Skip to content

AI 速递 2026-06-04

生成时间:2026/6/4 10:33:10(UTC: 2026-06-04T02:33:10.245Z)

OCC-RAG: Optimal Cognitive Core for Faithful Question Answering

Section titled “OCC-RAG: Optimal Cognitive Core for Faithful Question Answering”

👍 73 · arXiv

Recent progress in the development of language models has been defined by scale, with each generation absorbing more of the world’s knowledge into its weights. However, many practical applications ben…

From Activation to Causality: Discovery of Causal Visual Representations in the Human Brain

Section titled “From Activation to Causality: Discovery of Causal Visual Representations in the Human Brain”

👍 42 · arXiv

Identifying which brain regions represent a visual concept in the human brain is a central challenge in neuroscience. Existing approaches have localized coarse functional regions (e.g., faces, places)…

KVarN: Variance-Normalized KV-Cache Quantization Mitigates Error Accumulation in Reasoning Tasks

Section titled “KVarN: Variance-Normalized KV-Cache Quantization Mitigates Error Accumulation in Reasoning Tasks”

👍 27 · arXiv

Test-time scaling is a powerful approach to obtain better reasoning in large language models, but it becomes memory-bottlenecked during long-horizon decoding, as the KV-cache grows. KV-cache quantizat…

A Local Perturbation Theory for Cross-Domain Interference and Recovery in Multi-Domain RL

Section titled “A Local Perturbation Theory for Cross-Domain Interference and Recovery in Multi-Domain RL”

👍 24 · arXiv

Reinforcement learning (RL) post-training improves large language models (LLMs) on individual domains such as mathematical reasoning, code generation, question answering, and creative writing (CW), bu…

World Models Meet Language Models: On the Complementarity of Concrete and Abstract Reasoning

Section titled “World Models Meet Language Models: On the Complementarity of Concrete and Abstract Reasoning”

👍 23 · arXiv

World models and multimodal large language models (MLLMs) provide complementary capabilities for predicting future outcomes from static visual observations. World models can generate concrete visual r…

  • Plugin and skill installs now use an operator install policy instead of the old dangerous-code scanner path, with clearer doctor, CLI, ClawHub, and troubleshooting surfa…

链接https://github.com/openclaw/openclaw/releases/tag/v2026.6.2-beta.1

Changes since langchain-deepseek==1.0.1

chore(infra): bump langchain-tests floor to 1.1.9 (#37610) chore: bump idna from 3.10 to 3.15 in /libs/partners/deepseek (#37560) ci(infra): harden Dependabo…

链接https://github.com/langchain-ai/langchain/releases/tag/langchain-deepseek%3D%3D1.1.0

链接https://github.com/ollama/ollama/releases/tag/v0.30.4

  • Add crew trained agents file support
  • Add native Snowflake Cortex LLM provider
  • Add Databricks integration guide
  • Add Snowflake integration guide

链接https://github.com/crewAIInc/crewAI/releases/tag/1.14.7a1

链接https://github.com/aaif-goose/goose/releases/tag/v1.37.0

  • TUI controls now support F13-F24 keybindings, paste in searchable menus, and a compact reasoning-only status/title item (#25329, #25400, #25504).
  • Enterprise/admin flows now show mo…

链接https://github.com/openai/codex/releases/tag/rust-v0.137.0

Lovable signs multiyear deal with Google Cloud to up usage 5x, source says

Section titled “Lovable signs multiyear deal with Google Cloud to up usage 5x, source says”

Lovable and Google signed an expanded multiyear deal that involves a 5x expansion of Lovable’s footprint on Google Cloud, and expanded access to Anthropic Claude.

来源TechCrunch AI

Alphabet’s record-breaking $85B raise for Google’s AI business is a helluva good signal

Section titled “Alphabet’s record-breaking $85B raise for Google’s AI business is a helluva good signal”

If Alphabet’s record-breaking $85 billion stock sale signals investor appetite for AI-related offerings, we can see that investors are ready to chow.

来源TechCrunch AI

Google’s Dreambeans, its weirdest-named AI tool to date, will turn your life into a cartoon

Section titled “Google’s Dreambeans, its weirdest-named AI tool to date, will turn your life into a cartoon”

Dreambeans is a curated list of AI-illustrated “stories” culled from the personal data in your Google account.

来源TechCrunch AI

Amazon will show AI product images when you search for some reason

Section titled “Amazon will show AI product images when you search for some reason”

Amazon will use visual search and AI to show AI-generated product images that match your search queries. The retailer says it will help guide users to products.

来源TechCrunch AI

These two founders left Goldman and Meta to build voice AI for markets everyone else overlooked

Section titled “These two founders left Goldman and Meta to build voice AI for markets everyone else overlooked”

The startup’s own stack for Africa and Middle East is now handling more than 17,000 calls per day.

来源TechCrunch AI

Publishers will be able to opt out of AI Search, thanks to new regulation

Section titled “Publishers will be able to opt out of AI Search, thanks to new regulation”

U.K. regulators are requiring Google offer a tool allowing website publishers to opt-out of generative AI search features. The option will be tested in the U.K. then rolled out globally.

来源TechCrunch AI

Meta’s AI agent for WhatsApp Business is now available globally

Section titled “Meta’s AI agent for WhatsApp Business is now available globally”

WhatsApp will charge businesses for using its AI agent based on token usage.

来源TechCrunch AI

Coralogix raises $200M on bet that someone needs to watch the AI agents

Section titled “Coralogix raises $200M on bet that someone needs to watch the AI agents”

Coralogix is among a growing number of infrastructure firms betting that as AI systems move into production, demand will rise for tools that can monitor their behavior, troubleshoot failures, and provide the operational data needed to keep them running reliably.

来源TechCrunch AI