's latent space.
DeepSeek official website code exposed, Cyber security model application entry, including cyber-personal-v3 (personal version) and cyber-enterprise-v4 (enterprise version) two sets of application templates, personal version requires ID card face verification, enterprise version requires business license authorization letter, currently only small-scale gray testing.
Transformers releases66GitHub Copilot released v5.22.1 patch version, including the latest feature improvements and fixes.
Hugging Face Daily Papers4242WorkFactory builds coherent professional tasks through information planning and repair, producing the WorkPilot model, enhancing expert-level agent performance in professional work benchmarks. The framework starts from professional intent, constructs information support graphs to guide task instantiation, and identifies support failures and guides repairs through execution detection while preserving expected work.
Hugging Face Daily Papers6565In 18 languages, visual languages, and full-modal models, the retrieval head remains sparse in the newly added modalities (such as audio-visual) and has a high degree of cross-modal sharing. Full-modal models mainly retain the backbone model's retrieval heads; removing these retrieval heads eliminates almost all audio-visual retrieval capabilities.
Hugging Face Daily Papers6565LOWN adapts to skip image processing during diffusion visual language model decoding, reducing computational load while maintaining accuracy when predictions are consistent. Since visual markers account for 89-97% of input in the studied benchmarks, most decoding computations are used for repetitive processing of the same images.
The SoMark team has open-sourced the 1.3B document parsing model OXR, ranking first on three benchmarks: OmniDocBench (97.18), olmOCR-Bench (87.84), and DocParsingBench (87.76).
Why it matters: 1.3B model trained with RSI ranks first in three document parsing rankings, and provides engineering trade-offs between plug-in architecture and free API.
Hugging Face Daily Papers6565A-RSI refreshes training tasks based on learning signal decline and uses independently verifiable programs to generate answers, enhancing benchmark test performance without external labels. The model monitors reward changes and accuracy in training logs to refresh the task pool and utilizes a static model self-programming library to generate multimodal questions and their reference answers, while automatically verifying the properties of reward and refresh rules and auditing the generated answers. In eleven benchmark tests, A-RSI increased average performance from 57.77 to 61.05 with 800 new tasks.
Developer Divyansh Agrawal released the experimental model Qwenputer on October 22, built based on Alibaba's Qwen3.5-4B, which implements 32-bit RISC-V CPU and memory functions inside the neural network, allowing program execution and precise calculations without calling external tools like Python.
Why it matters: Developers compiled the RISC-V instruction set into the Qwen3.5-4B weights, demonstrating with specific benchmark numbers the improvement in computational precision within the model and the trade-offs in volume and speed.
Baidu has open-sourced the Wenxin Unlimited OCR 2 model, which supports parsing documents totaling over 3,600 pages at once, natively supporting layout element segmentation recognition and cross-page text merging, table recognition and merging. It scored 84.6 on the MPDocBench benchmark test and 77.1 on the MarathonDocBench benchmark test.
MarkTechPost5858Edge0 releases the open-source text-to-speech model Audio8-TTS-Realistic, with 0.6B parameters supporting zero-shot voice cloning for 23 languages, achieving an average error rate of 4.86% in the CV3 benchmark test, surpassing Qwen3-TTS-1.7B and VoxCPM2, but lower than the 4B-level Higgs TTS 3. The model employs a dual autoregressive architecture, including a slow branch with 24 layers and width 896, and a fast parallel prediction branch with 4 layers, trained using a learned realism critic.
Meta has launched the AI fashion app Style Sketch, where users can link their Instagram accounts to generate a virtual wardrobe and receive outfit recommendations, and can also try on clothes on a virtual model. The app supports iOS and Android systems and has already launched in Indonesia, Singapore, and Thailand.
The QQ HarmonyOS version App 9.2.80 (3570) beta upgrade has been launched, adding support for a simple mode, viewing space post views, enhanced group notification experience, selecting parts of plain text messages, AI repair for space albums, timed group mute and individual member mute, and chat number recognition features.
25 GPT-6.1 Terra agents run in Minecraft for 3 hours, aiming to build a prosperous civilization. They form a pseudo-government, establish rules, and develop complex coordination patterns. The agents initially focus on cutting down trees, then form a hierarchy, and communicate through approximately 5,900 messages.
Claude Code releases1515Claude Code v2.1.314 released, added write permission interception outside the working directory, optional identity group upstream routing, Bedrock upstream guard logs, idle conversation summary feature, fixed multiple session interruption, model switching, TLS connection, and tool call issues.
Gemini CLI releases1010Gemini CLI v0.67.0-nightly.20261030.gb633c0703 released, fixes case-insensitive isWithinRoot issue on Windows, identifies Gemini 3 models with dots, adds extension validation command and evaluation tests, handles leading slashes in file search queries, surfaces 429 rate limit cooling and countdown
SGLang releases00GitHub Copilot v0.5.23 released, providing code generation, review, security, and collaboration features, supporting custom integrations and advanced security features.
