ai Trend Report
Dashboard へ戻る
3
High impact
18
Mid impact
379
Signal watch
なぜこのサイトを作ったのか
私たちそれぞれが個別にAIを使って情報収集し、同じような3行要約を作るたびに、世界中で膨大な電力と計算リソースが消費されています。 本プロジェクトは、あらかじめ広範な情報を取得・集約しておくことで、個別のAI実行回数を減らし、地球環境（GPU/TPU負荷）に配慮した効率的な情報収集を目指す実験的なダッシュボードです。
#AIタグ
【毎日プロンプト#161】幽霊もタジタジ！？最恐の怪談を爆笑コメディに変える「お笑いツッコミ生成プロンプト」
・怖い怪談やホラーでも、現実的なツッコミを入れることで笑いに変えられるという発想を紹介
・ホラーが苦手な人でも楽しめるように、恐怖をコメディへ変換するプロンプトが主題
・本文では具体的なプロンプトや活用例の続きが案内されている
Action: 怪談文を入力すると、恐怖要素を残しつつ現実的なツッコミでコメディ化するプロンプトテンプレートを作成し、複数パターンを比較検証する。
#AIタグ
2026年7月21日に東京証券取引所から公表された最新の**「コーポレートガバナンス・コード（CGコード）」**の内容を、初心者の方でもわかるようにNotebooklmで整理した
・東京証券取引所が2026年7月21日に公表した最新のコーポレートガバナンス・コードを初心者向けに整理した内容
・筆者が自分用のメモとしてNoteにまとめた短い紹介記事
・NotebookLMを使って情報をわかりやすく整理したことが主題
Action: NotebookLMなどで制度・ガイドライン文書を要約する際は、原文リンクと更新日をメタデータとして保存し、差分確認できるノート構成を作る。
#AIタグ
R_581 問題分析にAIを
・AIによって受験勉強の進め方が変わる可能性がある
・変化の方向性にはさまざまなパターンがある
・AIを活用して変革するには、まず全体像の理解が必要
Action: 教育分野でのAI活用案を整理し、学習プロセスのどこを支援できるかをユースケースとして洗い出す
#AIタグ
【自己紹介】AI副業雑記はじめます【noteは人力】
・筆者は「ふたばまめ」として自己紹介している
・AIだけで作ったブログを始めたことをきっかけにnoteで記録を残す方針を述べている
・記事本文は導入のみで、詳細は続きに委ねられている
Action: AIで生成したブログ運用の記録項目（使用ツール、生成フロー、工数、品質課題）をテンプレート化して継続的に検証できるようにする
#AIタグ
Geminiを「自分の分身」にする方法！ 宇宙兄弟のFFS理論でAIに自分の魂を吹き込もう！【プロンプト付き】
・AIの回答が意図とズレる主因は、AIが利用者の性格や理念を十分に知らないことだと述べている
・Geminiを「自分の分身」や司令塔として機能させるために、個人の特性や教育観を反映させる考え方を紹介している
・宇宙兄弟のFFS理論とプロンプト活用を通じて、AIに自分らしい判断軸を持たせる方法を扱う内容と考えられる
Action: Gemini向けに、自分の価値観・判断基準・口調・禁止事項を整理したペルソナ定義プロンプトを作成し、定型テンプレートとして再利用できるようにする
#AIタグ
「セルフレビューさせて終わり」で本当に大丈夫か——AIエージェント時代に効く『敵対的検証』という一段深い仕組み
・同じAIエージェントに実装とレビューの両方を任せる運用には限界があるという問題提起
・AIの自己レビューだけでは見逃しや思考の偏りが残る可能性がある
・AIエージェント活用では、別視点から検証する『敵対的検証』の仕組みが重要になる
Action: 実装AIとは別モデル・別プロンプトの検証AIを用意し、レビュー観点を分離した二段階チェックフローを設計する
Hugging Face Papers
SWE-Pruner Pro：コードLLMはすでに何を刈り込むべきかを知っている
・記事タイトルのみが提示されており、本文情報は確認できない
・タイトルから、コード生成LLMによる不要部分の選別・削減手法に関する内容と推測される
・具体的な手法、評価結果、適用対象は本文不足のため不明
Action: 本文を確認し、提案手法の前提・評価指標・既存開発フローへの適用可否を整理してください。
Hugging Face Papers
TimeLens2: マルチモーダルLLMによる汎用動画時間的グラウンディング
・TimeLens2は、マルチモーダルLLMを用いた汎用的な動画の時間的グラウンディングを扱う研究とみられる
・与えられた内容はタイトルのみで、手法・評価・適用範囲などの詳細は確認できない
・動画理解、時系列区間の特定、自然言語と映像の対応付けに関心のある分野に関連する
Action: 論文本文を確認し、入力形式・対応タスク・ベンチマーク・推論コストを整理して既存の動画理解パイプラインに適用可能か評価する
Hugging Face Papers
EvolvingWorld：対話型文学世界においてロールプレイエージェントと世界モデルを共進化させるオープンスキーマフレームワーク
・対話型の文学世界を対象に、ロールプレイエージェントと世界モデルを同時に進化させる「EvolvingWorld」を提案している。
・オープンスキーマにより、固定的な構造に縛られず世界設定や登場人物の振る舞いを柔軟に拡張できることが主題と考えられる。
・提示情報がタイトルのみのため詳細な手法・評価結果は不明だが、AIエージェントによる物語生成やインタラクティブ体験への応用が想定される。
Action: ロールプレイAIのPoCとして、キャラクター状態と世界状態を別管理するオープンスキーマ設計を試作し、対話ログから両者を更新する仕組みを検証する。
Hugging Face Papers
HOMIE: マルチモーダル知的拡張による人間・物体中心の動画パーソナライゼーション
・HOMIEは、人間と物体の関係性に着目した動画パーソナライゼーション手法を扱う内容です。
・マルチモーダルな情報を活用して、動画の理解や強調表現を高度化することが主題と考えられます。
・本文情報がほぼ存在しないため、具体的な手法・評価・適用例の詳細は判断できません。
Action: 論文本文やプロジェクトページを確認し、入力モダリティ・モデル構成・評価指標を整理して既存の動画編集/推薦パイプラインに適用可能か検証する。
Hugging Face Papers
Apple-π：法則に基づく身体知能に向けた、動画を用いた思考のベンチマーク
・Apple-πは、動画を使って物理世界に関する推論・思考能力を評価するベンチマークを目指す内容です。
・主題は、法則に基づく身体知能（physical intelligence）を測るために、映像理解と推論を結びつける点にあります。
・本文情報が少ないため詳細な手法や結果は不明ですが、AIの動画理解と物理推論評価に関する研究トピックと考えられます。
Action: 動画理解モデルに対して、物理法則や因果関係を問う評価タスクを既存データセット上で試作し、現行モデルの推論限界を確認する。
Hugging Face Papers
ReflectWorld-MM: オープンエンドな動画ストリームのためのエンティティ指向マルチモーダルメモリシステム
・ReflectWorld-MMは、長時間・継続的な動画ストリームを対象にしたエンティティ指向のマルチモーダル記憶システムを扱う内容と考えられる
・映像内の対象や出来事を記憶・参照しながら、オープンエンドな動画理解を支援するアプローチが主題である
・本文情報がほぼタイトルのみのため、具体的な手法・評価・応用範囲の詳細は不明
Action: 論文本文を確認し、エンティティ表現・記憶更新方式・検索方法を整理して、動画理解パイプラインへ組み込めるか検討する
Hugging Face Papers
グループエントロピー制御付き方策最適化
・グループエントロピーを制御しながら方策最適化を行う手法を示すタイトル情報のみがある
・詳細な本文がなく、目的・手法・実験結果などの具体的内容は不明
・強化学習や方策最適化に関連するAI研究トピックと推測される
Action: 本文や論文概要を確認し、既存のPPO系手法との違いとエントロピー制御の実装ポイントを整理する
Hugging Face Papers
RynnBrain 1.1：より高い能力と汎化性を目指す身体性基盤モデル
- RynnBrain 1.1 は、より高い能力と汎化性能を目指した embodied foundation model に関する内容です。
- 提供本文にはタイトル以外の詳細がなく、具体的な技術改善点や評価結果は確認できません。
- ロボティクスや身体性AIの基盤モデル動向を追う上で、概要把握レベルの情報です。
Action: 本文や公開ページを確認し、モデルの改良点・ベンチマーク・対応タスクを整理して既存の embodied AI 手法と比較してください。
Hugging Face Papers
API呼び出しエージェントのための環境不要な合成データ生成
- API呼び出しエージェント向けの合成データ生成を扱う内容です。
- 実行環境に依存しない「environment-free」な手法が主題です。
- 本文情報がほぼないため、具体的な手法・効果・適用例は不明です。
Action: 本文や概要を確認し、対象API・生成データ形式・評価方法を整理して自分のエージェント開発に適用できるか検討する。
Hugging Face Papers
コーチとしてのLLM：検証不可能なタスクのための体験学習
- LLMを「コーチ」として活用し、正解を自動検証しにくいタスクに対する学習支援を扱う内容。
- 体験学習を通じて、対話や振り返りをベースにスキル向上を促すアプローチが主題。
- 記事本文の詳細が不足しているため、具体的な手法・評価方法・適用事例までは判断できない。
Action: 自分の業務で正解判定が難しいタスクを1つ選び、LLMにコーチ役をさせるプロンプトを作って振り返り対話の流れを試作する。
Hugging Face Papers
DiffGI: 高忠実度な薄肉シェル3D生成のための微分可能ジオメトリ画像
・薄肉シェル形状の高忠実度な3D生成を目的とした「DiffGI」を扱う内容
・微分可能なジオメトリ画像表現により、3D形状生成の最適化や学習をしやすくするアプローチと考えられる
・本文情報がほぼないため、具体的な手法・性能・適用例の確認が必要
Action: 論文本文やプロジェクトページを確認し、入力表現・学習方式・既存3D生成手法との差分を整理する。
Hugging Face Papers
LLMのポストトレーニングのための蒸留強化学習
・LLMのポストトレーニングにおける「Distilled Reinforcement Learning」がテーマ
・本文内容がタイトルのみで、手法・効果・適用例の詳細は確認できない
・強化学習と知識蒸留を組み合わせた学習効率化や性能改善の文脈が示唆される
Action: LLMのポストトレーニング手法として、強化学習と蒸留を組み合わせた既存研究を調査し、自分のモデル評価基盤で比較実験できるようにする。
Hugging Face Papers
JoyNexus: VLAモデル向けサービス指向マルチテナント事後学習
・JoyNexusはVLAモデル向けのサービス指向マルチテナント事後学習に関する内容。
・本文情報がタイトルのみのため、具体的な手法・評価・適用事例は判断できない。
・VLAモデルの運用効率化や複数利用者向け学習基盤に関するテーマである可能性が高い。
Action: 元論文や本文を確認し、VLAの対象タスク・学習方式・マルチテナント設計の要点を整理して社内適用可否を評価する。
Hugging Face Papers
セルフホスト型AIエージェントに対する自己状態攻撃：OS防御はどこまで有効か？
・セルフホスト型AIエージェントに対する「自己状態攻撃」とOS防御の限界を扱う内容です。
・提示された本文がタイトルのみのため、具体的な攻撃手法や検証結果までは判断できません。
・AIエージェントのローカル実行環境における安全性や隔離設計の重要性が示唆されます。
Action: AIエージェントをセルフホストする場合は、OS権限の最小化、コンテナ/VMによる分離、状態保存領域のアクセス制御をまず点検してください。
Hugging Face Papers
FlowMimic: オンライン動画編集データ生成とモダリティ模倣のための、ピクセルペア歪みフローフィールドによるマスク不要の視覚編集・生成
・FlowMimicという、マスク不要で画像・映像の視覚編集と生成を行う手法を扱う内容です。
・ピクセルペアを用いたWarped Flow Fieldにより、オンライン動画編集用データ生成やモダリティ模倣を目指しています。
・本文情報がほぼタイトルのみのため、具体的な手法詳細・性能・適用例までは判断できません。
Action: 論文本文や公開実装を確認し、既存の動画編集・画像変換パイプラインに組み込めるか、マスク不要編集の精度と計算コストを比較評価してください。
Hugging Face Papers
OpenLongTail: ロングテール運転データの生成的スケーリング
・ロングテールな運転データを生成的に拡張することをテーマにした内容。
・提示された本文情報がタイトルのみのため、手法・評価・適用範囲の詳細は不明。
・自動運転や運転データ拡張に関わるAI研究トピックとして分類できる。
Action: 論文本文や公開リポジトリを確認し、生成対象のシナリオ、使用モデル、評価指標を整理して既存の運転データ拡張手法と比較してください。
Hugging Face Papers
分布シフト下における忠実な生成のためのトークンレベル・オフポリシー学習
・分布シフトがある状況でも、生成結果の忠実性を高めるためのトークン単位のオフポリシー学習を扱う研究と考えられる
・学習時と推論時の分布差によって生じる生成品質や事実整合性の低下に対処することが主題
・詳細本文がないため手法や実験結果は不明だが、LLMの学習・評価・信頼性向上に関わるテーマ
Action: 自分の生成AIパイプラインで、トークン単位の報酬設計やオフポリシー評価が適用できる箇所を洗い出し、分布シフト時の忠実性指標を追加で計測する。
Hugging Face Papers
FlashRT：エージェントがリアルタイム・マルチモーダル・アプリケーションをデプロイするためのガイド用エージェントハーネス
・FlashRTは、エージェントによるリアルタイムなマルチモーダルアプリケーションのデプロイを支援するための仕組みと考えられる
・提示内容はタイトルのみで、アーキテクチャ、機能、性能、利用例などの詳細は不明
・関心領域としてはAIエージェント、リアルタイム処理、マルチモーダル入力/出力の実運用支援に属する
Action: 公開情報が少ないため、まずは論文やリポジトリの詳細を確認し、自社のリアルタイム音声・画像処理基盤に適用可能か評価する
Hugging Face Papers
DiFA: 拡散モデルのための推論時フォワードプロセス整合
・拡散モデルにおける推論時のフォワードプロセス整合（DiFA）を扱う内容。
・提供された本文がタイトルのみのため、手法の詳細、効果、適用例は判断できない。
・推論時のアライメント改善によって生成品質や安定性向上を狙う研究テーマと考えられる。
Action: 論文本文や要約を確認し、既存の拡散モデル推論パイプラインにどのように組み込めるかを調査する。
#AIタグ
【地域×AIを用いた学校祭】組織的DXと生徒の主体性が紡ぐ、持続可能な地域共創リノベーションの全貌
・学校祭を単なる校内行事ではなく、地域共創のプラットフォームとして再定義する導入内容
・AIや組織的DX、生徒の主体性を組み合わせた持続可能な運営・改革が主題
・本文は冒頭のみのため、具体的な実践手法や技術構成の詳細は未確認
Action: 学校行事や地域連携向けに、AI活用の企画管理・情報共有フローを整理した小規模な実証案を作成する
#AIタグ
「AI動画はもう限界かもしれない」そう思っていた僕が、Seedance 2.5で少しワクワクした話
・筆者は最近、AI動画に少し疲れを感じていた
・その中でSeedance 2.5に触れ、再び興味や期待を持った様子
・AI動画生成の限界感と、新しいモデルによる変化が主題と考えられる
Action: Seedance 2.5の特徴や既存AI動画生成モデルとの差分を整理し、評価用の比較観点（画質・操作性・生成速度・一貫性）を作成する
#AIタグ
【昼休み10分で月5万】おにぎりを食べながら、AIに「書いて」と言うだけ。午後の会議が終わる頃、記事ができている生活
【昼休み10分で月5万】おにぎりを食べながら、AIに「書いて」と言うだけ。午後の会議が終わる頃、記事ができている生活
#AIタグ
AIの「もっともらしい嘘」をハックする
・AIのハルシネーション（もっともらしい嘘）を前提に扱う重要性を述べている
・検証を防御設計として捉え、AI出力をそのまま信用しない姿勢を示している
・AI活用では生成性能だけでなく、確認・検証の仕組み作りが重要である
Action: AI出力を利用する処理に対して、根拠提示・再確認・人手レビューを組み合わせた検証フローを実装する
#AIタグ
一人ひとりが顧客の課題解決に尽力する文化「自分はずっと、こんな環境で働きたかったのかもしれない」と思えた
・SUPER STUDIOのシニアマネージャー山口昂治氏の経歴と役割が紹介されている
・アクセンチュアで基幹系システムの運用保守やSI、大規模案件のPL・PMを経験してきた
・現在は顧客課題の解決に加え、社内ルール整備など組織づくりにも関わっている
Action: 顧客課題の解決に必要なロールや開発手法の経験を棚卸しし、自チームの運用改善やルール整備の提案を1つ作成する
#AIタグ
桃太郎が腰を痛めたので、AIに「おばあさんの鬼退治」を書かせてみた
・桃太郎が腰を痛めたという設定で、おばあさんが代わりに鬼退治へ行く物語をAIに書かせる内容
・従来の昔話を別視点・別展開で再構成する発想が中心
・記事本文は導入のみで、なぜおばあさんが行くことになったのかという続きへの誘導になっている
Action: 昔話や既存ストーリーを題材に、役割置換や設定変更のプロンプトテンプレートを作って生成AIの出力差分を比較する。
#AIタグ
AIは「5000年ぶりの組織革命」なのか――コミュニケーションが可視化される時代のマネジメント変革
・AIは単なる業務効率化ツールではなく、組織運営やマネジメントのあり方を変える可能性がある
・コミュニケーションの可視化が進むことで、意思決定や組織構造の見直しが求められる
・技術導入だけでなく、組織設計やマネジメント手法の変革が重要なテーマとなる
Action: 社内コミュニケーションや意思決定フローを棚卸しし、AIで可視化・分析できる業務データの収集方針を設計する
#AIタグ
「AIって難しそう」その不安、実は9割が誤解です
・AIはエンジニア向けだけのものではなく、スマホと日本語だけでも使い始められると説明している
・初心者が挫折する主因は能力ではなく、専門用語中心で初心者向けでない情報環境にあると述べている
・筆者は初心者・シニア・個人事業主向けに、ChatGPTレッスンや相談、教材、継続コミュニティなどの支援サービスを提供している
Action: 非エンジニア向けに、専門用語を避けたChatGPT導入ガイドや初回オンボーディング手順を日本語で作成する。
#AIタグ
発想力がなくても、AIでビジネスは作れる「ずらし」は才能ではなく、翻訳できる技術である
・発想力や才能がなくても、AIを活用してビジネスを生み出せるという主張
・「ずらし」は属人的なひらめきではなく、翻訳可能な技術として扱える
・AIによってアイデア生成や価値提案の再構成がしやすくなる可能性がある
Action: 既存サービスの対象顧客・用途・提供価値を3軸で分解し、AIに「ずらし案」を10個生成させて新規プロトタイプ候補を整理する
#AIタグ
AIで変わる工場DX
・AIそのものではなく、AIを使いこなす競争相手が仕事を変えると訴えている
・変化はすでに静かに始まっており、気づいたときには競争ルールが変わっている可能性がある
・現状に不安を感じる人に向けて、時代変化への備えを促す導入文になっている
Action: 自社の工場業務を棚卸しし、AI導入で競争優位になり得る工程を1つ選んでPoC案を作成する
#AIタグ
中道改革連合・小川淳也代表、大事な『政権ビジョン』をAI作成のまま発表してしまう→「責任とれないから引用やめて」と自ら暴露ｗｗ
・中道改革連合が「競争力ある福祉国家」という政権ビジョン素案を掲げた内容が話題化
・記事タイトル上では、代表がAI作成文をそのまま発表したことや引用を控えるよう述べた点が問題視されている
・本文情報は限定的で、政治的話題を中心にAI利用の扱いと説明責任が論点になっている
Action: AI生成文を公開資料に使う場合は、生成物のレビュー責任者を明確化し、出典・検証フロー・注意書きを実装できる承認プロセスを整備する。
#AIタグ
不動産価値分析デューデリジェンス自動化事業 仕様書｜「ずらしの技術」で稼ぐビジネスモデル＃1
・不動産価値分析とデューデリジェンスの自動化事業に関する仕様書を示すタイトル情報のみがある
・「ずらしの技術」を活用したビジネスモデルの第1弾として位置付けられている
・本文情報が不足しており、具体的な機能・技術構成・実装要件は読み取れない
Action: 本文が不足しているため、記事全文を取得し、自動化対象業務・必要データ・分析ロジックを要件として整理してください
#AIタグ
経験はどう育つか——泥も余白も摩擦も、なぜ同じ話なのか
・知識は言葉で伝えられるが、経験は本人の身体を通じてしか得られないという前提を扱っている
・前回内容を踏まえ、本書の土台となる考え方を継続して掘り下げる構成になっている
・経験の形成における過程や感覚的要素の重要性を示唆する内容である
Action: 知識共有だけでなく、実際に手を動かして学べる小さな実験やハンズオンを設計する。
#AIタグ
AIに7人の登場人物を作って、事業をディベートさせてみた
・著者は業務効率化ではClaude Code、画像生成ではChatGPTを使っている
・最もよく使う用途は、経営判断時にAIへ複数キャラクターを設定して議論させること
・記事はその活用方法の導入部分で、詳細は続きに委ねられている
Action: 事業判断や設計レビュー用に、異なる立場のAIペルソナを複数用意して論点を議論させるプロンプトを作成する
#AIタグ
【2026年版】AI×Kindle出版のペンネーム戦略｜売れる著者名の決め方を完全解説
・AI×Kindle出版におけるペンネーム戦略を解説する動画の紹介
・売れる著者名の決め方が主題
・本文情報は少なく、詳細は動画視聴が前提
Action: AI関連の出版支援サービスを開発するなら、ペンネーム候補の生成・比較・商標チェック導線を設計する。
#AIタグ
スクショに矢印をつける仕事、AIにぜんぶ渡しました——300ブックマーク超えたガイド生成の裏側
・マニュアルや教材作成で発生するスクリーンショットへの矢印付け作業をAIに任せた話
・300ブックマーク超えのガイド生成の裏側について触れている
・本文が短く、具体的な手法や使用技術の詳細はこの抜粋だけでは不明
Action: スクリーンショット注釈作業の自動化候補として、画像認識や生成AIを使った矢印・説明文付与のワークフローを試作してください。
#AIタグ
「セミナーの話が右から左に抜ける」僕が、聞く力を取り戻すためにやったこと
・地域コミュニティの集まりとAIセミナーに参加したが、聞いていたはずの内容を十分に記憶できなかった
・うなずきやメモをしていても、帰宅後に内容の半分ほどしか思い出せないという課題を感じた
・セミナー内容の定着や「聞く力」を取り戻すための工夫や振り返りが主題と考えられる
Action: セミナー参加時は、要点を3つに絞ってリアルタイム記録し、終了後5分以内に要約メモを作る運用を試してください。
#LLMタグ
モーツァルトは楽器ごとに「音の動かし方」を変えている――交響曲第39番をMEIで音程分析してみた
・モーツァルトの交響曲第39番第1楽章をMEI形式の楽譜データから楽器別に分析している
・音程分布をヒストグラム化し、楽器ごとの音の進み方の違いを可視化している
・楽器ごとの役割や作曲上の書き分けをデータ分析で捉える試みである
Action: MEI楽譜を読み込んでパート別の音程差分を集計し、ヒストグラムを自動生成する分析スクリプトを作成する
LLMタグが付けられた新着記事 - Qiita
「GPUは4分の1、でも学習は終わらない」NVIDIAが宣言したポストトレーニング中心のAI経済
・NVIDIAはAI計算需要の中心が事前学習からポストトレーニングへ移ったと表明
・2026年7月17日の公式ブログで、その変化をNemotron 3 Ultraの事例で説明
・大規模モデル開発では、継続的な学習・調整工程の重要性が増していることを示唆
Action: 事前学習だけでなく、継続学習・ファインチューニング・評価を含むポストトレーニング前提でGPU予算とMLパイプラインを見直す。
#LLMタグ
Claude Fableを使う人は増える。Claude Fableに"使われる人"も増える。
・Claude Fableの記事は、優れたベンチマークや高い性能の話題から始まることが多い
・数日間の自律作業やコーディング能力向上、長時間の知的作業への強さが強調されている
・本文は途中までだが、Claude Fableの利便性と人間側への影響を示唆する導入になっている
Action: Claude Fableのような自律型AIを使う場合は、タスク委譲範囲・レビュー手順・停止条件を先に定義して運用してください。
LLMタグが付けられた新着記事 - Qiita
自作LLM評価ハーネスの実装解説——正解の独立検算・LLM-as-judgeの機械採点・Ollama num_ctxの実測対策
・自作のLLM比較評価で候補モデルが満点化し、差が出ない問題を扱っている
・判別力のある評価テストを作るための全体像と実装上の工夫を紹介している
・独立検算、LLM-as-judgeによる機械採点、Ollamaのnum_ctx実測対策が主題である
Action: 自社ユースケースに合わせて、独立検算付きの設問とLLM-as-judge採点を組み込んだ評価ハーネスを作成し、候補モデルの再比較を行う
#LLMタグ
ClaudeCodeから、Fable5 / GPT-5.6 / Grok4.5 / Kimi-K3を同時に使おう！
・ClaudeCodeから複数の最新LLMを同時に使うというテーマの記事
・Fable 5、GPT-5.6 Sol、Grok4.5、Kimi K3の比較や選び方への関心を扱っている
・本文は抜粋のみで詳細不明だが、用途に応じたモデル選定や併用が主題と考えられる
Action: ClaudeCode上で各モデルの得意分野を整理し、同じプロンプトで比較評価できる検証フローを作る
ITmedia NEWS 最新記事一覧
XのAndroidアプリ、1年かけ“過去最大級の刷新”も賛否 「UI変わりすぎ」「スペース使えない」の声も
XのAndroidアプリ、1年かけ“過去最大級の刷新”も賛否 「UI変わりすぎ」「スペース使えない」の声も
LLMタグが付けられた新着記事 - Qiita
DifyのParameter Extractorは思い通りに動かない — ヒアリングボット実装で踏んだクセと制御パターン
DifyのParameter Extractorは思い通りに動かない — ヒアリングボット実装で踏んだクセと制御パターン
#LLMタグ
【速報解説】Kimi K3 ── 2.8兆パラメータで世界最大、フロントエンド評価でFable 5を超えた
【速報解説】Kimi K3 ── 2.8兆パラメータで世界最大、フロントエンド評価でFable 5を超えた
#LLMタグ
Future of Work & Pray’s 未来人に聞く『AI時代の人間の可能性、"まだ、ここにない、出会い。"』～前編：カオス理論および数理脳科学者 津田一郎さんに聞く、カオス的労働観
Future of Work & Pray’s 未来人に聞く『AI時代の人間の可能性、"まだ、ここにない、出会い。"』～前編：カオス理論および数理脳科学者 津田一郎さんに聞く、カオス的労働観
機械学習タグが付けられた新着記事 - Qiita
RAGパイプライン全体像 — 取り込みから回答生成までの流れを初心者向けに整理する
RAGパイプライン全体像 — 取り込みから回答生成までの流れを初心者向けに整理する
LLMタグが付けられた新着記事 - Qiita
RAGパイプライン全体像 — 取り込みから回答生成までの流れを初心者向けに整理する
RAGパイプライン全体像 — 取り込みから回答生成までの流れを初心者向けに整理する
機械学習タグが付けられた新着記事 - Qiita
PyTorch 2.13のMPS FlexAttentionをM1 Maxで検証――疎な注意で最大7.83倍
PyTorch 2.13のMPS FlexAttentionをM1 Maxで検証――疎な注意で最大7.83倍
cs.LG updates on arXiv.org
Reinforcement Learning-Guided NSGA-II Enhanced with Gray Relational Coefficient for Multi-Objective Optimization: Application to NASDAQ Portfolio Optimization
Reinforcement Learning-Guided NSGA-II Enhanced with Gray Relational Coefficient for Multi-Objective Optimization: Application to NASDAQ Portfolio Optimization
cs.LG updates on arXiv.org
DocOCR-Eval: A Correction-Based Framework for OCR Tool Selection Without Ground Truth
DocOCR-Eval: A Correction-Based Framework for OCR Tool Selection Without Ground Truth
cs.LG updates on arXiv.org
Fully-sensorized smart-eyewear platform for on-device Machine Learning
Fully-sensorized smart-eyewear platform for on-device Machine Learning
cs.LG updates on arXiv.org
LLM Unlearning for Cyber Defense: A Survey on Methods, Challenges, and Emerging Threats
LLM Unlearning for Cyber Defense: A Survey on Methods, Challenges, and Emerging Threats
cs.LG updates on arXiv.org
Operator-Aware Mixed-Precision Tolerance Calibration for Tensor Kernels
Operator-Aware Mixed-Precision Tolerance Calibration for Tensor Kernels
cs.LG updates on arXiv.org
RouteCost: A Production-Inspired Multi-Stage Framework for Pre-Order Shipping Cost Estimation in E-Commerce
RouteCost: A Production-Inspired Multi-Stage Framework for Pre-Order Shipping Cost Estimation in E-Commerce
cs.LG updates on arXiv.org
Orthogonal Gradient Constraints Shape Noisy-Label Memorization Dynamics
Orthogonal Gradient Constraints Shape Noisy-Label Memorization Dynamics
cs.LG updates on arXiv.org
From Weights to Words: Expressing and Editing Preference Model Inferences in Natural Language
From Weights to Words: Expressing and Editing Preference Model Inferences in Natural Language
cs.LG updates on arXiv.org
Token-Level Cross-Modal Transformer with Contrastive Multi-Task Learning for Breast Cancer Subtype Classification and Survival Prediction
Token-Level Cross-Modal Transformer with Contrastive Multi-Task Learning for Breast Cancer Subtype Classification and Survival Prediction
cs.LG updates on arXiv.org
HantaWatch: Federated Learning for Hantavirus Genomic Surveillance
HantaWatch: Federated Learning for Hantavirus Genomic Surveillance
cs.LG updates on arXiv.org
OpenMHC: Accelerating the Science of Wearable Foundation Models
OpenMHC: Accelerating the Science of Wearable Foundation Models
cs.LG updates on arXiv.org
The Failures of Marginal Influence-Based Attribution Methods for Global Time Series Explanations
The Failures of Marginal Influence-Based Attribution Methods for Global Time Series Explanations
cs.LG updates on arXiv.org
Quantizing Recursive Reasoning Models
Quantizing Recursive Reasoning Models
cs.LG updates on arXiv.org
Diffusion-corrected Autoregressive Fourier Neural Operator for Droplet Evolution Prediction
Diffusion-corrected Autoregressive Fourier Neural Operator for Droplet Evolution Prediction
cs.LG updates on arXiv.org
BACON: Budgeted Human Calibration for Modeling and Evaluation with Multiple AI Judges
BACON: Budgeted Human Calibration for Modeling and Evaluation with Multiple AI Judges
cs.LG updates on arXiv.org
Normalized Rewards for Preference Optimization
Normalized Rewards for Preference Optimization
cs.LG updates on arXiv.org
KernelBench-Verified: Do LLM-Generated Kernels Actually Beat PyTorch?
KernelBench-Verified: Do LLM-Generated Kernels Actually Beat PyTorch?
cs.LG updates on arXiv.org
TRACE: Trajectory-Based Safety Patch Learning for LLM Post-Training Realignment
TRACE: Trajectory-Based Safety Patch Learning for LLM Post-Training Realignment
cs.LG updates on arXiv.org
RobustMAD: Evaluating Real-World Robustness of Multimodal Small Language Models for Deployable Anomaly Detection Assistants
RobustMAD: Evaluating Real-World Robustness of Multimodal Small Language Models for Deployable Anomaly Detection Assistants
cs.LG updates on arXiv.org
CIGPO: Contextual Information-Gain Policy Optimization for Multi-Turn Evidence-Reading LLM Agents
CIGPO: Contextual Information-Gain Policy Optimization for Multi-Turn Evidence-Reading LLM Agents
cs.LG updates on arXiv.org
Learning Structural Manipulability in Gate-Level Netlists Using Graph Neural Networks
Learning Structural Manipulability in Gate-Level Netlists Using Graph Neural Networks
cs.LG updates on arXiv.org
Let the Data Decide: Supervision Analysis, Capability Trade-offs, and Adaptive Objective Routing in Continued Pre-Training via Off-Policy Distillation
Let the Data Decide: Supervision Analysis, Capability Trade-offs, and Adaptive Objective Routing in Continued Pre-Training via Off-Policy Distillation
cs.LG updates on arXiv.org
Self-Evolving Just-In-Time Memory for Proactive Embodied Safety
Self-Evolving Just-In-Time Memory for Proactive Embodied Safety
cs.LG updates on arXiv.org
High-accuracy Low-Bit KV-Cache Quantization via Local Distribution Restoration
High-accuracy Low-Bit KV-Cache Quantization via Local Distribution Restoration
cs.LG updates on arXiv.org
Benchmarking Machine Learning Models for Multi-Omics-Based Breast Cancer Prediction
Benchmarking Machine Learning Models for Multi-Omics-Based Breast Cancer Prediction
cs.LG updates on arXiv.org
Learning Spatio-Temporal Foundation Models from Pure Synthetic Data
Learning Spatio-Temporal Foundation Models from Pure Synthetic Data
cs.LG updates on arXiv.org
SOS-LoRA: Static Orthogonal-Subspace Low-Rank Adaptation with Fixed Multi-Scale Scaling
SOS-LoRA: Static Orthogonal-Subspace Low-Rank Adaptation with Fixed Multi-Scale Scaling
cs.LG updates on arXiv.org
Comprehensive Evaluation of Machine Learning for Type 2 Diabetes Risk Prediction: Large-Scale External Validation and Fairness Analysis
Comprehensive Evaluation of Machine Learning for Type 2 Diabetes Risk Prediction: Large-Scale External Validation and Fairness Analysis
cs.LG updates on arXiv.org
More Than Memory: Task-Conditioned Signed FFN Writes in Long-Context Retrieval
More Than Memory: Task-Conditioned Signed FFN Writes in Long-Context Retrieval
cs.LG updates on arXiv.org
Feature Generation Using LLMs: An Evolutionary Algorithm Approach
Feature Generation Using LLMs: An Evolutionary Algorithm Approach
cs.LG updates on arXiv.org
Discovery by Dreaming: Cross-Domain Recombination in Artificial Memory
Discovery by Dreaming: Cross-Domain Recombination in Artificial Memory
cs.LG updates on arXiv.org
From Outcomes to Actions: Leveraging Hindsight for Long-Horizon Language Agent Training
From Outcomes to Actions: Leveraging Hindsight for Long-Horizon Language Agent Training
cs.LG updates on arXiv.org
Neural Controlled Differential Equations for EMT-Level Surrogate Modeling of Grid-Forming Inverters
Neural Controlled Differential Equations for EMT-Level Surrogate Modeling of Grid-Forming Inverters
cs.LG updates on arXiv.org
Quantifying Ranking Uncertainty in LLM Benchmarks
Quantifying Ranking Uncertainty in LLM Benchmarks
cs.LG updates on arXiv.org
AdaSurvMamba: Dynamic Fusion and Semantic Scanning for Multimodal Survival Analysis
AdaSurvMamba: Dynamic Fusion and Semantic Scanning for Multimodal Survival Analysis
cs.LG updates on arXiv.org
Reducing Per-Sample Harm in Stochastic Optimization
Reducing Per-Sample Harm in Stochastic Optimization
cs.LG updates on arXiv.org
Autonomous mechanistic discovery of colorectal cancer vulnerabilities via multi-scale AI swarms
Autonomous mechanistic discovery of colorectal cancer vulnerabilities via multi-scale AI swarms
cs.LG updates on arXiv.org
Preference-based Antibody Expression Ranking: Scaling with Large-scale Weak Supervision
Preference-based Antibody Expression Ranking: Scaling with Large-scale Weak Supervision
cs.LG updates on arXiv.org
PsiLogic: Chaos-Aware Active Cancellation for Adam with a Fair Cross-Domain Benchmark
PsiLogic: Chaos-Aware Active Cancellation for Adam with a Fair Cross-Domain Benchmark
cs.LG updates on arXiv.org
A Predict-then-Correct Loop Based on Few-Shot Continuous Contextual Bandit for Demand Forecasting
A Predict-then-Correct Loop Based on Few-Shot Continuous Contextual Bandit for Demand Forecasting
cs.LG updates on arXiv.org
Scaling Limits of Constant-Stepsize SGD at Flat Minima
Scaling Limits of Constant-Stepsize SGD at Flat Minima
cs.LG updates on arXiv.org
Retrieval is Enough: Training-Free Interpretability with a Tool-Using Agent
Retrieval is Enough: Training-Free Interpretability with a Tool-Using Agent
cs.LG updates on arXiv.org
EA-RMENet -- Path Loss Prediction in Urban Environments using Deep Learning
EA-RMENet -- Path Loss Prediction in Urban Environments using Deep Learning
cs.LG updates on arXiv.org
Compact convolutional neural networks for AI-based drone detection system
Compact convolutional neural networks for AI-based drone detection system
cs.LG updates on arXiv.org
K-IPO: Kendall-constrained Importance Preserving Oversampling for Imbalanced Tabular Data
K-IPO: Kendall-constrained Importance Preserving Oversampling for Imbalanced Tabular Data
cs.LG updates on arXiv.org
Leakage-Robust Evaluation and Data-Scale Sensitivity of Attention-Enhanced Multi-Task Learning for Joint Fault Diagnosis and Remaining Useful Life Estimation
Leakage-Robust Evaluation and Data-Scale Sensitivity of Attention-Enhanced Multi-Task Learning for Joint Fault Diagnosis and Remaining Useful Life Estimation
cs.LG updates on arXiv.org
Feedback Attribution and Representation Geometry: Metrics for Comparing Individual and Shared Rewards in MARL
Feedback Attribution and Representation Geometry: Metrics for Comparing Individual and Shared Rewards in MARL
cs.LG updates on arXiv.org
Building2Building: A Large Scale Benchmark for Generalizable Real-World Reinforcement Learning
Building2Building: A Large Scale Benchmark for Generalizable Real-World Reinforcement Learning
cs.LG updates on arXiv.org
Discrete Ricci Curvature on Protein Contact Graphs for Lightweight Fold Classification
Discrete Ricci Curvature on Protein Contact Graphs for Lightweight Fold Classification
cs.LG updates on arXiv.org
Capacity and Redundancy Trade-offs in Multi-Task Learning
Capacity and Redundancy Trade-offs in Multi-Task Learning
cs.LG updates on arXiv.org
Learning from World Feedback: Why Model Uncertainty Fails as a Risk Signal in Model-Based RL
Learning from World Feedback: Why Model Uncertainty Fails as a Risk Signal in Model-Based RL
cs.LG updates on arXiv.org
Privacy Cost as Equity Input: A Group Fairness Criterion for Differentially Private Machine Learning
Privacy Cost as Equity Input: A Group Fairness Criterion for Differentially Private Machine Learning
cs.LG updates on arXiv.org
Multimodal Attention-based Deep Learning for Emergency Triage with Electronic Health Records
Multimodal Attention-based Deep Learning for Emergency Triage with Electronic Health Records
cs.LG updates on arXiv.org
CLDRoute: Conditional Latent Diffusion for Routability Map Generation in Physical Design
CLDRoute: Conditional Latent Diffusion for Routability Map Generation in Physical Design
cs.LG updates on arXiv.org
A Framework for Early Sepsis Prediction via Self-Supervised (JEPA) and Federated Representation Learning
A Framework for Early Sepsis Prediction via Self-Supervised (JEPA) and Federated Representation Learning
cs.LG updates on arXiv.org
Building a Neural Network from Scratch: Implementation, Evaluation, and Optimization
Building a Neural Network from Scratch: Implementation, Evaluation, and Optimization
cs.LG updates on arXiv.org
Effects of width-dependent model hyperparameters and $\ell_2$-regularization on the loss landscape of two-layer ReLU networks
Effects of width-dependent model hyperparameters and $\ell_2$-regularization on the loss landscape of two-layer ReLU networks
cs.LG updates on arXiv.org
Half the Experts, All the Code: One-Shot Domain Pruning of Mixture-of-Experts LLMs for Coding
Half the Experts, All the Code: One-Shot Domain Pruning of Mixture-of-Experts LLMs for Coding
cs.LG updates on arXiv.org
Graph-Embedded Intuitionistic Fuzzy Broad Learning System: A Multi-view Framework
Graph-Embedded Intuitionistic Fuzzy Broad Learning System: A Multi-view Framework
cs.LG updates on arXiv.org
BG4Sea: Biogeochemical Seasonal Forecastability via Progressive Information Scaling
BG4Sea: Biogeochemical Seasonal Forecastability via Progressive Information Scaling
cs.LG updates on arXiv.org
The Anatomy of a Truth Direction: Knowledge-Dependent Dimensionality, a Relational Law, and a Convergent Category Geometry in Small Language Models
The Anatomy of a Truth Direction: Knowledge-Dependent Dimensionality, a Relational Law, and a Convergent Category Geometry in Small Language Models
cs.LG updates on arXiv.org
Robust Losses from Univariate Base Functions for Noisy-Label Learning
Robust Losses from Univariate Base Functions for Noisy-Label Learning
cs.LG updates on arXiv.org
On the Potential of Graph Neural Networks as Metamodels for Supply Chain Optimization: Dataset, Architectures, and Directions
On the Potential of Graph Neural Networks as Metamodels for Supply Chain Optimization: Dataset, Architectures, and Directions
cs.LG updates on arXiv.org
MultiLoReFT: Decoupling Shared and Modality-Specific Subspaces in Multimodal Learning via Low-Rank Representation Fine-Tuning
MultiLoReFT: Decoupling Shared and Modality-Specific Subspaces in Multimodal Learning via Low-Rank Representation Fine-Tuning
cs.LG updates on arXiv.org
Value-Monotonicity Matters: A Concordance Loss for Deep Survival Prediction
Value-Monotonicity Matters: A Concordance Loss for Deep Survival Prediction
cs.LG updates on arXiv.org
Interpretable Anomaly and Drift Detection with Gaussian Mixture Models
Interpretable Anomaly and Drift Detection with Gaussian Mixture Models
cs.LG updates on arXiv.org
Honest Physical-Support Inference after Latent Dictionary Learning: Collision Singularities and Minimax Resolution
Honest Physical-Support Inference after Latent Dictionary Learning: Collision Singularities and Minimax Resolution
cs.LG updates on arXiv.org
First-Order Predictable but Pairwise Fragile: Local Task Adaptation in Trained Transformers
First-Order Predictable but Pairwise Fragile: Local Task Adaptation in Trained Transformers
cs.LG updates on arXiv.org
Beyond Memory Leaderboards: Evaluating Scientific Memory as Budgeted Context Restoration
Beyond Memory Leaderboards: Evaluating Scientific Memory as Budgeted Context Restoration
cs.LG updates on arXiv.org
Principled Direction-Free Intrinsic Motivation through Model-Free Epistemic Free-Energy Estimators
Principled Direction-Free Intrinsic Motivation through Model-Free Epistemic Free-Energy Estimators
cs.LG updates on arXiv.org
Bridging battery design and health assessment through virtual sensing and physics-informed learning
Bridging battery design and health assessment through virtual sensing and physics-informed learning
cs.LG updates on arXiv.org
HyBDM: Multi-Scale Hybrid Experts for Time Series Forecasting with Bidirectional Dependency Modeling
HyBDM: Multi-Scale Hybrid Experts for Time Series Forecasting with Bidirectional Dependency Modeling
cs.LG updates on arXiv.org
Certified-Gap Dual-Price Policies for Real-Time Truckload Bid Acceptance with Relocating, Clock-Constrained Resources
Certified-Gap Dual-Price Policies for Real-Time Truckload Bid Acceptance with Relocating, Clock-Constrained Resources
cs.LG updates on arXiv.org
TVGL-CFM:Generating and Forecasting Time-Varying Trajectories of Dynamic Networks with Conditional Flow Matching
TVGL-CFM:Generating and Forecasting Time-Varying Trajectories of Dynamic Networks with Conditional Flow Matching
cs.LG updates on arXiv.org
When Can Safe Controllers Adapt? Information before Commitment
When Can Safe Controllers Adapt? Information before Commitment
cs.LG updates on arXiv.org
Enhancing Personalized Bladder Cancer Treatment Through Reinforcement Learning: A Recurrent Patient State Transition Decision Support Framework
Enhancing Personalized Bladder Cancer Treatment Through Reinforcement Learning: A Recurrent Patient State Transition Decision Support Framework
cs.LG updates on arXiv.org
Investigation of Polycystic Ovary Syndrome (PCOS) Diagnosis Using Machine Learning Approaches
Investigation of Polycystic Ovary Syndrome (PCOS) Diagnosis Using Machine Learning Approaches
cs.LG updates on arXiv.org
CADENCE: Closing the Reasoning Gap via Coverage-Adaptive On-Policy Distillation
CADENCE: Closing the Reasoning Gap via Coverage-Adaptive On-Policy Distillation
cs.LG updates on arXiv.org
SurvCF(t): Counterfactual Explanations for Survival Analysis in Predictive Maintenance Multivariate Time Series Data
SurvCF(t): Counterfactual Explanations for Survival Analysis in Predictive Maintenance Multivariate Time Series Data
cs.LG updates on arXiv.org
TurboVec: A Case Study in Cost-Efficient Private Retrieval for Enterprise RAG via Codebook-Oblivious Quantization
TurboVec: A Case Study in Cost-Efficient Private Retrieval for Enterprise RAG via Codebook-Oblivious Quantization
cs.LG updates on arXiv.org
Periodic Bootstrap Thompson Sampling For Periodically Non-Stationary Bandit Problems
Periodic Bootstrap Thompson Sampling For Periodically Non-Stationary Bandit Problems
cs.LG updates on arXiv.org
Counterfactual Shapley Credit Assignment
Counterfactual Shapley Credit Assignment
cs.LG updates on arXiv.org
Regularize or Localize: When Training-Time KV-Cache Geometry Pays Under Quantization
Regularize or Localize: When Training-Time KV-Cache Geometry Pays Under Quantization
cs.LG updates on arXiv.org
Interpretable Machine Learning for Air Pollution and Respiratory Health Prediction: A Socioeconomic Subgroup Analysis
Interpretable Machine Learning for Air Pollution and Respiratory Health Prediction: A Socioeconomic Subgroup Analysis
cs.LG updates on arXiv.org
ChemFusion: A Multimodal Cross-Attention Network for Reaction Yield Prediction
ChemFusion: A Multimodal Cross-Attention Network for Reaction Yield Prediction
cs.LG updates on arXiv.org
Apeliotes: A Diffusion-Based Modeling Framework for km-scale Multi-Level Atmospheric Fields
Apeliotes: A Diffusion-Based Modeling Framework for km-scale Multi-Level Atmospheric Fields
cs.LG updates on arXiv.org
Solver-Hard Is Not Model-Hard: A Hardness-Controlled Diagnostic for LLM Constraint Reasoning
Solver-Hard Is Not Model-Hard: A Hardness-Controlled Diagnostic for LLM Constraint Reasoning
cs.LG updates on arXiv.org
What does a Bayes-filtered transformer believe? A predictive Monte Carlo approach
What does a Bayes-filtered transformer believe? A predictive Monte Carlo approach
cs.LG updates on arXiv.org
Persistent Sparse Autoencoders: Learning Feature Timescales in Language Models
Persistent Sparse Autoencoders: Learning Feature Timescales in Language Models
cs.LG updates on arXiv.org
Robust Assamese Speech Recognition through Controlled Fine-Tuning of Whisper Models
Robust Assamese Speech Recognition through Controlled Fine-Tuning of Whisper Models
cs.LG updates on arXiv.org
Explaining and Tuning Transformer-based LLMs in Arithmetic Tasks with Human Strategies
Explaining and Tuning Transformer-based LLMs in Arithmetic Tasks with Human Strategies
cs.LG updates on arXiv.org
DADIR: Density-Aware Data-level Imbalanced Regression Framework
DADIR: Density-Aware Data-level Imbalanced Regression Framework
cs.LG updates on arXiv.org
DynImmune-BERT: Dynamic Immune Repertoire Modeling with Neural ODE Driven Continuous Transformers
DynImmune-BERT: Dynamic Immune Repertoire Modeling with Neural ODE Driven Continuous Transformers
cs.LG updates on arXiv.org
Distilled Reinforcement Learning for LLM Post-training
Distilled Reinforcement Learning for LLM Post-training
cs.LG updates on arXiv.org
Node4All: Learning Node Representation Beyond Datasets
Node4All: Learning Node Representation Beyond Datasets
cs.LG updates on arXiv.org
AIGB-R1: Self-Evolving Generative Auto-Bidding via Hierarchical Planner-Executor Optimization
AIGB-R1: Self-Evolving Generative Auto-Bidding via Hierarchical Planner-Executor Optimization
cs.LG updates on arXiv.org
An Iterative Geometric Approach to Optimizing Separating Hyperplanes
An Iterative Geometric Approach to Optimizing Separating Hyperplanes
cs.LG updates on arXiv.org
Lookahead Branching for Neural Network Verification
Lookahead Branching for Neural Network Verification
cs.LG updates on arXiv.org
DRNOISE: Benchmarking Deep Research Agents in Misleading Evidence Environments
DRNOISE: Benchmarking Deep Research Agents in Misleading Evidence Environments
cs.LG updates on arXiv.org
WAR: Workload-Aware Rollouts for Synchronous Agentic Reinforcement Learning
WAR: Workload-Aware Rollouts for Synchronous Agentic Reinforcement Learning
cs.LG updates on arXiv.org
Rationalizing Boltzmann Rationality: An Axiomatic Characterization of Entropy-Regularized Policies
Rationalizing Boltzmann Rationality: An Axiomatic Characterization of Entropy-Regularized Policies
cs.LG updates on arXiv.org
TAPAS: Throughput-adaptive Perception for Autonomous Systems
TAPAS: Throughput-adaptive Perception for Autonomous Systems
cs.LG updates on arXiv.org
Rethinking the Suitability of Reinforcement Learning Algorithms Under Practical Transfer Constraints
Rethinking the Suitability of Reinforcement Learning Algorithms Under Practical Transfer Constraints
cs.LG updates on arXiv.org
When Drift Detectors cry Wolf: False Alarm Rates in continuous ML Monitoring
When Drift Detectors cry Wolf: False Alarm Rates in continuous ML Monitoring
cs.LG updates on arXiv.org
A multiverse-consensus pipeline for reproducible feature selection in untargeted LC-MS metabolomics
A multiverse-consensus pipeline for reproducible feature selection in untargeted LC-MS metabolomics
cs.LG updates on arXiv.org
CoEvoP&R: Co-Evolving Placement Objectives with Routing Feedback via Large Language Models
CoEvoP&R: Co-Evolving Placement Objectives with Routing Feedback via Large Language Models
cs.LG updates on arXiv.org
CORAL: Learning Amyloid Fibril Ligand Docking with Cooperative Binding Rewards
CORAL: Learning Amyloid Fibril Ligand Docking with Cooperative Binding Rewards
cs.LG updates on arXiv.org
Grounded verification of chemical and materials reasoning: detection is the bottleneck
Grounded verification of chemical and materials reasoning: detection is the bottleneck
cs.LG updates on arXiv.org
Kernelized Linear Attention: Breaking the Capacity Wall with Symmetric Cones
Kernelized Linear Attention: Breaking the Capacity Wall with Symmetric Cones
cs.LG updates on arXiv.org
Decoder-Preserving Sparse Autoencoders: Which Readouts Survive Sparse Compression?
Decoder-Preserving Sparse Autoencoders: Which Readouts Survive Sparse Compression?
cs.LG updates on arXiv.org
Abliteration Is Not a Scalpel: Off-Target Effects of Refusal Removal on Decision Disposition Across Model Families
Abliteration Is Not a Scalpel: Off-Target Effects of Refusal Removal on Decision Disposition Across Model Families
cs.LG updates on arXiv.org
Calibrated Alzheimer's Conversion Risk in Mild Cognitive Impairment: Persistent Homology of Clinical Trajectories with Conformal Guarantees
Calibrated Alzheimer's Conversion Risk in Mild Cognitive Impairment: Persistent Homology of Clinical Trajectories with Conformal Guarantees
cs.LG updates on arXiv.org
Residual-Guided Multi-Resolution Refinement of Foundation Models: A Case Study in Drought Forecasting
Residual-Guided Multi-Resolution Refinement of Foundation Models: A Case Study in Drought Forecasting
cs.LG updates on arXiv.org
Retrieval-Augmented Interpretable Learning: Towards Task-Specific Zero-Shot Models in Healthcare
Retrieval-Augmented Interpretable Learning: Towards Task-Specific Zero-Shot Models in Healthcare
cs.LG updates on arXiv.org
Lightweight Wrappers for Adapting Time Series Foundation Models to Regional Drought Forecasting
Lightweight Wrappers for Adapting Time Series Foundation Models to Regional Drought Forecasting
cs.LG updates on arXiv.org
After the Euclidean Highway: Hyperbolic Expert AI as the Next Innovation
After the Euclidean Highway: Hyperbolic Expert AI as the Next Innovation
cs.LG updates on arXiv.org
One-step lowest-variance selection in a Gaussian random-field model motivated by masked diffusion: Total correlation and a square root collision threshold
One-step lowest-variance selection in a Gaussian random-field model motivated by masked diffusion: Total correlation and a square root collision threshold
cs.LG updates on arXiv.org
FailureAtlas: A Taxonomy of Failure Modes in Multi-Provider LLM Serving Infrastructure
FailureAtlas: A Taxonomy of Failure Modes in Multi-Provider LLM Serving Infrastructure
cs.LG updates on arXiv.org
Program Synthesis for Simulation-Based Inference: Joint Model Selection and Parameter Estimation
Program Synthesis for Simulation-Based Inference: Joint Model Selection and Parameter Estimation
cs.LG updates on arXiv.org
Volatility-Aware Extreme Event Detection in High-Frequency Financial Markets
Volatility-Aware Extreme Event Detection in High-Frequency Financial Markets
cs.LG updates on arXiv.org
CoCurve: Cross-Module Co-Pruning Curvature for Training-Free Structured LLM Pruning
CoCurve: Cross-Module Co-Pruning Curvature for Training-Free Structured LLM Pruning
cs.LG updates on arXiv.org
A Weisfeiler-Leman Characterization of Global-Attention Graph Transformers for Mixed-Integer Linear Programs
A Weisfeiler-Leman Characterization of Global-Attention Graph Transformers for Mixed-Integer Linear Programs
cs.LG updates on arXiv.org
AGG: Jacobian-Aggregated Group Gradient for Efficient GRPO Training of Diffusion Models
AGG: Jacobian-Aggregated Group Gradient for Efficient GRPO Training of Diffusion Models
cs.LG updates on arXiv.org
ANNLib: A Development Framework for Efficient Approximate Nearest Neighbor Search
ANNLib: A Development Framework for Efficient Approximate Nearest Neighbor Search
cs.LG updates on arXiv.org
Concentration and Mean-Square Bounds for Contractive Stochastic Approximation: A Unified Elementary Approach
Concentration and Mean-Square Bounds for Contractive Stochastic Approximation: A Unified Elementary Approach
cs.LG updates on arXiv.org
Trustworthy Protein-Ligand Binding Affinity Prediction via Reliability-Aware Multi-Engine Fusion
Trustworthy Protein-Ligand Binding Affinity Prediction via Reliability-Aware Multi-Engine Fusion
cs.LG updates on arXiv.org
Optimizing the Preconditioner: A Black-box Online-to-Nonconvex Conversion with Static Regret Minimization Oracles
Optimizing the Preconditioner: A Black-box Online-to-Nonconvex Conversion with Static Regret Minimization Oracles
cs.LG updates on arXiv.org
PoLoRA: A Preconditioned Orthogonalized LoRA Optimizer
PoLoRA: A Preconditioned Orthogonalized LoRA Optimizer
cs.LG updates on arXiv.org
Can Transformers Really Do It All? On the Compatibility of Inductive Biases Across Tasks
Can Transformers Really Do It All? On the Compatibility of Inductive Biases Across Tasks
cs.LG updates on arXiv.org
TypiCore: A Hybrid Active Query Strategy for Class-Incremental Learning on Time Series
TypiCore: A Hybrid Active Query Strategy for Class-Incremental Learning on Time Series
cs.LG updates on arXiv.org
Selectivity Matters: Source Node Influence Pruning for Unsupervised Graph Domain Adaptation
Selectivity Matters: Source Node Influence Pruning for Unsupervised Graph Domain Adaptation
cs.LG updates on arXiv.org
GeneSpeak-FP: Target and Compound Retrieval from Observed Cell-Level Perturbation Signatures
GeneSpeak-FP: Target and Compound Retrieval from Observed Cell-Level Perturbation Signatures
cs.LG updates on arXiv.org
Beyond Objective Expressivity: Geometry Preservation in Multimodal Contrastive Learning
Beyond Objective Expressivity: Geometry Preservation in Multimodal Contrastive Learning
cs.LG updates on arXiv.org
Uncovering Latent Reasoning Strategies in Language Models
Uncovering Latent Reasoning Strategies in Language Models
cs.LG updates on arXiv.org
Planning with Transformers: Chain of Computation and Structured Context Windows
Planning with Transformers: Chain of Computation and Structured Context Windows
cs.LG updates on arXiv.org
MXSens: Sensitivity-Aware Mixed-Precision Quantization for Efficient LLM Inference
MXSens: Sensitivity-Aware Mixed-Precision Quantization for Efficient LLM Inference
cs.LG updates on arXiv.org
Towards Reliable Zero-Shot Crowd Forecasting: Evaluating Time Series Foundation Models for Special Event Pedestrian Forecasting
Towards Reliable Zero-Shot Crowd Forecasting: Evaluating Time Series Foundation Models for Special Event Pedestrian Forecasting
cs.LG updates on arXiv.org
Generalize and Guide: Decomposing Rewards for Few-Shot Inverse Reinforcement Learning
Generalize and Guide: Decomposing Rewards for Few-Shot Inverse Reinforcement Learning
cs.LG updates on arXiv.org
FIFA World Cup 2026 as a Contamination-Free Benchmark for LLM Forecasting Agents: Four Models, a Bookmaker, and 104 Matches
FIFA World Cup 2026 as a Contamination-Free Benchmark for LLM Forecasting Agents: Four Models, a Bookmaker, and 104 Matches
cs.LG updates on arXiv.org
Feature Attribution-Based Explainability Analysis of Deep Learning Models in Predictive Process Monitoring
Feature Attribution-Based Explainability Analysis of Deep Learning Models in Predictive Process Monitoring
cs.LG updates on arXiv.org
The Concept of Representation in ML: Beyond Plato and Aristotle
The Concept of Representation in ML: Beyond Plato and Aristotle
cs.LG updates on arXiv.org
Phasor Attention: Mean Root Square Normalization for Phase Manifold Preservation
Phasor Attention: Mean Root Square Normalization for Phase Manifold Preservation
cs.LG updates on arXiv.org
Theoretical Foundations of $\max$@$k$ Reinforcement Learning
Theoretical Foundations of $\max$@$k$ Reinforcement Learning
cs.LG updates on arXiv.org
Mobius Learning: Cyclic Depth Folding in Transformers
Mobius Learning: Cyclic Depth Folding in Transformers
cs.LG updates on arXiv.org
Distributional Soft Bellman Operator under the Cram\'er Geometry
Distributional Soft Bellman Operator under the Cram\'er Geometry
cs.LG updates on arXiv.org
A Geometric Perspective on Stabilizing Value Conflict Resolution
A Geometric Perspective on Stabilizing Value Conflict Resolution
cs.LG updates on arXiv.org
Topological Signatures of Context-Level Reliability in TabPFN
Topological Signatures of Context-Level Reliability in TabPFN
cs.LG updates on arXiv.org
DiFA: Inference-Time Forward-Process Alignment for Diffusion Models
DiFA: Inference-Time Forward-Process Alignment for Diffusion Models
cs.LG updates on arXiv.org
Harness Engineering for LLM-Driven GPU Kernel Generation
Harness Engineering for LLM-Driven GPU Kernel Generation
cs.LG updates on arXiv.org
Information-Based Exploration via Random Features for Reinforcement Learning
Information-Based Exploration via Random Features for Reinforcement Learning
cs.LG updates on arXiv.org
fSRD: Fuzzy Spectral Region Decomposition -- Automated Multi Operator Koopman Representations via an Adaptive Spectral Learning Architecture
fSRD: Fuzzy Spectral Region Decomposition -- Automated Multi Operator Koopman Representations via an Adaptive Spectral Learning Architecture
cs.LG updates on arXiv.org
MADA-RL: Multi-Agent Debate-Aware Reinforcement Learning for Parameter-Efficient Reasoning in Compact Models
MADA-RL: Multi-Agent Debate-Aware Reinforcement Learning for Parameter-Efficient Reasoning in Compact Models
cs.LG updates on arXiv.org
FlashPDE: A Drop-in Fused Triton Operator Library for Neural PDE Solvers
FlashPDE: A Drop-in Fused Triton Operator Library for Neural PDE Solvers
cs.LG updates on arXiv.org
L1 Augmented Attention as an Improved Vector Similarity Metric
L1 Augmented Attention as an Improved Vector Similarity Metric
cs.LG updates on arXiv.org
SEE: Structure-aware Exploring \& Exploiting for Long-horizon GUI Agent Trajectory Synthesis
SEE: Structure-aware Exploring \& Exploiting for Long-horizon GUI Agent Trajectory Synthesis
cs.LG updates on arXiv.org
SGN: A Similarity-based Generative Network for Data Generation under Distribution Shift
SGN: A Similarity-based Generative Network for Data Generation under Distribution Shift
cs.LG updates on arXiv.org
Sobek: Streaming Equivariant Tensor Product Convolutions
Sobek: Streaming Equivariant Tensor Product Convolutions
cs.LG updates on arXiv.org
Generalised Bellman recurrence and three dualities in sequential decision-making
Generalised Bellman recurrence and three dualities in sequential decision-making
cs.LG updates on arXiv.org
SelectInfer: Selective Neuron Loading and Computation for On-Device LLMs
SelectInfer: Selective Neuron Loading and Computation for On-Device LLMs
cs.LG updates on arXiv.org
Enhancing Rubric-based RL via Self-Distillation
Enhancing Rubric-based RL via Self-Distillation
cs.LG updates on arXiv.org
The Label Complexity of Class-Conditional Coverage under Distribution Shift
The Label Complexity of Class-Conditional Coverage under Distribution Shift
cs.LG updates on arXiv.org
Empowering On-Device Model Adaptation with an Edge AI Inference Accelerator
Empowering On-Device Model Adaptation with an Edge AI Inference Accelerator
cs.LG updates on arXiv.org
LLM-as-a-Coach: Experiential Learning for Non-Verifiable Tasks
LLM-as-a-Coach: Experiential Learning for Non-Verifiable Tasks
cs.LG updates on arXiv.org
Manifold-Constrained Hyper-Connections for Parameter-Efficient Finetuning
Manifold-Constrained Hyper-Connections for Parameter-Efficient Finetuning
cs.LG updates on arXiv.org
Do Language Models Dream of Binding Molecules? Benchmarking LLMs under Spatial Constraints
Do Language Models Dream of Binding Molecules? Benchmarking LLMs under Spatial Constraints
cs.LG updates on arXiv.org
Totally Positive Matrices and the Highest-Order Coefficients of the Characteristic Polynomial
Totally Positive Matrices and the Highest-Order Coefficients of the Characteristic Polynomial
cs.LG updates on arXiv.org
Differentiable Logic Gate Networks for Low-Latency EEG Classification on Edge Devices
Differentiable Logic Gate Networks for Low-Latency EEG Classification on Edge Devices
cs.LG updates on arXiv.org
The Calibration Channel Determines the Bayes-Error Proxy: An Exact Law for Temperature-Induced Distortion
The Calibration Channel Determines the Bayes-Error Proxy: An Exact Law for Temperature-Induced Distortion
cs.LG updates on arXiv.org
OR Else: A Differentiable Trust Region for Policy Optimization
OR Else: A Differentiable Trust Region for Policy Optimization
cs.LG updates on arXiv.org
A Continual Validation, Updating, and Decision-Making Framework for Self-Adaptive Digital Twins via Robust Model Predictive Control: A Case Study in Additive Manufacturing
A Continual Validation, Updating, and Decision-Making Framework for Self-Adaptive Digital Twins via Robust Model Predictive Control: A Case Study in Additive Manufacturing
cs.LG updates on arXiv.org
FlashRT: Agent Harness for Guiding Agents to Deploy Real-Time Multimodal Applications
FlashRT: Agent Harness for Guiding Agents to Deploy Real-Time Multimodal Applications
cs.LG updates on arXiv.org
Three-Body Scattering for Generative Modeling
Three-Body Scattering for Generative Modeling
cs.LG updates on arXiv.org
Causal Discovery on Irregular Time Series
Causal Discovery on Irregular Time Series
cs.LG updates on arXiv.org
A Survey on GNN-based Link Prediction: Techniques, Applications, and Challenges
A Survey on GNN-based Link Prediction: Techniques, Applications, and Challenges
cs.LG updates on arXiv.org
Masked Diffusion Language Models are Strong and Steerable Text-Based World Models for Agentic RL
Masked Diffusion Language Models are Strong and Steerable Text-Based World Models for Agentic RL
cs.LG updates on arXiv.org
Comparing Spectrogram Front-Ends for Abnormal Heart-Sound Detection with a Convolutional Neural Network
Comparing Spectrogram Front-Ends for Abnormal Heart-Sound Detection with a Convolutional Neural Network
cs.LG updates on arXiv.org
Overcoming the BCI Calibration Bottleneck: A Clinically-Grounded Architecture using Riemannian Alignment and Stochastic Weight Averaging
Overcoming the BCI Calibration Bottleneck: A Clinically-Grounded Architecture using Riemannian Alignment and Stochastic Weight Averaging
cs.LG updates on arXiv.org
FinBench: Time-Gated Calibration and Uncertainty Benchmarking for Agentic Financial Forecasting
FinBench: Time-Gated Calibration and Uncertainty Benchmarking for Agentic Financial Forecasting
cs.LG updates on arXiv.org
Composable Verification Pipelines for Multi-Agent Systems
Composable Verification Pipelines for Multi-Agent Systems
cs.LG updates on arXiv.org
A Novel Hybrid Quantum Reservoir Computing (nHQRC) for Phase Transition Detection in Non-Equilibrium Dynamical Systems
A Novel Hybrid Quantum Reservoir Computing (nHQRC) for Phase Transition Detection in Non-Equilibrium Dynamical Systems
cs.LG updates on arXiv.org
Depth Estimators Are Implicit Neural Fields for 3D Scene Geometry Inpainting and Reconstruction
Depth Estimators Are Implicit Neural Fields for 3D Scene Geometry Inpainting and Reconstruction
cs.LG updates on arXiv.org
It Depends on the Dataset: When a Brain-Encoding Model's Predicted Responses Beat Their Visual Backbone for Video Memorability
It Depends on the Dataset: When a Brain-Encoding Model's Predicted Responses Beat Their Visual Backbone for Video Memorability
cs.LG updates on arXiv.org
SLT: Robust Quantum Neural Networks for Noisy-Label Medical Image Classification via Supermartingale-based Label Transition
SLT: Robust Quantum Neural Networks for Noisy-Label Medical Image Classification via Supermartingale-based Label Transition
cs.LG updates on arXiv.org
Efficient EEG Seizure Detection Using INT8 Quantization, Channel Pruning, and Spiking Neural Networks
Efficient EEG Seizure Detection Using INT8 Quantization, Channel Pruning, and Spiking Neural Networks
cs.LG updates on arXiv.org
On Hardware-Aware Design and Optimization of Edge Intelligence
On Hardware-Aware Design and Optimization of Edge Intelligence
cs.LG updates on arXiv.org
Depth-Regularized JEPA World Models Learn More Transferable Representations from Real Outdoor Robot Data
Depth-Regularized JEPA World Models Learn More Transferable Representations from Real Outdoor Robot Data
cs.LG updates on arXiv.org
Monte Carlo Dropout Uncertainty and Entropy-Thresholded Selective Prediction for Architecture-Agnostic Brain Tumor MRI Triage
Monte Carlo Dropout Uncertainty and Entropy-Thresholded Selective Prediction for Architecture-Agnostic Brain Tumor MRI Triage
cs.LG updates on arXiv.org
ECG-LLM: Foundation Model for ECG-Based Cardiac Reasoning
ECG-LLM: Foundation Model for ECG-Based Cardiac Reasoning
cs.LG updates on arXiv.org
SGMCE: Segment-Grounded Morphological Concept Explanation for Malaria Parasite Species Identification in Thick Blood Smears
SGMCE: Segment-Grounded Morphological Concept Explanation for Malaria Parasite Species Identification in Thick Blood Smears
cs.LG updates on arXiv.org
RegionFM: Interpretable Region-Based Brain MRI Classification Using Foundation Model Embeddings
RegionFM: Interpretable Region-Based Brain MRI Classification Using Foundation Model Embeddings
cs.LG updates on arXiv.org
Lipschitz Continuity in Deep Learning: A Systematic Review of Theoretical Foundations, Estimation Methods, Regularization Approaches, and Certifiable Robustness
Lipschitz Continuity in Deep Learning: A Systematic Review of Theoretical Foundations, Estimation Methods, Regularization Approaches, and Certifiable Robustness
cs.LG updates on arXiv.org
AEVAL: From Anecdotal to Deterministic Testing for Agentic Skill Workflows
AEVAL: From Anecdotal to Deterministic Testing for Agentic Skill Workflows
cs.LG updates on arXiv.org
Boundary-Seeking GAN-Augmented TabTransformer for Adversarially Robust Intrusion Detection
Boundary-Seeking GAN-Augmented TabTransformer for Adversarially Robust Intrusion Detection
cs.LG updates on arXiv.org
Joint-Embedding Predictive Architecture for Sensor-based Activity Recognition
Joint-Embedding Predictive Architecture for Sensor-based Activity Recognition
cs.LG updates on arXiv.org
MTSSL: Meta-Thresholding Semi-Supervised Learning
MTSSL: Meta-Thresholding Semi-Supervised Learning
cs.LG updates on arXiv.org
AoA: Theorem Proving Agent over Abstract Syntax Tree of Redesigned Language
AoA: Theorem Proving Agent over Abstract Syntax Tree of Redesigned Language
cs.LG updates on arXiv.org
Disentangling Model and Human Data Uncertainty in Apparent Facial Age Estimation
Disentangling Model and Human Data Uncertainty in Apparent Facial Age Estimation
cs.LG updates on arXiv.org
Think, Plan, Paint: Layout-Aware Reasoning for Controllable Image Generation in Unified Models
Think, Plan, Paint: Layout-Aware Reasoning for Controllable Image Generation in Unified Models
cs.LG updates on arXiv.org
Back to the museum: Investigation of the acceptance of Android Andrea with and without emotion simulation in a museum
Back to the museum: Investigation of the acceptance of Android Andrea with and without emotion simulation in a museum
cs.LG updates on arXiv.org
Spectral-Morphological Attention U-Net: An Efficient Network for Active Wildfire Detection
Spectral-Morphological Attention U-Net: An Efficient Network for Active Wildfire Detection
cs.LG updates on arXiv.org
Foresight Residual RL for Long-Horizon Robot Manipulation with Vision-Language-Action Models
Foresight Residual RL for Long-Horizon Robot Manipulation with Vision-Language-Action Models
cs.LG updates on arXiv.org
From Modalities to Propositions: A Language-Centric Framework for Multimodal Intelligence
From Modalities to Propositions: A Language-Centric Framework for Multimodal Intelligence
cs.LG updates on arXiv.org
Exact Network Surgery: Functional Invariance and Gradient Plasticity in Reactive Computational Graphs
Exact Network Surgery: Functional Invariance and Gradient Plasticity in Reactive Computational Graphs
cs.LG updates on arXiv.org
Mapping Order in Semicrystalline Polymers using Machine Learning of Nanobeam Electron Diffraction
Mapping Order in Semicrystalline Polymers using Machine Learning of Nanobeam Electron Diffraction
cs.LG updates on arXiv.org
Harnessing disorder to decouple extension and shear in kirigami metamaterials
Harnessing disorder to decouple extension and shear in kirigami metamaterials
cs.LG updates on arXiv.org
Backpropagation-Free Trunk Training via the Split Forward Gradients
Backpropagation-Free Trunk Training via the Split Forward Gradients
cs.LG updates on arXiv.org
De-floored Principal Component Regression: When Rank Selection Alone Is Insufficient for Prediction
De-floored Principal Component Regression: When Rank Selection Alone Is Insufficient for Prediction
cs.LG updates on arXiv.org
How Do You Choose Your AI Component? An Interview Study of Secure AI Integration in Practice
How Do You Choose Your AI Component? An Interview Study of Secure AI Integration in Practice
cs.LG updates on arXiv.org
OpenLanguageModel: Readable and Composable Small-Language-Model Pretraining for Education and Research
OpenLanguageModel: Readable and Composable Small-Language-Model Pretraining for Education and Research
cs.LG updates on arXiv.org
The Value of Depth in Message Passing on Sparse Graphs: A Kesten-Stigum Dichotomy
The Value of Depth in Message Passing on Sparse Graphs: A Kesten-Stigum Dichotomy
cs.LG updates on arXiv.org
Semi-Supervised Conditional Diffusion via Label Augmentation
Semi-Supervised Conditional Diffusion via Label Augmentation
cs.LG updates on arXiv.org
Semi-Supervised Conditional Generative Learning through Stochastic Interpolation and Sufficient Representations
Semi-Supervised Conditional Generative Learning through Stochastic Interpolation and Sufficient Representations
cs.LG updates on arXiv.org
Dropout and Random Gradient Masking Are Asymptotically Equivalent in Large ResNets
Dropout and Random Gradient Masking Are Asymptotically Equivalent in Large ResNets
cs.LG updates on arXiv.org
Demodulation of chaotic signals using convolutional neural network
Demodulation of chaotic signals using convolutional neural network
cs.LG updates on arXiv.org
Diagnosing Correctness Probes under Self-Judgement Confounding
Diagnosing Correctness Probes under Self-Judgement Confounding
cs.LG updates on arXiv.org
Identity-Paired Progressive Depth Training: When Trainability Persists Beyond Expressibility
Identity-Paired Progressive Depth Training: When Trainability Persists Beyond Expressibility
cs.LG updates on arXiv.org
Explainable Lightweight Compact Deep Models for Speech Emotion Recognition
Explainable Lightweight Compact Deep Models for Speech Emotion Recognition
cs.LG updates on arXiv.org
Trace-Based On-Policy Distillation for Masked Diffusion Language Models
Trace-Based On-Policy Distillation for Masked Diffusion Language Models
cs.LG updates on arXiv.org
Hierarchical Wireless Foundation Model for Multi-Task Optimization
Hierarchical Wireless Foundation Model for Multi-Task Optimization
cs.LG updates on arXiv.org
Transferable Low-Rank Convolutional Bases for Onboarding Unseen Medical Imaging Modalities
Transferable Low-Rank Convolutional Bases for Onboarding Unseen Medical Imaging Modalities
cs.LG updates on arXiv.org
A Method for Learning Value Systems in Generative AI
A Method for Learning Value Systems in Generative AI
cs.LG updates on arXiv.org
Optimizing Clinical Trial Protocols Using EHR-Derived Heterogeneous Treatment Effects
Optimizing Clinical Trial Protocols Using EHR-Derived Heterogeneous Treatment Effects
cs.LG updates on arXiv.org
Pediatric Bone Age Prediction Using Deep Learning
Pediatric Bone Age Prediction Using Deep Learning
cs.LG updates on arXiv.org
What Do They See? Interpreting Complex Road Scenarios Through the Eyes of Vision-Language-Action Models for Safe and Trustworthy Autonomous Vehicle Learning
What Do They See? Interpreting Complex Road Scenarios Through the Eyes of Vision-Language-Action Models for Safe and Trustworthy Autonomous Vehicle Learning
cs.LG updates on arXiv.org
Tight Sample Bounds for Renyi and Min-Entropy Estimation
Tight Sample Bounds for Renyi and Min-Entropy Estimation
cs.LG updates on arXiv.org
Twisted Schr\"odinger Bridge Matching
Twisted Schr\"odinger Bridge Matching
cs.LG updates on arXiv.org
PriorProof: A Point-in-Time Measure of Technique Novelty for Formal Proofs
PriorProof: A Point-in-Time Measure of Technique Novelty for Formal Proofs
cs.LG updates on arXiv.org
Increasing Line Outage Localization Performance with Ensemble Classifiers
Increasing Line Outage Localization Performance with Ensemble Classifiers
cs.LG updates on arXiv.org
WHALE: A Scalable Unified Model for Recommendation with Wukong-HSTU Architecture
WHALE: A Scalable Unified Model for Recommendation with Wukong-HSTU Architecture
cs.LG updates on arXiv.org
Robust Chance-Constrained Optimization using a Continuous Parameter Space Wasserstein-2 Ambiguity Set of Gaussian Mixtures
Robust Chance-Constrained Optimization using a Continuous Parameter Space Wasserstein-2 Ambiguity Set of Gaussian Mixtures
cs.LG updates on arXiv.org
Otap:Structure-Aware Optimal Transport for Evaluating Planning and Execution in Agent Trajectories
Otap:Structure-Aware Optimal Transport for Evaluating Planning and Execution in Agent Trajectories
cs.LG updates on arXiv.org
ThRIve: Thermally Robust CNN Inference via Low-Rank Adaptation in Heterogeneous PIM Architectures
ThRIve: Thermally Robust CNN Inference via Low-Rank Adaptation in Heterogeneous PIM Architectures
cs.LG updates on arXiv.org
A Multi-Model Hybrid Defense Approach Against White-box Adversarial Attacks in Computer Network Traffic
A Multi-Model Hybrid Defense Approach Against White-box Adversarial Attacks in Computer Network Traffic
cs.LG updates on arXiv.org
The Geometry of Semantic Space: A Continuous Geometric Framework for the Transformer Architecture
The Geometry of Semantic Space: A Continuous Geometric Framework for the Transformer Architecture
cs.LG updates on arXiv.org
OrderMoE: An expert similarity driven distributed edge MoE inference
OrderMoE: An expert similarity driven distributed edge MoE inference
cs.LG updates on arXiv.org
Non-Asymptotic Best Policy Identification Guarantees in Online Reinforcement Learning
Non-Asymptotic Best Policy Identification Guarantees in Online Reinforcement Learning
cs.LG updates on arXiv.org
Rate-Distortion-Perception Theory: Redefining the Fundamental Limits of Information Representation
Rate-Distortion-Perception Theory: Redefining the Fundamental Limits of Information Representation
cs.LG updates on arXiv.org
Lossless but Not Free: An Empirical Anatomy of Speculative Decoding on Consumer Hardware
Lossless but Not Free: An Empirical Anatomy of Speculative Decoding on Consumer Hardware
cs.LG updates on arXiv.org
Interpreting Quantum Learning Models via Stochastic Processes
Interpreting Quantum Learning Models via Stochastic Processes
cs.LG updates on arXiv.org
Self-Modifying Lean Proof Agents with Verifier-Grounded Benchmark Coevolution
Self-Modifying Lean Proof Agents with Verifier-Grounded Benchmark Coevolution
cs.LG updates on arXiv.org
Stringological sequence prediction II: Right-to-left automaticity and related complexity measures
Stringological sequence prediction II: Right-to-left automaticity and related complexity measures
cs.LG updates on arXiv.org
Taurus: Accelerating Out-of-Core Graph Neural Network Inference on Billion-Scale Graphs
Taurus: Accelerating Out-of-Core Graph Neural Network Inference on Billion-Scale Graphs
cs.LG updates on arXiv.org
Team DACTYL at PAN 2026: Bayesian Data Mixing and Empirical X-risk Minimization for AI-text Detection
Team DACTYL at PAN 2026: Bayesian Data Mixing and Empirical X-risk Minimization for AI-text Detection
cs.LG updates on arXiv.org
Quantifying Diversity of Thought: A Predictive Law of Weighted LLM Ensemble Lift
Quantifying Diversity of Thought: A Predictive Law of Weighted LLM Ensemble Lift
cs.LG updates on arXiv.org
Kernel Regression with Tensor Trains and Hadamard Overparameterization
Kernel Regression with Tensor Trains and Hadamard Overparameterization
cs.LG updates on arXiv.org
Efficient Sequential Evaluation of Large Language Models
Efficient Sequential Evaluation of Large Language Models
cs.LG updates on arXiv.org
Feature-Guided Diffusion for Non-Differentiable Inverse Rendering
Feature-Guided Diffusion for Non-Differentiable Inverse Rendering
cs.LG updates on arXiv.org
DA-MergeLoRA: Hypernetwork-Based LoRA Merging for Few-Shot Test-Time Domain Adaptation
DA-MergeLoRA: Hypernetwork-Based LoRA Merging for Few-Shot Test-Time Domain Adaptation
cs.LG updates on arXiv.org
The Dimension of Nonterminating Resampling Computations
The Dimension of Nonterminating Resampling Computations
cs.LG updates on arXiv.org
Panache: One-Pass Motif Discovery at Every Window Length
Panache: One-Pass Motif Discovery at Every Window Length
cs.LG updates on arXiv.org
SALT: Salience-Aware Lexical Trie for Long-Context Compression
SALT: Salience-Aware Lexical Trie for Long-Context Compression
cs.LG updates on arXiv.org
Token-Level Off-Policy Learning for Faithful Generation Under Distribution Shift
Token-Level Off-Policy Learning for Faithful Generation Under Distribution Shift
cs.LG updates on arXiv.org
FlowSonic: Stable Zero-Shot Music Editing via High-Order Trajectory Integration
FlowSonic: Stable Zero-Shot Music Editing via High-Order Trajectory Integration
cs.LG updates on arXiv.org
Can AI Agents Really Complete RTL-to-GDS? Lessons from Benchmarking Tool-Interactive EDA Workflows
Can AI Agents Really Complete RTL-to-GDS? Lessons from Benchmarking Tool-Interactive EDA Workflows
cs.LG updates on arXiv.org
The Curvature Shadow: An Apparent Failure of Maximum-Entropy Equilibrium Selection is a Removable Artifact
The Curvature Shadow: An Apparent Failure of Maximum-Entropy Equilibrium Selection is a Removable Artifact
cs.LG updates on arXiv.org
COLIP-2: Olfaction-Vision-Language Embeddings
COLIP-2: Olfaction-Vision-Language Embeddings
cs.LG updates on arXiv.org
ZifaMem: Structured Memory for Persona, Preference, and Emotional Continuity in AI Companions
ZifaMem: Structured Memory for Persona, Preference, and Emotional Continuity in AI Companions
cs.LG updates on arXiv.org
Detection, Attribution, Narration: An End-to-End Pipeline for Explainable Money Mule Identification
Detection, Attribution, Narration: An End-to-End Pipeline for Explainable Money Mule Identification
cs.LG updates on arXiv.org
Semantic Color Naturalness Breaker: Preventing Illegitimate Colorization via Content-Aware Color Priors
Semantic Color Naturalness Breaker: Preventing Illegitimate Colorization via Content-Aware Color Priors
cs.LG updates on arXiv.org
Online learning of neural state-space models
Online learning of neural state-space models
cs.LG updates on arXiv.org
Brain-Aligned Multi-Stream Video Transformers with Sparse Self-Selection
Brain-Aligned Multi-Stream Video Transformers with Sparse Self-Selection
cs.LG updates on arXiv.org
LFM: Leveraging Foundation Models for Source-Free Universal Domain Adaptation
LFM: Leveraging Foundation Models for Source-Free Universal Domain Adaptation
cs.LG updates on arXiv.org
An efficient adaptive dimension selection algorithm for multidimensional probit graded response models
An efficient adaptive dimension selection algorithm for multidimensional probit graded response models
cs.LG updates on arXiv.org
Early Yield Prediction for Sugar Beet Fields using Satellite Data -- Learnings from Specialized Vision Transformers
Early Yield Prediction for Sugar Beet Fields using Satellite Data -- Learnings from Specialized Vision Transformers
cs.LG updates on arXiv.org
Equality, Equity, and Causality in Fairness Research: A Commentary on Cheng (2026)
Equality, Equity, and Causality in Fairness Research: A Commentary on Cheng (2026)
cs.LG updates on arXiv.org
An Adjoint-Sensitivity Framework for Lost-in-the-Middle Phenomena in Causal Residual Transformers
An Adjoint-Sensitivity Framework for Lost-in-the-Middle Phenomena in Causal Residual Transformers
cs.LG updates on arXiv.org
Measuring Monosemanticity in Sparse Autoencoders via Latent Activation Coherence
Measuring Monosemanticity in Sparse Autoencoders via Latent Activation Coherence
cs.LG updates on arXiv.org
ETAS: An Effect-Typed Language for Agent Systems
ETAS: An Effect-Typed Language for Agent Systems
cs.LG updates on arXiv.org
Reasoning as a Double-Edged Sword: Architecture and Cross-Stage Robustness in Vision-Language-Action Models
Reasoning as a Double-Edged Sword: Architecture and Cross-Stage Robustness in Vision-Language-Action Models
cs.LG updates on arXiv.org
Organization of computation in reservoir computing
Organization of computation in reservoir computing
cs.LG updates on arXiv.org
Entanglement geometry separates circuit cutting, classical hardness, and trainability
Entanglement geometry separates circuit cutting, classical hardness, and trainability
cs.LG updates on arXiv.org
Beyond the Edge of Chaos: Stability-Expressivity Transfer in Reservoir Forecasting
Beyond the Edge of Chaos: Stability-Expressivity Transfer in Reservoir Forecasting
cs.LG updates on arXiv.org
Chemical filters for ultra-high-throughput materials screening and generation
Chemical filters for ultra-high-throughput materials screening and generation
cs.LG updates on arXiv.org
AutoEncoder-Compressed Parallel Split Learning for Pre-trained Model Fine-Tuning
AutoEncoder-Compressed Parallel Split Learning for Pre-trained Model Fine-Tuning
cs.LG updates on arXiv.org
Value-Aware Prediction for Robust Multi-Agent Coordination Under Communication Loss
Value-Aware Prediction for Robust Multi-Agent Coordination Under Communication Loss
cs.LG updates on arXiv.org
PRIME: Plasticity Recovery in Multi-Agent Environments for UAV-Assisted Emergency Communication Networks
PRIME: Plasticity Recovery in Multi-Agent Environments for UAV-Assisted Emergency Communication Networks
cs.LG updates on arXiv.org
Aggregate in the Advantage, Not the Ratio: A Canonical-Form Analysis of Cooperative Multi-Agent Policy Optimization
Aggregate in the Advantage, Not the Ratio: A Canonical-Form Analysis of Cooperative Multi-Agent Policy Optimization
cs.LG updates on arXiv.org
PAMD: Structured Adaptive Distances for Bisimulation Representations in Visual Reinforcement Learning
PAMD: Structured Adaptive Distances for Bisimulation Representations in Visual Reinforcement Learning
cs.LG updates on arXiv.org
Remote Awareness of Seafloor Images Collected by AUVs over Low-Bandwidth Communication Links
Remote Awareness of Seafloor Images Collected by AUVs over Low-Bandwidth Communication Links
cs.LG updates on arXiv.org
Adaptive Adversaries: A Multi-Turn, Multi-LLM Benchmark for LLM Agent Security
Adaptive Adversaries: A Multi-Turn, Multi-LLM Benchmark for LLM Agent Security
cs.LG updates on arXiv.org
Hardware Mechanisms to Dynamically Throttle AI Performance
Hardware Mechanisms to Dynamically Throttle AI Performance
cs.LG updates on arXiv.org
WorldCupArena: Fine-Grained Evaluation of Language Models and Deep-Research Agents on Football Forecasting
WorldCupArena: Fine-Grained Evaluation of Language Models and Deep-Research Agents on Football Forecasting
cs.LG updates on arXiv.org
SciForma: Structure-Faithful Generation of Scientific Diagrams
SciForma: Structure-Faithful Generation of Scientific Diagrams
cs.LG updates on arXiv.org
How Does Alignment Tuning Shape Representations of Sycophancy and Related Cue-Induced Biases in LLMs?
How Does Alignment Tuning Shape Representations of Sycophancy and Related Cue-Induced Biases in LLMs?
cs.LG updates on arXiv.org
COVAriance-Induced Fairness Gap Penalty for Subgroup-Fair Clustering
COVAriance-Induced Fairness Gap Penalty for Subgroup-Fair Clustering
cs.LG updates on arXiv.org
ClouDens: Operational Context-Aware Anomaly Detection for Large-scale Cloud System Monitoring
ClouDens: Operational Context-Aware Anomaly Detection for Large-scale Cloud System Monitoring
cs.LG updates on arXiv.org
EVOLVE: Efficient Learned Volume Compression with Variable-Rate Encoding on a Cross-Domain Database
EVOLVE: Efficient Learned Volume Compression with Variable-Rate Encoding on a Cross-Domain Database
cs.LG updates on arXiv.org
Certified Training for Convolutional Perturbations
Certified Training for Convolutional Perturbations
cs.LG updates on arXiv.org
PPL-Factory: Task-Aware and Budget-Aware Data Selection from Language Modeling to Reasoning
PPL-Factory: Task-Aware and Budget-Aware Data Selection from Language Modeling to Reasoning
cs.LG updates on arXiv.org
Unveiling Invariant and Transferable Latent Factors Across Heterogeneous Environments via ATLAS
Unveiling Invariant and Transferable Latent Factors Across Heterogeneous Environments via ATLAS
cs.LG updates on arXiv.org
Vector Search As Nearest Neighbor Matching: RAG-based Policy Learning in Causal Inference
Vector Search As Nearest Neighbor Matching: RAG-based Policy Learning in Causal Inference
cs.LG updates on arXiv.org
Patch Policy: Efficient Embodied Control via Dense Visual Representations
Patch Policy: Efficient Embodied Control via Dense Visual Representations
cs.LG updates on arXiv.org
The Many Senses of Visual Similarity: A Text-Prompted Image Perceptual Metric
The Many Senses of Visual Similarity: A Text-Prompted Image Perceptual Metric
cs.LG updates on arXiv.org
Hierarchical Reinforcement Learning for Air Combat at DARPA's AlphaDogfight Trials
Hierarchical Reinforcement Learning for Air Combat at DARPA's AlphaDogfight Trials
cs.LG updates on arXiv.org
Ghosts in Neural Networks: Existence, Structure and Role of Infinite-Dimensional Null Space
Ghosts in Neural Networks: Existence, Structure and Role of Infinite-Dimensional Null Space
cs.LG updates on arXiv.org
Automated Reinforcement Learning: An Overview
Automated Reinforcement Learning: An Overview
cs.LG updates on arXiv.org
Posts of Peril: Detecting Information About Hazards in Text
Posts of Peril: Detecting Information About Hazards in Text
cs.LG updates on arXiv.org
A Survey of Features Used for Representing Black-box Single-objective Continuous Optimization
A Survey of Features Used for Representing Black-box Single-objective Continuous Optimization
cs.LG updates on arXiv.org
Regret Optimality of Sample Average Approximation for Data-Driven Newsvendor Problems: A General Optimization Perspective
Regret Optimality of Sample Average Approximation for Data-Driven Newsvendor Problems: A General Optimization Perspective
cs.LG updates on arXiv.org
Online-Score-Aided Federated Learning for Resource-Constrained Wireless Clients with Continual Data Arrival
Online-Score-Aided Federated Learning for Resource-Constrained Wireless Clients with Continual Data Arrival
cs.LG updates on arXiv.org
Distributed solar generation forecasting using attention-based deep neural networks for cloud movement prediction
Distributed solar generation forecasting using attention-based deep neural networks for cloud movement prediction
cs.LG updates on arXiv.org
BADTV: Unveiling Backdoor Threats in Third-Party Task Vectors
BADTV: Unveiling Backdoor Threats in Third-Party Task Vectors
cs.LG updates on arXiv.org
Mediator: Memory-efficient LLM Merging with Less Parameter Conflicts and Uncertainty Based Routing
Mediator: Memory-efficient LLM Merging with Less Parameter Conflicts and Uncertainty Based Routing
cs.LG updates on arXiv.org
PLayer-FL: A Principled Approach to Personalized Layer-wise Cross-Silo Federated Learning
PLayer-FL: A Principled Approach to Personalized Layer-wise Cross-Silo Federated Learning
cs.LG updates on arXiv.org
AdaGC: Enhancing LLM Pretraining Stability via Adaptive Gradient Clipping
AdaGC: Enhancing LLM Pretraining Stability via Adaptive Gradient Clipping
cs.LG updates on arXiv.org
On the Interpolation Effect of Score Smoothing in Diffusion Models
On the Interpolation Effect of Score Smoothing in Diffusion Models
cs.LG updates on arXiv.org
ExMAG: Learning of Maximally Ancestral Graphs
ExMAG: Learning of Maximally Ancestral Graphs
cs.LG updates on arXiv.org
Potential failures of physics-informed machine learning in traffic flow modeling: theoretical and experimental analysis
Potential failures of physics-informed machine learning in traffic flow modeling: theoretical and experimental analysis
cs.LG updates on arXiv.org
Prismatic Synthesis: Gradient-based Data Diversification Boosts Generalization in LLM Reasoning
Prismatic Synthesis: Gradient-based Data Diversification Boosts Generalization in LLM Reasoning
cs.LG updates on arXiv.org
Sign-SZPO: Provable Preference-based Reinforcement Learning with an Unknown Link Function
Sign-SZPO: Provable Preference-based Reinforcement Learning with an Unknown Link Function
cs.LG updates on arXiv.org
Can Interpretation Predict Behavior on Unseen Data?
Can Interpretation Predict Behavior on Unseen Data?
cs.LG updates on arXiv.org
Attentions Under the Microscope: A Comparative Study of Resource Utilization for Variants of Self-Attention
Attentions Under the Microscope: A Comparative Study of Resource Utilization for Variants of Self-Attention
cs.LG updates on arXiv.org
Symmetric Behavior Regularized Policy Optimization
Symmetric Behavior Regularized Policy Optimization
cs.LG updates on arXiv.org
Geometric origin of adversarial vulnerability in deep learning
Geometric origin of adversarial vulnerability in deep learning
cs.LG updates on arXiv.org
NIRVANA: Structured Pruning Reimagined for Large Language Model Compression
NIRVANA: Structured Pruning Reimagined for Large Language Model Compression
cs.LG updates on arXiv.org
Multivariate Time Series Forecasting with Gate-Based Quantum Reservoir Computing on NISQ Hardware
Multivariate Time Series Forecasting with Gate-Based Quantum Reservoir Computing on NISQ Hardware
cs.LG updates on arXiv.org
Interpret Policies in Deep Reinforcement Learning using SILVER with RL-Guided Labeling: A Model-level Approach to High-dimensional and Multi-action Environments
Interpret Policies in Deep Reinforcement Learning using SILVER with RL-Guided Labeling: A Model-level Approach to High-dimensional and Multi-action Environments
cs.LG updates on arXiv.org
When Fewer Layers Break More Chains: Layer Pruning Harms Test-Time Scaling in LLMs
When Fewer Layers Break More Chains: Layer Pruning Harms Test-Time Scaling in LLMs
cs.LG updates on arXiv.org
BBOPlace-Bench: Benchmarking Black-Box Optimization for Chip Placement
BBOPlace-Bench: Benchmarking Black-Box Optimization for Chip Placement
cs.LG updates on arXiv.org
InertialAR: Autoregressive 3D Molecule Generation with Inertial Frames
InertialAR: Autoregressive 3D Molecule Generation with Inertial Frames
cs.LG updates on arXiv.org
ProtoTSNet: Interpretable Multivariate Time Series Classification With Prototypical Parts
ProtoTSNet: Interpretable Multivariate Time Series Classification With Prototypical Parts
cs.LG updates on arXiv.org
Financial Management System for SMEs: Real-World Deployment of Accounts Receivable and Cash Flow Prediction
Financial Management System for SMEs: Real-World Deployment of Accounts Receivable and Cash Flow Prediction
cs.LG updates on arXiv.org
ProDER: A Continual Learning Approach for Fault Prediction in Evolving Smart Grids
ProDER: A Continual Learning Approach for Fault Prediction in Evolving Smart Grids
cs.LG updates on arXiv.org
mHC-GNN: Manifold-Constrained Hyper-Connections for Graph Neural Networks
mHC-GNN: Manifold-Constrained Hyper-Connections for Graph Neural Networks
cs.LG updates on arXiv.org
GlyRAG: Context-Aware Retrieval-Augmented Framework for Blood Glucose Forecasting
GlyRAG: Context-Aware Retrieval-Augmented Framework for Blood Glucose Forecasting
cs.LG updates on arXiv.org
Sparsity-Aware Low-Rank Representation for Efficient Fine-Tuning of Large Language Models
Sparsity-Aware Low-Rank Representation for Efficient Fine-Tuning of Large Language Models
cs.LG updates on arXiv.org
Hybrid Mamba-Attention Neural Architecture for Channel Estimation
Hybrid Mamba-Attention Neural Architecture for Channel Estimation