Skip to content
Cut to the source.The real-time AI news wire.
Saved in a cookie.

Search tips

  • "exact phrase" the words together, in this order
  • -word leave out stories with the word
  • topic:robotics a topic by ID or name; quote names with spaces: topic:"open-weight models"
  • category:"AI agents" a category
  • company:openai a company, person, product or organization (also entity:)
  • source:forbes.com a publisher by site or name
  • type:analysis news, analysis, opinion, research paper and other kinds
  • after:2026-11-01 stories on or after that New York day; before:2026-11-10 on or before it
Subscribe
NewsletterAboutModels & ResearchAI AgentsAI DevelopmentInfrastructure & HardwareImage, Video & AudioRobotics & Autonomous SystemsBusiness & IndustryPolicy & RegulationSafety & SecurityBenchmarks & EvaluationOpen Source & CommunityAI in Society
Models & ResearchAI AgentsAI DevelopmentInfrastructure & HardwareImage, Video & AudioRobotics & Autonomous SystemsBusiness & IndustryPolicy & RegulationSafety & SecurityBenchmarks & EvaluationOpen Source & CommunityAI in Society
Robotics & Autonomous SystemsBusiness & IndustryPolicy & RegulationSafety & SecurityBenchmarks & EvaluationOpen Source & CommunityAI in Society

15 NovModels & Researchcasrai.org

Sketching Speeds Up LLM Context Attribution

Sketching Speeds Up LLM Context Attribution

AIAI-generated summary Researchers from Google Research and Yale University released a preprint introducing a randomized sketching algorithm for language model context attribution. The team validated their algorithm using the Llama-3.2-1B-Instruct model on an NVIDIA H100 GPU to estimate leave-one-out prompt attribution without evaluating each token individually.

NVIDIA Google Llama Long Context Inference Optimization

Earlier:
Sketching Algorithm Cuts Context Attribution Cost for…
Read at casrai.org

Share onXLinkedInBlueskyEmail

Latest stories

  • GPT-6.2 Astra, the #2 Frontier Model is better and 20%-600% cheaper, but has mixed reviews; 6.2 Sol likely main workhorse

    20 November 2026

    Latent Space — November 17, 2026 — GPT-6.2 Astra and Sol lower benchmark costs. Early testers disagree on what Astra is like to work with.

  • Qwen Code v0.28.0: Qwen4-Max support, unified agent execution paths

    20 November 2026

    devotd.net — The new Qwen Code release supports model-specific settings for Qwen4-Max and unifies default execution paths for terminal sessions, headless sessions…

  • Nvidia weighs more funding for Samsung Electro-Mechanics (009150)

    20 November 2026

    AI Stock Wire — Samsung Electro-Mechanics said Nvidia is considering more funding for its Sejong and Busan plants, with output from 2029, after the two amended their substrate deal.

  • Clay Adds Sales Workflow Building and Previews to ChatGPT Plugin

    20 November 2026

    Unite.AI — Clay announced on November 17, 2026 that its relaunched MCP plugin can construct and preview sales workflows directly inside ChatGPT, covering prospect selection, data enrichment and campaign launch. The plugin connects...

  • LeBron James sues AI‑powered betting startup over unauthorized likeness

    20 November 2026

    aiunderstanding.org — NBA star LeBron James filed a lawsuit in New York federal court accusing PlayVision’s Bron app of using an AI‑generated version of his image, voice and nickname in a viral ad without permission.

TruncatedTruncated

Continuous coverage across AI research, agentic systems, infrastructure, and regulation.

Content

  • Latest
  • Archive
  • Topics
  • Newsletter
  • RSS Feed

About

  • About
  • Developers
  • Publishers
  • Advertise

Legal

  • Impressum
  • Privacy Policy
  • Terms & Conditions
  • Disclaimer
  • Contact

© 2026 Truncated. All rights reserved.

Truncated™ is a trademark of Three Things Media.