Skip to content

issues Search Results · language:Dune language:Python language:JavaScript language:JavaScript language:Java language:Java

Filter by

55M results  (795 ms)

55M results

oyatai-0331/AInews
事実・基礎: LoRA: Low-Rank Adaptation of Large Language Models

テーマ: AI news and lab-focused LLM research: model/platform releases, open-weight models, AI agents/dev tools, RAG, efficient fine-tuning/inference, LLM evaluation, safety/regulation/business adoption 日付: ...
ai-news
auto-research
domain:research
llm-research
related-work
  • github-actions[bot]
  • Opened 
    23 minutes ago
  • #209

oyatai-0331/AInews
事実・基礎: Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks

テーマ: AI news and lab-focused LLM research: model/platform releases, open-weight models, AI agents/dev tools, RAG, efficient fine-tuning/inference, LLM evaluation, safety/regulation/business adoption 日付: ...
ai-news
auto-research
domain:research
llm-research
related-work
  • github-actions[bot]
  • Opened 
    23 minutes ago
  • #207

oyatai-0331/AInews
事実・基礎: ReAct: Synergizing Reasoning and Acting in Language Models

テーマ: AI news and lab-focused LLM research: model/platform releases, open-weight models, AI agents/dev tools, RAG, efficient fine-tuning/inference, LLM evaluation, safety/regulation/business adoption 日付: ...
ai-news
auto-research
domain:research
llm-research
related-work
  • github-actions[bot]
  • Opened 
    23 minutes ago
  • #208

open-gsd/gsd-core
enhance(execute): isolated-executor scope-overreach / rejected worktree must fail safe — never default recovery to editing main (split from #1265)

This was generated by AI during triage. Split from #1265 (second observation by @electroheadfx). #1265 tracks the original stale-HEAD worktree-base bug; this issue tracks the distinct safety concern surfaced ...
approved-enhancement
area: workflow
enhancement
needs-triage
ready-for-agent
  • trek-e
  • 1
  • Opened 
    23 minutes ago
  • #1292

oyatai-0331/AInews
事実・基礎: LLaMA: Open and Efficient Foundation Language Models

テーマ: AI news and lab-focused LLM research: model/platform releases, open-weight models, AI agents/dev tools, RAG, efficient fine-tuning/inference, LLM evaluation, safety/regulation/business adoption 日付: ...
ai-news
auto-research
domain:research
llm-research
related-work
  • github-actions[bot]
  • Opened 
    23 minutes ago
  • #204

oyatai-0331/AInews
事実・基礎: Community-driven AI innovation comes alive with Llama 2 (Meta blog)

テーマ: AI news and lab-focused LLM research: model/platform releases, open-weight models, AI agents/dev tools, RAG, efficient fine-tuning/inference, LLM evaluation, safety/regulation/business adoption 日付: ...
ai-news
auto-research
domain:research
llm-research
related-work
  • github-actions[bot]
  • Opened 
    23 minutes ago
  • #205

oyatai-0331/AInews
事実・基礎: Announcing Mistral 7B

テーマ: AI news and lab-focused LLM research: model/platform releases, open-weight models, AI agents/dev tools, RAG, efficient fine-tuning/inference, LLM evaluation, safety/regulation/business adoption 日付: ...
ai-news
auto-research
domain:research
llm-research
related-work
  • github-actions[bot]
  • Opened 
    23 minutes ago
  • #206

oyatai-0331/AInews
考察: RAGパイプラインにおいて、検索結果に対して明示的な出典プロビネンス（ソーススニペット＋メタデータ）をユーザーに提示すると、ユーザー主観の信頼度と自動的なファクトチェック一致率が有意に改善し、誤情報（hallucination）率が低下する。

テーマ: AI news and lab-focused LLM research: model/platform releases, open-weight models, AI agents/dev tools, RAG, efficient fine-tuning/inference, LLM evaluation, safety/regulation/business adoption 日付: ...
ai-news
auto-research
domain:research
hypothesis
llm-research
  • github-actions[bot]
  • Opened 
    23 minutes ago
  • #203

oyatai-0331/AInews
考察: LLMを用いた自動評価（LLMジャッジ）は出力スタイルやモデルバイアスに引きずられやすく、エージェント／生成タスクでの自動スコアは人間評価に比べて一様に高めに出る傾向があり、人手検証では平均で10%以上のスコア低下が観察される。

テーマ: AI news and lab-focused LLM research: model/platform releases, open-weight models, AI agents/dev tools, RAG, efficient fine-tuning/inference, LLM evaluation, safety/regulation/business adoption 日付: ...
ai-news
auto-research
domain:research
hypothesis
llm-research
  • github-actions[bot]
  • Opened 
    23 minutes ago
  • #202

oyatai-0331/AInews
考察: 分野特化の知識ベースを使ったRAG（密ベクトル検索＋生成）システムは、その分野の事実質問に対して、同サイズの非検索ベースLLMよりF1/Factualityで少なくとも10%高い性能を示す。

テーマ: AI news and lab-focused LLM research: model/platform releases, open-weight models, AI agents/dev tools, RAG, efficient fine-tuning/inference, LLM evaluation, safety/regulation/business adoption 日付: ...
ai-news
auto-research
domain:research
hypothesis
llm-research
  • github-actions[bot]
  • Opened 
    23 minutes ago
  • #200
Issue origami icon

Learn how you can use GitHub Issues to plan and track your work.

Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub Issues
ProTip! Restrict your search to the title by using the in:title qualifier.
Issue origami icon

Learn how you can use GitHub Issues to plan and track your work.

Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub Issues
ProTip! Restrict your search to the title by using the in:title qualifier.