AI 速递 2026-06-25
生成时间：2026/6/25 10:05:03（UTC: 2026-06-25T02:05:03.945Z）
Qwen-AgentWorld: Language World Models for General Agents
Section titled “Qwen-AgentWorld: Language World Models for General Agents”👍 90 · arXiv
A world model predicts environment dynamics based on current observations and actions, serving as a core cognitive mechanism for reasoning and planning. In this work, we investigate how world modeling…
NatureBench: Can Coding Agents Match the Published SOTA of Nature-Family Papers?
Section titled “NatureBench: Can Coding Agents Match the Published SOTA of Nature-Family Papers?”👍 49 · arXiv
We introduce NatureBench, a cross-discipline benchmark of 90 tasks distilled from peer-reviewed Nature-family publications, designed to evaluate whether AI coding agents can move beyond reproduction t…
MobileForge: Annotation-Free Adaptation for Mobile GUI Agents with Hierarchical Feedback-Guided Policy Optimization
Section titled “MobileForge: Annotation-Free Adaptation for Mobile GUI Agents with Hierarchical Feedback-Guided Policy Optimization”👍 35 · arXiv
MLLM-based mobile GUI agents have made substantial progress in UI understanding and action execution, but adapting them to real target apps remains costly because mobile apps are numerous, frequently …
MemGUI-Agent: An End-to-End Long-Horizon Mobile GUI Agent with Proactive Context Management
Section titled “MemGUI-Agent: An End-to-End Long-Horizon Mobile GUI Agent with Proactive Context Management”👍 33 · arXiv
MLLM-based mobile GUI agents have made substantial progress on short-horizon tasks, yet remain unreliable on long-horizon tasks that require retaining intermediate facts across many steps and app tran…
OpenThoughts-Agent: Data Recipes for Agentic Models
Section titled “OpenThoughts-Agent: Data Recipes for Agentic Models”👍 28 · arXiv
Agentic language models dramatically expand the applications of AI yet little is publicly known about how to curate training data for broadly capable agents. Existing open efforts such as SWE-Smith, S…
OpenClaw v2026.6.11-beta.1
Section titled “OpenClaw v2026.6.11-beta.1”2026.6.11
Section titled “2026.6.11”Highlights
Section titled “Highlights”- More capable channel control: Slack relay mode, native Mattermost
/oc_queue, and per-DM model overrides make channel operations easier to automate and tune. (#947…
链接：https://github.com/openclaw/openclaw/releases/tag/v2026.6.11-beta.1
LangChain langchain-openrouter==0.2.4
Section titled “LangChain langchain-openrouter==0.2.4”Changes since langchain-openrouter==0.2.3
release(openrouter): 0.2.4 (#38381)
chore(openrouter): bump openrouter floor to 0.9.2, drop file workaround (#38216)
test(openrouter): cover `cache_control…
链接：https://github.com/langchain-ai/langchain/releases/tag/langchain-openrouter%3D%3D0.2.4
Ollama v0.30.11-rc0
Section titled “Ollama v0.30.11-rc0”What’s Changed
Section titled “What’s Changed”- launch: add thinking capability detection to opencode by @hoyyeva in https://github.com/ollama/ollama/pull/15434
- launch: auto-install Claude Code by @hoyyeva in https://github…
链接：https://github.com/ollama/ollama/releases/tag/v0.30.11-rc0
LlamaIndex v0.14.23
Section titled “LlamaIndex v0.14.23”Release Notes
Section titled “Release Notes”[2026-06-24]
Section titled “[2026-06-24]”llama-index-callbacks-argilla [0.5.0]
Section titled “llama-index-callbacks-argilla [0.5.0]”- chore(deps): bump the uv group across 32 directories with 3 updates ([#21664](https://github.com/run-llama/llama_index/pu…
链接：https://github.com/run-llama/llama_index/releases/tag/v0.14.23
CrewAI 1.14.8a5
Section titled “CrewAI 1.14.8a5”What’s Changed
Section titled “What’s Changed”Features
Section titled “Features”- Make declarative refs work across flows and crews (#6326)
Bug Fixes
Section titled “Bug Fixes”- Fix JSON schema flow state kickoff inputs (#6325)
Documentation
Section titled “Documentation”- Nest One Card per St…
链接：https://github.com/crewAIInc/crewAI/releases/tag/1.14.8a5
OpenAI Codex CLI rust-v0.142.1
Section titled “OpenAI Codex CLI rust-v0.142.1”New Features
Section titled “New Features”- Added opt-in Windows system proxy support for authentication, including PAC, WPAD, static proxies, and bypass rules. (#26708)
Changelog
Section titled “Changelog”Full Changelog: https://github.com/opena…
链接：https://github.com/openai/codex/releases/tag/rust-v0.142.1
Anthropic says Alibaba illicitly extracted Claude AI model capabilities
Section titled “Anthropic says Alibaba illicitly extracted Claude AI model capabilities”Article URL: https://www.reuters.com/world/china/anthropic-says-alibaba-illicitly-extracted-claude-ai-model-capabilities-2026-06-24/ Comments URL: https://news.ycombinator.com/item?id=48664814 Points: 53
Comments: 87
Section titled “Comments: 87”Big AI labs are hiring philosophers
Section titled “Big AI labs are hiring philosophers”Comments URL: https://news.ycombinator.com/item?id=48662452 Points: 127
Comments: 113
Section titled “Comments: 113”For most of the world, open-source AI is the only way forward
Section titled “For most of the world, open-source AI is the only way forward”Article URL: https://techstrong.ai/articles/for-most-of-the-world-open-source-ai-is-the-only-way-forward/ Comments URL: https://news.ycombinator.com/item?id=48660839 Points: 203
Comments: 131
Section titled “Comments: 131”RubyLLM: A Ruby framework for all major AI providers
Section titled “RubyLLM: A Ruby framework for all major AI providers”Article URL: https://rubyllm.com/ Comments URL: https://news.ycombinator.com/item?id=48660711 Points: 347
Comments: 55
Section titled “Comments: 55”OpenAI and Broadcom unveil LLM-optimized inference chip
Section titled “OpenAI and Broadcom unveil LLM-optimized inference chip”Article URL: https://openai.com/index/openai-broadcom-jalapeno-inference-chip/ Comments URL: https://news.ycombinator.com/item?id=48659257 Points: 141
Comments: 1
Section titled “Comments: 1”Reid Hoffman says SpaceX ‘not an AI company’, xAI ‘complete train wreck’
Section titled “Reid Hoffman says SpaceX ‘not an AI company’, xAI ‘complete train wreck’”Article URL: https://fortune.com/2026/06/24/reid-hoffman-spacex-musk-openai-anthropic-gen-z-mistake/ Comments URL: https://news.ycombinator.com/item?id=48658647 Points: 222
Comments: 255
Section titled “Comments: 255”Haystack: Open-Source AI Framework for Production Ready Agents, RAG
Section titled “Haystack: Open-Source AI Framework for Production Ready Agents, RAG”Article URL: https://haystack.deepset.ai/ Comments URL: https://news.ycombinator.com/item?id=48658095 Points: 86
Comments: 21
Section titled “Comments: 21”How to burst the AI bubble: Strike at its roots
Section titled “How to burst the AI bubble: Strike at its roots”Article URL: https://arstechnica.com/gadgets/2026/06/how-to-burst-the-ai-bubble-strike-at-its-roots/ Comments URL: https://news.ycombinator.com/item?id=48657518 Points: 62