NVIDIA、AI工場向け蓄電池の設計指針を提示

AdaSR: Adaptive Streaming Reasoning with Hierarchical Relative Policy Optimization

arXiv cs.CL (Computation and Language) ・ 2026-06-12

AdaSR、入力を逐次処理する「ストリーミング推論」を提案

CORA: Analyzing and bridging thinking-answer gap in Multimodal RLVR via Consistency-Oriented Reasoning Alignment

arXiv cs.CL (Computation and Language) ・ 2026-06-12

CORA、マルチモーダルRLVRの「思考と回答のずれ」を是正

Flood and Harvest: The Provable Necessity of Trivia for Generating Valuable Mathematics via the Lens of Language Generation in the Limit

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-12

価値ある数学生成に「雑多な命題」が不可欠であることを理論的に論証

Beyond task performance: Decoding bioacoustic embeddings with speech features

arXiv cs.LG (Machine Learning) ・ 2026-06-12

生物音響の埋め込みを音声特徴で解読し透明性を検証

Graph Structured Combinatorial Semi-Bandit with Nonlinear Reward Associations through Separable Signals

arXiv cs.LG (Machine Learning) ・ 2026-06-12

グラフ構造の組合せ半バンディットを分離信号で解く

Which Directions Matter? Sparse Design for Affine Robust Optimization

arXiv cs.LG (Machine Learning) ・ 2026-06-12

アフィンロバスト最適化で重要な不確実性方向を疎に選択

Graph Diffusion Residuals for Control-Function Instrumental Variables

arXiv cs.LG (Machine Learning) ・ 2026-06-12

制御関数IV推定にグラフ拡散の残差を活用

Characterizing Cultural Localization in AI-Generated Stories

arXiv cs.CL (Computation and Language) ・ 2026-06-12

AI生成の物語における文化的ローカライズを分析

Regulating the Machine Contributor: Governance and Policy Alignment in Open Source

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-12

オープンソースへのAIエージェント貢献を統治する枠組みを論じる

AudioDER: A Deduplication-Enhanced Reasoning Dataset for Post-Training Large Audio-Language Models

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-12

音声言語モデルの事後学習向け推論データセット「AudioDER」

SIMMER: Benchmarking Latent Failures in LLM Executable Planning with a World Model

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-12

世界モデルでLLM実行計画の潜在的失敗を測る「SIMMER」

ORCA: A Platform for Open-Source Dexterity Research

arXiv cs.LG (Machine Learning) ・ 2026-06-12

ORCA、オープンソースの器用さ研究プラットフォーム

Rethinking Global Average Pooling: Your Classifier Is Secretly a Multi-Instance Learner

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-12

大域平均プーリングの分類器は実は多重インスタンス学習器と指摘

Provably Safe, Yet Scalable Reinforcement Learning

arXiv cs.LG (Machine Learning) ・ 2026-06-12

証明可能な安全性とスケール性を両立する強化学習

From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-12

LLMが「チャットボット」から永続自律AIへ移行する転換を概念化

The Perceived Fragility of Explanations in Audio Models: Manipulation of Attribution with Unchanged Predictions

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-12

音声モデルの説明は予測を変えず操作できるその脆弱性を検証

Learning to Hear Hesitation: Continual Learning for Disfluency-Aware ASR

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-12

言いよどみを聞き取る継続学習でASRを頑健化

A Low-Rank Subspace Analysis of LLM Interventions

arXiv cs.LG (Machine Learning) ・ 2026-06-12

LLM介入を低ランク部分空間で分析、副作用を抑制

Discovery under Hypothesis Redundancy: A Geometric Theory of Discovery Bottlenecks

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-12

仮説の冗長性で生じる発見の頭打ちを幾何学的に理論化

Can Deep Neural Networks Improve Compression of Very Large Scientific Data?

arXiv cs.LG (Machine Learning) ・ 2026-06-12

深層NNで大規模科学データの誤差有界圧縮を改善できるか

Achieving Precise Text-To-Cypher Via Grounded Knowledge Graph Data Generation

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-12

知識グラフ接地のデータ生成で高精度Text-to-Cypherを実現

ScoreGate: Adaptive Chunk Selection for Retrieval-Augmented Generation via Dual-Score Statistical Fusion

arXiv cs.CL (Computation and Language) ・ 2026-06-12

ScoreGate、RAGの取得チャンク数を適応的に選択

Decoupled Mixture-of-Experts for Parametric Knowledge Injection

arXiv cs.CL (Computation and Language) ・ 2026-06-12

分離型MoEでパラメトリックな知識注入を改善

Spatio-Temporal Audio Language Modeling for Dynamic Sound Sources

arXiv cs.CL (Computation and Language) ・ 2026-06-12

動的な音源の時空間を扱う音声言語モデル

Knowledge Graph Enhanced Memory-Augmented Retrieval for Long Context Modeling

arXiv cs.CL (Computation and Language) ・ 2026-06-12

知識グラフ強化の記憶検索で長文脈を一貫処理

The Holistic Storage of Verb+Up Phrases in Text-based and Audio-based Language Models

arXiv cs.CL (Computation and Language) ・ 2026-06-12

句動詞の全体的記憶を文字・音声モデルで分析

EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments

arXiv cs.CL (Computation and Language) ・ 2026-06-11

動的環境でのLLMエージェント評価基盤EvoArena、パッチ型記憶EvoMemも提案

Learning to Reason by Analogy via Retrieval-Augmented Reinforcement Fine-Tuning

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

類推による推論をRAGとRL微調整で学習するRA-RFT

Influcoder: Distilling Decoders' Gradient Influence Rankings into an Encoder for Data Attribution

arXiv cs.CL (Computation and Language) ・ 2026-06-11

Influcoder、勾配影響ランキングを蒸留しデータ帰属を高速化

HyperTool: Beyond Step-Wise Tool Calls for Tool-Augmented Agents

arXiv cs.CL (Computation and Language) ・ 2026-06-11

HyperTool、ツール呼び出しをコードに畳み込みMCP精度を15.7%から35.3%へ

Operadic consistency: a label-free signal for compositional reasoning failures in LLMs

arXiv cs.CL (Computation and Language) ・ 2026-06-11

LLMの合成推論の失敗をラベル不要で検出する「オペラド整合性」を提案

SkMTEB: Slovak Massive Text Embedding Benchmark and Model Adaptation

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

SkMTEB、スロバキア語の埋め込みベンチと小型モデル公開

From Tokens to Faces: Investigating Discrete Speech Representations for 3D Facial Animation

arXiv cs.CL (Computation and Language) ・ 2026-06-11

音声駆動3D顔アニメに最適な音声表現を比較、音素情報が鍵と判明

AgentBeats: Agentifying Agent Assessment for Openness, Standardization, and Reproducibility

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

AgentBeats、judge agentで評価を標準化するAAAを提唱

Distribution-Agnostic Robust Trajectory Optimization via Chance-Constrained Reinforcement Learning

arXiv cs.LG (Machine Learning) ・ 2026-06-11

機会制約RLで分布非依存のロバスト軌道最適化、地球-火星遷移で検証

Beyond the Commitment Boundary: Probing Epiphenomenal Chain-of-Thought in Large Reasoning Models

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

思考連鎖に『コミットメント境界』、以降の手順は無影響と判明

Reward Modeling for Multi-Agent Orchestration

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

OrchRM、多agent統括の品質を自己教師で評価し効率10倍

EvTexture++: Event-Driven Texture Enhancement for Video Super-Resolution

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

EvTexture++、イベント信号で動画超解像のテクスチャ強化

Uncertainty-Aware Hybrid Retrieval for Long-Document RAG

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

UMG-RAG、粒度を信頼度推定として扱う訓練不要のRAG検索

AgentRivet: an automated system for producing Rivet routines from journal publications

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

AgentRivet、論文から欠落Rivetルーチンを自動生成するLLM系

Graphical Causal Reasoning for Root Cause Analysis in Cloud Networks

arXiv cs.LG (Machine Learning) ・ 2026-06-11

クラウド網障害の根本原因分析、因果グラフ探索で85.7%の再現率

Measurement-Calibrated Multi-Camera Fusion for Vision-Based Indoor Localization

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

単眼誤差の定量化で多カメラ融合を校正する屋内測位手法

Leveraging Audio-LLMs to Filter Speech-to-Speech Training Data

arXiv cs.CL (Computation and Language) ・ 2026-06-11

音声LLMで音声翻訳の学習データを選別、ASR-BLEU最大+1.4

Heterogeneous LiDAR Early Fusion and Learned Re-Ranking Strategy for Robust Long-Term Place Recognition in Unstructured Environments

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

異種LiDAR早期融合と再ランクで非構造環境の場所認識を頑健化

Ontology Memory-Augmented ASR Correction for Long Text-Speech Interleaved Conversations

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

オントロジー記憶で長い対話のASR誤り訂正を文脈接地

Clustering Node Attributed Networks with Graph Neural Networks and Self Learning

arXiv cs.LG (Machine Learning) ・ 2026-06-11

GNNと自己学習の反復でノード属性グラフをクラスタリング

Mod-Guide: An LLM-based Content Moderation Feedback System to Address Insensitive Speech toward Indigenous Ethnic and Religious Minority Communities

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

Mod-Guide、少数派への無神経表現を扱うLLM節度支援系

SmartFont: Dynamic Condition Allocation for Few-Shot Font Generation

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

SmartFont、大域と局所条件を多段配分する少数事例フォント生成

An LLM System for Autonomous Variational Quantum Circuit Design

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-11

LLMで量子回路を自律設計する閉ループagent枠組み

From Passive Generation to Investigation: A Proactive Scientific Peer Review Agent

arXiv cs.CL (Computation and Language) ・ 2026-06-11

能動的に論文を調査する査読エージェントProReviewer、5品質次元で最高評価

Enhanced Low-Density Region Exploration in Classifier-Guided Diffusion Models Through Modified Reverse Diffusion Sampling

arXiv cs.LG (Machine Learning) ・ 2026-06-11

追加学習なしで拡散モデルの低密度領域を探索、希少サンプル網羅を改善

Navigating the Safety-Fidelity Trade-off: Massive-Variate Time Series Forecasting for Power Systems via Probabilistic Scenarios

arXiv cs.LG (Machine Learning) ・ 2026-06-11

電力系統向け確率予測ベンチマークPowerPhase、最大3.7万チャネル

SkillCAT: Contrastive Assessment and Topology-Aware Skill Self-Evolution for LLM Agents

arXiv cs.CL (Computation and Language) ・ 2026-06-11

SkillCAT、LLMエージェントのスキル自己進化を検証付き3段階に分離

Physics-Guided Spatiotemporal Learning for Coastal Wave Peak Period Estimation from Video

arXiv cs.LG (Machine Learning) ・ 2026-06-11

物理誘導の時空間学習で沿岸映像から波のピーク周期を推定

Clipping Makes Distributed and Federated Asynchronous SGD Robust to Stragglers

arXiv cs.LG (Machine Learning) ・ 2026-06-11

勾配クリッピングが非同期SGDを遅延耐性にすると理論的に正当化

TimeLens: On-Device Artifact Recognition with Retrieval-Augmented Question Answering for the Grand Egyptian Museum

arXiv cs.CL (Computation and Language) ・ 2026-06-11

TimeLens、大エジプト博物館向けオンデバイス文化財認識とRAG型QA

ComAct: Reframing Professional Software Manipulation via COM-as-Action Paradigm

arXiv cs.CL (Computation and Language) ・ 2026-06-11

ComAct、COM経由の決定論的操作で産業CADソフトをエージェント制御

PolyAlign: Conditional Human-Distribution Alignment

arXiv cs.CL (Computation and Language) ・ 2026-06-11

PolyAlign、文脈に応じた人間の応答分布へLLMを整合

SICI: A Semantic-Pragmatic Complexity Index Reveals Regime Shifts in LLM Stance Detection

arXiv cs.CL (Computation and Language) ・ 2026-06-11

意味・語用論的複雑性指標SICI、LLM立場検出のレジーム変化を解明

MemRefine: LLM-Guided Compression for Long-Term Agent Memory

arXiv cs.CL (Computation and Language) ・ 2026-06-11

MemRefine、LLM判断で予算内にエージェント長期記憶を圧縮

NTS-CoT: Mitigating Hallucinations in LLM-based News Timeline Summarization with Chain-of-Thought Reasoning

arXiv cs.CL (Computation and Language) ・ 2026-06-11

NTS-CoT、思考連鎖でニュース年表要約の幻覚を抑制

Reroute, Don't Remove: Recoverable Visual Token Routing for Vision-Language Models

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-10

Reroute、視覚トークンを除去でなく再ルーティングしVLM推論を効率化

Context-Driven Incremental Compression for Multi-Turn Dialogue Generation

arXiv cs.CL (Computation and Language) ・ 2026-06-10

C-DIC、多ターン対話の文脈を逐次圧縮し長対話を安定化

DIRECT: When and Where Should You Allocate Test-Time Compute in Embodied Planners?

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-10

DIRECT、身体エージェントのテスト時計算をプロンプト別に配分

Which Models Are Our Models Built On? Auditing Invisible Dependencies in Modern LLMs

arXiv cs.CL (Computation and Language) ・ 2026-06-10

ModSleuth、公開成果物からLLMの依存グラフを再構成し監査

Verifiable Environments Are LEGO Bricks: Recursive Composition for Reasoning Generalization

arXiv cs.CL (Computation and Language) ・ 2026-06-10

RACES、検証可能環境を再帰合成しRL推論の汎化を拡張

Adjoint Method versus Physics-Informed Neural Networks in PDE-Constrained Inverse Problems

arXiv cs.LG (Machine Learning) ・ 2026-06-10

PDE 制約逆問題で随伴法と PINN を同条件で公平比較

Fourier Features Let Agents Learn High Precision Policies with Imitation Learning

arXiv cs.LG (Machine Learning) ・ 2026-06-10

Fourier 特徴で点群方策が高精度ロボット操作を獲得

PROJECTMEM: A Local-First, Event-Sourced Memory and Judgment Layer for AI Coding Agents

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-10

projectmem、コーディングエージェントに局所優先の記憶層を追加

Measuring Epistemic Resilience of LLMs Under Misleading Medical Context

arXiv cs.CL (Computation and Language) ・ 2026-06-10

MedMisBench、誤誘導文脈下でのLLM医療判断の脆弱性を測定

The Standard Interpretable Model: A general theory of interpretable machine learning to deductively design interpretable methods using Lagrangian mechanics

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-10

SIM、ラグランジュ力学で解釈可能手法を演繹的に設計する理論

Holding the FP8 Quality Ceiling at 8-Bit Weights and Activations: INT8 and GGUF Post-Training Quantization of Ideogram 4.0 for Consumer GPUs

arXiv cs.LG (Machine Learning) ・ 2026-06-10

Ideogram 4.0 を INT8 量子化、民生 GPU で FP8 品質を維持

DiffCold: A Diffusion-based Generative Model for Cold-Start Item Recommendation

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-10

DiffCold、拡散生成でコールドスタート推薦のシーソー難を解消

Re-evaluating Confidence Remasking in Masked Diffusion Language Models

arXiv cs.LG (Machine Learning) ・ 2026-06-10

拡散言語モデルの remasking 手法 WINO、再評価で効果は限定的

MLT-Dedup: Efficient Large-Scale Online Video Deduplication via Multi-Level Representations and Spatial-Temporal Matching

arXiv cs.LG (Machine Learning) ・ 2026-06-10

MLT-Dedup、多層表現で大規模オンライン動画の重複排除を効率化

OpenMedReason: Scientific Reasoning Supervision for Medical Vision-Language Models

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-10

OpenMedReason、約45万件の医療多モーダル推論コーパスを公開

A Controlled Study of Decoding-Time Truthfulness Methods on Instruction-Tuned LLMs

arXiv cs.CL (Computation and Language) ・ 2026-06-10

CHAIR、層ごとのlogitから幻覚を検出する教師あり手法

nD-RoPE: A Generalized RoPE for n-Dimensional Position Embedding

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-10

nD-RoPE、n次元位置埋め込みへRoPEを分解なく一般化

Augmenting Molecular Language Models with Local $n$-gram Memory

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-10

MolGram、局所n-gram記憶で分子言語モデルを効率改善

Bridging the Morphology Gap: Adapting VLA Models to Dexterous Manipulation via Intent-Conditioned Fine-Tuning

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-10

InDex、意味継承でVLAモデルを器用な多関節ハンドへ適応

DAM-VLA: Decoupled Asynchronous Multimodal Vision Language Action model

arXiv cs.LG (Machine Learning) ・ 2026-06-10

DAM-VLA、モダリティ別の非同期処理で実機操作成功率を倍増

FORT-Searcher: Synthesizing Shortcut-Resistant Search Tasks for Training Deep Search Agents

arXiv cs.CL (Computation and Language) ・ 2026-06-10

FORT、近道耐性のある探索課題を合成し深層探索エージェントを訓練

On the Limits of LLM-as-Judge for Scientific Novelty Assessment

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-10

LLM審判の科学的新規性評価の限界、RQ-Benchで検証

Simplicity Suffices for Parameter Noise Injection in Stochastic Gradient Descent

arXiv cs.LG (Machine Learning) ・ 2026-06-10

SGD へのノイズ注入、単純な等方性手法で十分と実証

Decoding Multimodal Cues: Unveiling the Implicit Meaning Behind Hateful Videos

arXiv cs.CL (Computation and Language) ・ 2026-06-10

IARE、根拠付きで有害動画を説明可能に検出する枠組み

uva-irlab-conv at SemEval-2026 Task 8: Multi-Turn RAG with Learned Sparse Retrieval and Listwise Reranking

arXiv cs.CL (Computation and Language) ・ 2026-06-10

SemEval-2026 Task 8、疎検索とリスト再順位で多ターンRAG

Toward Generalist Autonomous Research via Hypothesis-Tree Refinement

arXiv cs.CL (Computation and Language) ・ 2026-06-10

Arbor、仮説ツリー改良で長期間の自律研究ループを運用

An Ontology-Guided Multi-Anchor Graph Retrieval Framework for Traffic Legal Liability Determination

arXiv cs.CL (Computation and Language) ・ 2026-06-10

OMAGR、オントロジー誘導の多アンカーグラフ検索で交通法責任を判定

Grammar-Constrained Decoding Can Jailbreak LLMs into Generating Malicious Code

arXiv cs.CL (Computation and Language) ・ 2026-06-10

CodeSpear、文法制約デコードを悪用しLLMに悪性コードを生成させる

Lius: Translation Model Based Instructional Lingustic Using Continual Instruction Tuning In Kupang Malay

arXiv cs.CL (Computation and Language) ・ 2026-06-10

Lius、継続的指示調整で低資源クパン・マレー語の翻訳を改善

EEVEE: Towards Test-time Prompt Learning in the Real World for Self-Improving Agents

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

EEVEE、複数データセット対応のテスト時プロンプト学習を実現

Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories

arXiv cs.CL (Computation and Language) ・ 2026-06-09

Data2Story、データから検証可能なマルチモーダル記事を自動生成

Flaws in the LLM Automation Narrative

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

LLMの専門家代替論に疑義、人間専門家が平均で上回ると報告

ABC-Bench: An Agentic Bio-Capabilities Benchmark for Biosecurity

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

ABC-Bench、LLMエージェントのバイオセキュリティ関連能力を評価

First-Order Trajectory Matching: Fast Ensemble Predictions of Chaotic, Turbulent, Stochastic Systems

arXiv cs.LG (Machine Learning) ・ 2026-06-09

FTM、軌跡から確率流速を学習し安価なアンサンブル予測

DMT: Demographic Conditioning, Morphology-Enhanced Transformer for Cuffless Blood Pressure Estimation from PPG Signals

arXiv cs.LG (Machine Learning) ・ 2026-06-09

DMT、PPGから人口統計条件付きTransformerでカフレス血圧推定

TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

TRACE、エージェントRLのロールアウト予算をツリー状に配分

Towards Autonomous Accelerator Design: FPGA Accelerator Generation with SECDA

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

SECDA-DSE、LLMでFPGAアクセラレータの設計空間探索を誘導

PhantomBench: Benchmarking the Non-existential Threat of Language Models

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

PhantomBench、存在しない概念へのLLMのハルシネーションを評価

RoboNaldo: Accurate, Stable and Powerful Humanoid Soccer Shooting via Motion-Guided Curriculum Reinforcement Learning

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

RoboNaldo、動作誘導カリキュラムRLでヒューマノイドの強シュート

VISTA: A Versatile Interactive User Simulation Toolkit for Agent Evaluation

arXiv cs.CL (Computation and Language) ・ 2026-06-09

VISTA、エージェント評価向けの多用途ユーザシミュレーション基盤

A History-Aware Visually Grounded Critic for Computer Use Agents

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

HiViG、計算機操作エージェント向けの履歴認識・視覚接地批評

T1-Bench: Benchmarking Multi-Scenario Agents in Real-World Domains

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

T1-Bench、25ドメインの多シナリオエージェントを高忠実評価

Generative Archetype-Grounded Item Representations for Sequential Recommendation

arXiv cs.CL (Computation and Language) ・ 2026-06-09

GenAIR、対象顧客像で接地した生成的アイテム表現で推薦改善

Generative Explainability for Next-Generation Networks: LLM-Augmented XAI with Mutual Feature Interactions

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

LLM拡張XAIで次世代ネットワークの説明を自然言語生成

Democratising Camera Trap AI: An Open-Source Model for Detecting UK Mammals

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

英国哺乳類検出の開源モデルを公開、mAP 0.984を達成

Trace Only What You Need: Structure-Aware On-Demand Hypergraph Memory for Long-Document Question Answering

arXiv cs.CL (Computation and Language) ・ 2026-06-09

DocTrace、長文書QA向けの構造認識・オンデマンド超グラフ記憶

Task Robustness via Re-Labelling Vision-Action Robot Data

arXiv cs.LG (Machine Learning) ・ 2026-06-09

TREAD、VLMでロボットの視覚行動データを再ラベルし頑健化

Role-Agent: Bootstrapping LLM Agents via Dual-Role Evolution

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

Role-Agent、単一LLMがエージェントと環境を兼ね共進化

Range Penalization: Theoretical Insights with Applications in Federated Learning

arXiv cs.LG (Machine Learning) ・ 2026-06-09

範囲正則化、連合学習で量子化に資する重みの極性クラスタ化

Ethical and Technical Limits of Deepfake Speech Datasets

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

ディープフェイク音声39データセットを監査、公平性評価は困難

Pose-ICL: 3D-Aware In-Context Learning for Pose-Controllable Subject Customization

arXiv cs.AI (Artificial Intelligence) ・ 2026-06-09

Pose-ICL、3D認識の文脈内学習でポーズ制御可能な被写体生成

Janus: A Benchmark for Goal-Conditioned Information Distortion in LLMs

arXiv cs.CL (Computation and Language) ・ 2026-06-09

JANUS、真実の事実を選択的に扱うLLMの情報歪曲を測定

Attention-Discounted Adaptive Sampler for Masked Diffusion Language Models

arXiv cs.CL (Computation and Language) ・ 2026-06-09

ADAS、マスク拡散言語モデルの並列復号を再ランクで安全化

Recovering the Zipfian Distribution in Unsupervised Term Discovery

arXiv cs.CL (Computation and Language) ・ 2026-06-09

教師なし語彙発見、グラフクラスタリングがK-means等を上回る

N-GRPO: Embedding-Level Neighbor Mixing for Enhanced Policy Optimization

arXiv cs.CL (Computation and Language) ・ 2026-06-09

N-GRPO提案、埋め込み近傍の混合でGRPOの探索を強化

Infini Memory: Maintainable Topic Documents for Long-Term LLM Agent Memory

arXiv cs.CL (Computation and Language) ・ 2026-06-09

Infini Memory、トピック文書でエージェントの長期記憶を保守

How Does Reasoning Flow? Tracing Attention-Induced Information Flow for Targeted RL in LLMs

arXiv cs.CL (Computation and Language) ・ 2026-06-09

FlowTracer、attention情報流でRLのトークン信用割当を精緻化

ParaBridge: Bridging Paralinguistic Perception and Dialogue Behavior in Speech Language Models

arXiv cs.CL (Computation and Language) ・ 2026-06-09

ParaBridge、音声LLMのパラ言語知覚と対話行動の溝を埋める