LIVE · 05/12
agent-fw★RELlangchain-core==1.4.0langchain-core==1.4.0[langchain-releases]vscodeRELVisual Studio Code 1.120 Release NotesVisual Studio Code 1.120 Release Notes[vscode-updates]vscodeRELVisual Studio Code 1.119 Release NotesVisual Studio Code 1.119 Release Notes[vscode-updates]vscodeRELVisual Studio Code 1.118 Release NotesVisual Studio Code 1.118 Release Notes[vscode-updates]vscodeRELVisual Studio Code 1.117 Release NotesVisual Studio Code 1.117 Release Notes[vscode-updates]vscodeRELVisual Studio Code 1.116 Release NotesVisual Studio Code 1.116 Release Notes[vscode-updates]vscodeRELVisual Studio Code 1.115 Release NotesVisual Studio Code 1.115 Release Notes[vscode-updates]vscodeRELVisual Studio Code 1.114 Release NotesVisual Studio Code 1.114 Release Notes[vscode-updates]vscodeRELVisual Studio Code 1.113 Release NotesVisual Studio Code 1.113 Release Notes[vscode-updates]vscodeRELVisual Studio Code 1.112 Release NotesVisual Studio Code 1.112 Release Notes[vscode-updates]vscodeRELVisual Studio Code 1.111 Release NotesVisual Studio Code 1.111 Release Notes[vscode-updates]vscodeRELVisual Studio Code 1.110 Release NotesVisual Studio Code 1.110 Release Notes[vscode-updates]vscodeRELVisual Studio Code 1.109 Release NotesVisual Studio Code 1.109 Release Notes[vscode-updates]cursorRELCursor in Microsoft TeamsCursor in Microsoft Teams[cursor-changelog]researchFlashSVD v1.5: Making Low-Rank Transformers Inference Actually FastFlashSVD v1.5: Making Low-Rank Transformers Inference Actually Fast[arxiv-cs-ai]researchSingle-Configuration Attack Success Rate Is Not Enough: Jailbreak Evaluations Should Report Distributional Attack SuccessSingle-Configuration Attack Success Rate Is Not Enough: Jailbreak Evaluations Should Report Distributional Attack Success[arxiv-cs-ai]researchProactBench: Beyond What The User Asked ForProactBench: Beyond What The User Asked For[arxiv-cs-ai]researchRethinking Agentic Search with Pi-Serini: Is Lexical Retrieval Sufficient?Rethinking Agentic Search with Pi-Serini: Is Lexical Retrieval Sufficient?[arxiv-cs-ai]researchContext-Augmented Code Generation: How Product Context Improves AI Coding Agent Decision Compliance by 49%Context-Augmented Code Generation: How Product Context Improves AI Coding Agent Decision Compliance by 49%[arxiv-cs-ai]researchComputer Use at the Edge of the Statistical PrecipiceComputer Use at the Edge of the Statistical Precipice[arxiv-cs-ai]researchExecution Envelopes: A Shared Admission Contract for Backend AI Execution RequestsExecution Envelopes: A Shared Admission Contract for Backend AI Execution Requests[arxiv-cs-ai]researchDo not copy and paste! Rewriting strategies for code retrievalDo not copy and paste! Rewriting strategies for code retrieval[arxiv-cs-ai]researchMazocarta: A Seeded Procedural Deckbuilder for Instrumented Game DevelopmentMazocarta: A Seeded Procedural Deckbuilder for Instrumented Game Development[arxiv-cs-ai]researchWhat Software Engineering Looks Like to AI Agents? -- An Empirical Study of AI-Only Technical Discourse on MoltBookWhat Software Engineering Looks Like to AI Agents? -- An Empirical Study of AI-Only Technical Discourse on MoltBook[arxiv-cs-ai]
Today 56
Total 498
Major 23
Active sources 25/51
Updated just now
langchain-core==1.4.0
Featured · 注目 Featured

langchain-core==1.4.0 langchain-core==1.4.0

Changes since langchain-core==0.3.86 chore(infra): merge v1.4 into master ( #37350 ) chore: bump urllib3 from 2.6.3 to 2.7.0 in /libs/core ( #37329 ) fix(core): avoid eager pydantic.v1 import in @depr

release langchain-releases · 13h ago
Daily Summary

今日の更新

Today's Updates

Today 56 ▼ 81%
Yesterday 288
7-day 502
Last 7 days
34
48
26
14
36
288
56
05/06 05/07 05/08 05/09 05/10 05/11 05/12
Last 7 days article counts
DateCount
2026-05-0634
2026-05-0748
2026-05-0826
2026-05-0914
2026-05-1036
2026-05-11288
2026-05-1256
Top categories · 7d
主要な更新 Top stories 05/12 · 10 件
  1. 01 agent-fw ★ REL langchain-core==1.4.0 langchain-core==1.4.0 Changes since langchain-core==0.3.86 chore(infra): merge v1.4 into master ( #37350 ) chore: bump urllib3 from 2.6.3 to 2.7.0 in /libs/core ( #37329 ) fix(core): avoid eager pydantic.v1 import in @depr [langchain-releases]
  2. 02 vscode REL +11 Visual Studio Code 1.120 Release Notes Visual Studio Code 1.120 Release Notes Monthly release notes for VS Code 1.120. [vscode-updates]
  3. 03 cursor REL Cursor in Microsoft Teams Cursor in Microsoft Teams Cursor is now available in Microsoft Teams. [cursor-changelog]
  4. 04 research FlashSVD v1.5: Making Low-Rank Transformers Inference Actually Fast FlashSVD v1.5: Making Low-Rank Transformers Inference Actually Fast arXiv:2605.08314v1 Announce Type: cross Abstract: SVD-based Low-rank compression reduces transformer parameters and nominal FLOPs, but these savings often translate poorly into real LLM serving speedu [arxiv-cs-ai]
  5. 05 research Single-Configuration Attack Success Rate Is Not Enough: Jailbreak Evaluations Should Report Distributional Attack Success Single-Configuration Attack Success Rate Is Not Enough: Jailbreak Evaluations Should Report Distributional Attack Success arXiv:2605.09070v1 Announce Type: cross Abstract: Many jailbreak attack research papers report attack success rates for a limited number of parameter settings, even though there are many combinations [arxiv-cs-ai]
  6. 06 research A Dataset of Agentic AI Coding Tool Configurations A Dataset of Agentic AI Coding Tool Configurations arXiv:2605.08435v1 Announce Type: new Abstract: Agentic AI coding tools such as Claude Code and OpenAI Codex execute multi-step coding tasks with limited human oversight. To steer these tools, develop [arxiv-cs-se]
  7. 07 research EvidenT: An Evidence-Preserving Framework for Iterative System-Level Package Repair EvidenT: An Evidence-Preserving Framework for Iterative System-Level Package Repair arXiv:2605.08621v1 Announce Type: new Abstract: Frequent toolchain updates and growing ISA diversity have made system-level software package repair increasingly important. Diagnosing and repairing bui [arxiv-cs-se]
  8. 08 claude Claude Opus 4.7 Claude Opus 4.7 [anthropic-news]
  9. 09 claude Claude Design Anthropic Labs Claude Design Anthropic Labs [anthropic-news]
  10. 10 claude OpenAI CodexとClaude Codeの「AIコーディング支援のコスト感」の違い (no English title) はじめに OpenAI Codex と Claude Code を両方使っていると、単純な「どちらが賢いか」とは別に、かなり現実的な差が見えてきます。 それが コスト感 です。 ここでいうコストは、月額料金だけではありません。 どれくらい作 [qiita-claude]
🔥 Today's Top 3 importance × recency
  1. langchain-core==1.4.0 langchain-core==1.4.0 langchain-releases 13h ago
  2. CodeQL 2.25.3 adds Swift 6.3 support CodeQL 2.25.3 adds Swift 6.3 support github-changelog 3d ago
  3. Visual Studio Code 1.120 Release Notes Visual Studio Code 1.120 Release Notes vscode-updates just now

Timeline 498 total · page 1/17

TODAY 30 entries
NEW blog local-llm 2h ago · qiita-llm

個人的によく使う、エンジニアの学習効率を上げるClaudeプロンプト３選

AI要約 個人的によく使う学習効率化のプロンプトを3つ、 実際のやり取り例とあわせてメモしてみます。 1. 横展開プロンプト：「〇〇という概念と同等の重要度の概念を3つ教えて」 用途 言語・フレームワーク・アーキテクチャなど、新しい概念を一つ学んだタ

#llm#qiita
qiita.com →
個人的によく使う、エンジニアの学習効率を上げるClaudeプロンプト３選 og
NEW blog tech-news 3h ago · techcrunch

Thinking Machines wants to build an AI that actually listens while it talks Thinking Machines wants to build an AI that actually listens while it talks

EN Right now, every AI model you've ever used works the same way. You talk, it listens. It responds, you listen. Thinking Machines is trying to change that by building a model that processes your input a

EN Right now, every AI model you've ever used works the same way. You talk, it listens. It responds, you listen. Thinking Machines is trying to change that by building a model that processes your input a

#news#techcrunch
techcrunch.com →
Thinking Machines wants to build an AI that actually listens while it talks og
NEW paper research 4h ago · arxiv-cs-ai

FlashSVD v1.5: Making Low-Rank Transformers Inference Actually Fast FlashSVD v1.5: Making Low-Rank Transformers Inference Actually Fast

EN arXiv:2605.08314v1 Announce Type: cross Abstract: SVD-based Low-rank compression reduces transformer parameters and nominal FLOPs, but these savings often translate poorly into real LLM serving speedu

EN arXiv:2605.08314v1 Announce Type: cross Abstract: SVD-based Low-rank compression reduces transformer parameters and nominal FLOPs, but these savings often translate poorly into real LLM serving speedu

#arxiv#paper
arxiv.org →
FlashSVD v1.5: Making Low-Rank Transformers Inference Actually Fast og
NEW paper research 4h ago · arxiv-cs-ai

Single-Configuration Attack Success Rate Is Not Enough: Jailbreak Evaluations Should Report Distributional Attack Success Single-Configuration Attack Success Rate Is Not Enough: Jailbreak Evaluations Should Report Distributional Attack Success

EN arXiv:2605.09070v1 Announce Type: cross Abstract: Many jailbreak attack research papers report attack success rates for a limited number of parameter settings, even though there are many combinations

EN arXiv:2605.09070v1 Announce Type: cross Abstract: Many jailbreak attack research papers report attack success rates for a limited number of parameter settings, even though there are many combinations

#arxiv#paper
arxiv.org →
Single-Configuration Attack Success Rate Is Not Enough: Jailbreak Evaluations Should Report Distributional Attack Success og
NEW paper research 4h ago · arxiv-cs-ai

ProactBench: Beyond What The User Asked For ProactBench: Beyond What The User Asked For

EN arXiv:2605.09228v1 Announce Type: cross Abstract: Most LLM benchmarks score how well a model responds to explicit requests. They leave unmeasured a different conversational ability: noticing and actin

EN arXiv:2605.09228v1 Announce Type: cross Abstract: Most LLM benchmarks score how well a model responds to explicit requests. They leave unmeasured a different conversational ability: noticing and actin

#arxiv#paper
arxiv.org →
ProactBench: Beyond What The User Asked For og
healthy harness v0.1 powered by Astro + Claude Opus Cloudflare Pages