「RAG」の検索結果
307 件A growing body of work establishes that large language models are not mere statistical memorizers, but are cap
For a finite set $O$ of Boolean functions, we consider the class of propositional formulas built using the fun
Graph-based surrogate models offer a promising route to accelerate computational fluid dynamics (CFD) simulati
Sparsity is a powerful structural resource in optimization and statistics. We develop frameworks for leveragin
Recent advancements in transformer length generalization theory enable us to reliably predict when a transform
When an external reference set (an anchor) is used to decompose an LLM-judge panel's error into a quality sign
Quantization schemes based on randomized rotations have recently received renewed attention, including the rol
Charts are structured visual compositions whose elements have distinct functional roles, semantic corresponden
Reinforcement Learning with Verifiable Rewards (RLVR) has been central to the recent success of Large Reasonin
Exploration in reinforcement learning (RL) remains a fundamental challenge. Recent goal-conditioned RL strateg
Predictive process monitoring aims at forecasting various aspects of running processes. Among the different ta
We address the challenge of scalable uncertainty quantification in large-scale scientific applications, where
This paper proposes a level-set-based physics-driven neural network solver (LSPDNN) for 3-D electromagnetic in
Chagas disease is a major cause of cardiomyopathy in Latin America. Cardiac magnetic resonance (CMR) imaging c
In data-driven training, multivariate time-series forecasting is usually optimized with a scalar loss averaged
Reliable video generation requires more than high-quality frames to form a coherent story: a model must mainta
We apply transfer learning (TL) to construct data-driven models of inclusive electron-nucleus cross sections.
Universal machine-learning interatomic potentials (u-MLIPs) aim to generalize across diverse configurations. B
Accurate diagnostic and risk-prediction models are important for supporting clinical decision-making during in
Industrial process monitoring is fundamental to the safety and economic performance of modern process plants.
Representing a 3D scene as multi-view images allows 2D VLMs to reason in 3D by reusing priors from pre-trainin
To mitigate the scalability bottleneck in the radio access network (RAN) in federated edge learning (FEEL), ov
We study zeroth-order optimization of non-convex functions with the aid of directional hints, which are cheap
Urban region embeddings have shown promising results in diverse urban sensing tasks such as crime, income, and
Detailed routing remains a dominant runtime bottleneck in physical design due to increasing complexity of desi
Agentic AI systems built on large language models fail in two persistent ways that scaling does not fix: they
Slice sampling is a Markov chain Monte Carlo algorithm that draws its next state uniformly from a "slice"---a
In nonconvex optimization problems arising in geometric machine learning, data augmentation is commonly used t
Large language models are increasingly deployed as agents that plan over long horizons and act through externa
Dexterous manipulation involves contact-rich and fine-grained interactions with the physical world, posing sig
Mid-training, the stage between pre-training and alignment, is where a model's per-domain data composition is
Ensuring the safety of autonomous driving is a critical challenge. Scenario-based testing is a systematic proc
Text-to-SQL systems translate natural language queries into executable SQL, democratizing access to structured
Automatic fact-checking systems assess the veracity of claims given evidence from relevant documents. Large La
Analysts in emerging equity markets keep answering the same questions. Did fundamentals match the market's res
Benchmark scores are a central currency in model releases: they inform purchasing decisions, shape public trus
Large language model (LLM)-powered agents can be accurate on average yet unreliable in production, a discrepan
Threat hunting increasingly depends on converting unstructured knowledge (e.g., Cyber Threat Intelligence repo
The transition from single-core to multi-core architectures in safety-critical embedded systems introduces sig
Long-form instructional videos require automatic chaptering to support browsing, navigation, and knowledge acc
Streaming automatic speech recognition (ASR) for real-time voice agents and full-duplex dialogue must provide
Onboard object detection in Earth observation is constrained by limited computational resources and the absenc
An action chunk can span several stages of a manipulation task, yet a label for its first step describes only
Large Language Model (LLM) agents are evolving from single-session tools toward long-term personal assistants
Large language model (LLM)-based multi-agent systems (MAS) achieve strong performance by employing specialized
KV cache is evolving from a serving optimization into an external memory substrate for long-term LLM agents. I
Personalized memory helps LLM agents deliver stable, tailored assistance by storing and reusing user-specific
Multi-agent large language models solve complex tasks by coordinating several policies in a shared environment
Modeling long-term user behavior is central to sequential recommendation and billion-scale industrial recommen
In persistent interactions, long contexts may encode an evolving process rather than a fixed record: later eve
Training capable cyber agents is often treated primarily as a problem of model scale, yet open-weight post-tra
Full-duplex spoken dialogue systems enable simultaneous listening and speaking, but their audio-only perceptio
Bimanual manipulation policies require large and diverse training datasets, yet collecting demonstrations on p
Travel survey data are essential for transportation planning and travel behavior analysis, yet collecting larg
Agent memory systems have demonstrated significant potential in long-term dialogue, personalized assistants, a
Hallucination detection is crucial for large language models (LLMs), as hallucinated content creates significa
This report analyzes Qiushi Engine v0.8 across all 40 test tasks in AstaBench E2E-Bench-Hard, a benchmark that
Vision-language models (VLMs) augmented with retrieval-augmented generation (RAG) benefit from access to exter
Sparse autoencoder (SAE)-based steering has been widely used to address knowledge conflicts by guiding LLMs to
Aerial Object Goal Navigation (ObjectNav) requires an unmanned aerial vehicle (UAV) to locate a described targ
We present the IGT system for PolyFiQA Task 2 of the FinMMEval Lab at CLEF 2026, a multilingual financial ques
We study scheming in LLM agents, in which agents covertly pursue misaligned goals. Our focus is to understand
The Indonesian Digital Library of Culture (Perpustakaan Digital Budaya Indonesia, PDBI; budaya-indonesia.org)
Firms making inventory decisions have access to operational data, optimization tools, and large language model
Speech deepfakes can mimic a speaker's voice convincingly enough to deceive listeners and automated systems. T
Evaluating first-stage retrievers in large-scale production RAG requires a benchmark that pairs a large-scale
Large Language Models are increasingly deployed as information intermediaries, yet measuring their political b
Remote sensing multimodal large language models (RS-MLLMs) have advanced scene understanding and visual questi
Oncology care operates at constant pressure of absorbing rapidly evolving evidence base in biomedicine. The Am
Recursive self-improvement (RSI) requires a concrete mechanism through which an AI system observes its capabil
Key-value (KV) cache eviction is essential for scaling long-context inference in Large Language Models. Howeve
NumPy is a widely used Python library for numerical scientific computing, known for its declarative APIs and i
World models are increasingly used as policy-in-the-loop imagination environments, where reliable rollouts req
Autoregressive (AR) video diffusion models have shown great potential in real-time video generation. Recent me
Autonomous 3D active mapping requires a space robot to choose where to sense while building the geometry neede
Vision transformers (ViTs) have achieved remarkable generalization across visual domains, yet little is known
State-of-the-art vision models process images in their entirety, lacking the ability to selectively zoom in on
Partially Relevant Video Retrieval (PRVR) retrieves untrimmed videos when queries describe only short moments.
Object 6D pose estimation formulations have progressively reduced reliance on object-specific priors, evolving
Reasoning segmentation converts an implicit linguistic conclusion into a precise mask, requiring both semantic
High-fidelity vehicle assets are essential for controllable traffic scene generation, particularly for synthes
Cultural heritage collections often contain contemporary and historical visual records of the same physical ob
While 3D Gaussian Splatting (3DGS) has emerged as a powerful representation for real-time novel view synthesis
A simultaneous localization and mapping (SLAM) method using a monocular camera and a low-cost inertial measure
Gaussian splats provide a fast, high-fidelity representation for 3D objects but are often constructed from inc
Multiple Instance Learning (MIL) is widely used for weakly supervised learning, particularly in digital pathol
Text-to-motion (T2M) generation maps natural language to human joint movements, aiding gaming, VR, and robotic
World models take multimodal inputs like text, photos, and diagrams to generate dynamic scenes in accordance w
Video Large Language Models (VideoLLMs) are increasingly deployed in safety-critical applications such as cont
Object detectors have shown remarkable performance in various fields, among these medical imaging, surveillanc
VLMs have shown promise for autonomous driving, yet still suffer from hallucination, weak spatio-temporal perc
Video generation models have recently attracted substantial attention for their ability to generate visually c
Generalist robot policies carry broad manipulation priors from large-scale data, but specializing them to a ne
Workspace analysis measures where a robot can place its end effector. For visually guided manipulation, reacha
Input devices for robotic microsurgery are frequently described as preserving the surgeon's trained technique,
Autonomous exploration on uneven terrain requires ground robots to balance exploration efficiency, coverage co
This paper presents a soft robotic drummer for accurate and efficient drum rolls. High-frequency drum rolls re
Surface coverage with task-redundant manipulators is challenging because each surface point may admit multiple
TacClip is a minimally encumbering wearable device for recording fingertip deformation caused by contact force
We propose a new approach for synthesizing large sets of winning strategies in stochastic parity games (2.5-pl
Offline multi-agent payoff models are estimated under a logging distribution but used on distributions induced
Many pedestrian trajectory prediction algorithms have been proposed to improve the safety of navigation for mo
The software supply chain has become an increasingly exposed attack surface because of its reliance on intrica
Purpose: Accurate CT protocol selection is critical for diagnostic quality and patient safety, yet the current
Key--value (KV) cache compression is an effective way to reduce the memory overhead of large language model (L
Counterfactual explanations formalize "what-if" scenarios by identifying modifications to an input instance th
Deep learning is an efficient technique to monitor the real time welding process, reducing post-welding repair
Multimodal large language models often capture visual-linguistic correlations but struggle to predict how loca
Chain-of-Thought (CoT) improves the reasoning ability of Large Language Models (LLMs) but incurs substantial c
Ask a language model to respond "very excitedly," and its output is typically only mildly more energetic. We q
Late-interaction models such as ColBERT achieve strong effectiveness by representing each document with many t
Chain-of-thought (CoT) reasoning improves the reasoning ability of large language models by introducing interm
Adaptive Conformal Inference (ACI) extends conformal prediction to non-exchangeable settings by adjusting the
Reasoning agents increasingly rely on external tools such as web search to answer complex queries. Reinforceme
We study the control of Markov decision processes in which the quality of a policy is evaluated by a dynamic,
LLM-based digital twins promise to reduce repeated human data collection by generating person- specific respon
Multimodal Large Language Models (MLLMs) show strong progress on vision-language tasks, yet their reliability
Natural Language Processing (NLP) in the climate domain requires models to process heterogeneous text sources,
A wrong number is worse than no answer. Across factuality-critical domains -- audience metrics, scheduling and
Democratizing access to the knowledge held in large corpora of tables such as data lakes is emerging as a cent
Financial scenarios are diverse and complex, spanning varying data conditions, tool configurations, and workfl
Large language models (LLMs) increasingly shape communication, learning, work, creativity, and decision-making
When you merge two fine-tuned models from the same base checkpoint by simply averaging their weights, you impl
Large language models (LLMs) have demonstrated strong performance in table understanding. However, they typica
The same underlying computational problem is solved across unrelated fields under different names: recursive B
Converting in-service reinforced-concrete (RC) building blueprints into simulation-ready models---structured f
Taxonomy induction aims to organize concept sets into coherent hierarchical structures. Recent LLM-based metho
Demographic synthetic survey panels are often validated by matching aggregate answers to published surveys. We
Ethical evaluation of Large Language Models (LLMs) often characterizes model values as static and monolithic.
Post-hoc sparse attention accelerates long-context prefill by routing each query to a small set of token-level
Diffusion Large Language Models (dLLMs) generate text via bidirectional iterative denoising, naturally support
Minimal-pair benchmarks such as BLiMP evaluate linguistic knowledge by testing whether language models (LMs) p
Autoregressive language models generate one token per decoding step, limiting the useful output of each forwar
Retrieval-augmented generation (RAG) enables large language models (LLMs) to answer questions by accessing ext
Robotic ophthalmic surgery offers high precision but introduces a "sensory gap" by decoupling the surgeon from
Weak gravitational lensing shear and convergence trace the distribution of baryonic and dark matter across spa
Accurate 3D plant organ segmentation is fundamental to automated phenotyping. Existing approaches rely on anno
In real-world applications, pedestrian trajectory prediction models rely on inputs from detection and tracking
Dot maps, which visualize individual data points as dots over a geographic region, are widely used across dive
We present our solution to the LUMPI track of the UCF UrbanTwin Sim2Real LiDAR Challenge at the 6th DriveX Wor
Agricultural parcel polygons play a fundamental role in geospatial applications such as precision agriculture,
Transferring human hand demonstrations to robotic grippers has recently emerged as a cost-effective solution f
Automated fingermark identification is the foundation of forensic investigation, yet progress in the field is
Privacy-sensitive surveillance systems could benefit from large vision-language models (VLMs), but such models
Climate change is increasing the severity and unpredictability of natural disasters. In time-critical crises s
Morphology of the sub-basal nerve plexus (SNP) reflects peripheral nerve health, and corneal confocal microsco
SLAM systems based on 3D Gaussian Splatting (3DGS) have recently demonstrated promising reconstruction accurac
Despite the rapid progress of Multimodal Large Language Models (MLLMs) in 2D vision-language tasks, robust mul
Out-of-distribution (OOD) detection is critical for safe deployment of medical AI systems. Recently, test-time
Recent image-to-3D generation models built on flow-matching diffusion Transformers (DiT) can produce high-fide
This paper presents a novel framework designed to enhance key object identification in autonomous driving. Exi
Large Vision-Language Models (LVLMs) have achieved strong performance on diverse visual tasks, yet their abili
Localizing a font into new languages is a highly intricate task requiring precise design adaptation of glyphs,
Purpose: This study aims to develop an AI framework applicable for postoperative imaging for automated measure
Synthesizing photorealistic driving videos along specified trajectories is essential for scalable closed-loop
Connected robotics is an emerging 6G application where mobile robots follow natural-language instructions to m
Human videos are an abundant source of dexterous manipulation behaviors, but they lack tactile information tha
Vision-Language-Action (VLA) policies are commonly adapted to new manipulation settings through additional gra
Contact-rich assembly remains challenging because it requires submillimeter spatial accuracy and reliable inte
Target-based LiDAR-camera extrinsic calibration is a prerequisite for multi-sensor fusion in robotics. However
How robot body configuration shapes human intervention during approach remains underexplored. We conducted a w
World-Action Models inherit world knowledge from video-generative priors, and channel it into executable contr
Local pedestrian-vehicle forecasting spans heterogeneous physical scales: pedestrians combine root locomotion
Estimating the 6D pose of textureless objects without prior CAD models remains a critical challenge due to the
Humanoid locomotion requires control policies that remain stable under imperfect sensing while exploiting temp
Autonomous mobile robots performing person-following tasks often suffer from temporary occlusions and sensor t
Robotic manipulation often requires acting on information that is no longer visible, yet Vision-Language-Actio
Reinforcement learning has become the leading paradigm in legged locomotion, enabling complex behaviors from b
Distributed Dexterous Manipulation (DDM) is a novel paradigm that presents significant control challenges due
Distributed learning control for multirobot systems (MRS) offers significant flexibility in presence of uncert
Trust-Hub reuses host circuits: several files differ mainly in the inserted Trojan. When gates from sibling va
We consider dimensionality reduction for high-dimensional observations accompanied by a supplied partition int
We study the identity straight-through estimator (STE) for training a two-layer binary-activation network with
In many scientific disciplines, weak signals of interest are obscured by dominant nuisance signals that are se
When non-expert users ask LLMs for assistance, their queries can often have misconceptions (e.g., "How do I pa
Authorship signals matter in settings where writing style carries identity: digital forensics, plagiarism anal
High-quality structured organic reaction data are essential for developing artificial intelligence for chemist
Sequential memory agents process long documents by reading chunks one after another while maintaining a compac
Tokenization forms the foundation of modern Natural Language Processing (NLP) systems by transforming raw text
Although multimodal Large Language Models (MLLMs) excel in diverse tasks, their scalability remains limited by
Although professional workflows leverage large language models widely, the interpretation for auditing unconst
Recent studies report that automated red-teaming finds more vulnerabilities, at lower cost, than human red-tea
Enterprise AI is evolving into an Enterprise Operating System where autonomous AI agents can plan, reason, use
Large language models (LLMs) have shown strong potential for translating natural-language (NL) requirements in
Recursive Super-Resolution (SR) extends fixed-scale SR to extreme magnification by repeatedly feeding predicti
Reinforcement learning (RL) across multiple domains can broaden the reasoning capabilities of large language m
Randomized benchmarking can hide classical temporal correlations because its Clifford-twirled response is even
Existing approaches to visual attribute value extraction (AVE) primarily rely on static product images, failin
Graph-agentic retrieval-augmented generation combines structured evidence with adaptive controllers that can p
Although highly effective in vision and language domains, applying in-context learning to robotics remains cha
Active mapping requires a robot to select camera viewpoints that efficiently reconstruct an unknown 3D scene.
Urban traffic monitoring plays a critical role in safety analysis, congestion management, and incident respons
Neural inertial odometry has demonstrated strong potential for motion estimation in challenging environments,
The generation of safety-critical traffic scenarios is essential for training and evaluating autonomous vehicl
Existing unlearning approaches typically rely on post hoc weight adaptation or distillation, leading to duplic
Principal component analysis (PCA) can rotate away from its population target when a covariance matrix is esti
Vision-language-action (VLA) models have shown promising generalization for language-conditioned robot manipul
Grounding natural-language instructions into reliable and executable actions remains a fundamental challenge f
Multi-fingered dexterous manipulation remains a frontier for real-world reinforcement learning (RL) due to the
Robot manipulation foundation models require scalable evaluation and data generation across diverse scenarios,
We give a necessary and sufficient condition for the existence of power-one sequential tests in an i.i.d. comp
Dynamical symbolic regression methods identify governing differential equations from noisy data, balancing int
Conformal counterfactual inference enables network operators to use logged telemetry to reliably answer 'what-
We study regression under bounded responses in terms of excess mean squared error. When the comparator class i
Leveraging their inherent sparse event-driven computation, spiking neural networks (SNNs) offer a promising pa
Vision-language models are increasingly used as reward functions for robotic learning, but this role requires
Reliable 3D understanding of the surrounding environment is a core requirement for autonomous driving. Multi-v
World-action models (WAM) predict future states to guide robot actions, enabling learning from both action-fre
Human-robot interaction (HRI) enables intuitive and intelligent collaboration between humans and robots in rea
Achieving robust SLAM in large-scale underground coal mines with complex structures and severe degeneracies re
World-action models guide action generation with predicted future observations, but vision-centric predictions
Diffusion probabilistic models can capture the multi-modal, interaction-rich distribution of joint future traj
Multi-vehicle cooperative autonomous driving enhances the safety and reliability of autonomous driving systems
Autonomous robots continuously encounter objects, changes, and situations, and every event admitted into cogni
Three-dimensional scene graphs (3DSGs) have emerged as a promising approach for building geometrically grounde
Large smart-farming deployments generate continuous scientific data from spatially distributed sensors, includ
Rollcast is a probabilistic forecasting method for univariate time series that combines a compact set of rolli
Logit-based knowledge distillation for autoregressive language models usually aligns teacher and student next-
This paper presents a new method for headland coverage path planning for arable fields. Several earlier approa
For robots to operate reliably in real-world environments, they need to perceive their surroundings, act, and
An approval committee is Hare-core stable if no coalition meeting the Hare quota can strictly improve by movin
We introduce the first Probably Approximately Correct (PAC) learning framework for general-sum concurrent stoc
We introduce an uncoupled learning algorithm which, when employed by all players of an arbitrary $N$-player no
The Gromov-Wasserstein (GW) distance provides a principled framework for aligning metric measure (mm) spaces b
Many NLP tasks, such as summarization and extractive question answering, reduce to retrieving relevant content
Causal inference is the practice of estimating the effect of a treatment or intervention from data. It traditi
Reinforcement learning typically optimizes average reward. For generative policies, the average can hide an im
Neural Architecture Search (NAS) has emerged as a powerful paradigm for automatically designing deep neural ne
This chapter reconstructs the Hopfield network as a physical theory of memory rather than merely an early neur
This chapter reconstructs the McCulloch-Pitts program as a physics of neural computation rather than the famil
This paper is the first in a series on Turn-Based Combat Arena, a configurable framework for turn-based strate
Many statistical models involve parameter-dependent normalizing constants that are computationally intractable
We extend recent work establishing an equivalence between one-layer transformers and nearest-neighbor classifi
Reconstructing a damaged musical fragment is an inverse problem: the observed sequence contains partial inform
Black-Box Optimization (BBO) has broad applications, while traditional algorithms such as evolutionary algorit
Price extraction from websites is a key task for market monitoring, price comparison, and business analytics i
Neural network mixed-effects models (NMMs) have gained traction by combining the strong representation and pre
Agent evaluations often use one benchmark to choose a workflow and then search for task types where its advant
Uncoupled no-regret dynamics provide a decentralized route to equilibrium, but prior guarantees for individual
In resin transfer moulding, complete saturation of the fibre preform is necessary before the resin front reach
Large language models are increasingly required to generate responses that satisfy multiple competing objectiv
Structured pruning uses surrogate objectives because direct task evaluation over every feasible mask is too ex
While modern generative models excel at modeling complex data, precise inference-time control in conditional g
Causal representation learning (CRL) aims to recover latent causal variables and their structural relations fr
In metric social choice, voters rank candidates by their distances in an unknown metric space. A voting rule u
Decision-time equilibrium search carried poker to superhuman play, but it has so far relied on tractable subga
Representation learning begins when training changes the features that define similarity between data. A froze
Synthetic data can improve statistical inference when real data are scarce, but naively treating synthetic sam
In the framework of network dynamics, learning models, and neural tangent kernels (NTK), we show that the corr
Reliable prediction of time-varying channel state information (CSI) is essential for efficient wireless commun
Global sharpness of a sampling bound does not determine whether the bound is attainable on a particular fixed
In the study of Nash equilibria of finite-player games, one often seeks equilibria that are compatible with pr
Kolmogorov-Arnold Networks (KANs) replace fixed activations in deep architectures with learnable univariate ed
We address the simultaneous prediction of multiple high-dimensional physical fields governed by linear equalit
The simple exclusion process (SEP) is a paradigmatic model for nonequilibrium transport, yet the rich dynamics
Stochastic gradient descent (SGD) is typically analyzed at a deterministic horizon chosen before the algorithm
We analyze exact-metric, Metropolis-adjusted Dikin walks by keeping the proposal determinant and reverse quadr
The functional annotation of genes in non-model organisms remains a significant challenge in computational bio
In this paper, we introduce ES-AHD, a novel framework that fundamentally integrates Evolution Strategy (ES) in
Low-earth-orbit (LEO) satellites enable high-resolution, large-scale Earth observation for applications such a
Developments in high-performance computing (HPC) technology continue to drastically increase quantities of ava
Reversible computing is a novel paradigm that has recently emerged and extends traditional forwards-only compu
In a deliberative poll, once submissions outnumber what anyone will read, some mechanism chooses which argumen
Analog circuit topology synthesis remains challenging because useful designs occupy a tiny fraction of a combi
Symbolic protocol verification models the network attacker as a Dolev--Yao (DY) intruder, which does everythin
We give a randomized voting rule with expected metric distortion $5/2$, improving the previous best upper boun
Classical seismic data reconstruction relies on manually designed structural priors and iterative operators, w
We reconstruct the mentor--student network through which documented scholarly training passed across roughly n
We present an optical random access memory (ORAM) based on warm cesium (Cs) atomic vapor and demonstrate its o
We propose a randomized social choice rule called Mixed Integrated Veto (MIV) with metric distortion of $5/2$,
Automated machine learning (AutoML) systems search for pipelines within a space of preprocessing operators, le
Liquid democracy permits voters to vote directly or delegate their votes to others. Existing algorithmic analy
We find no evidence of critical scaling in the Schelling segregation model, in either the Moore neighborhood o
We study the group-fair distortion of metric facility assignment problems, where a set of agents, partitioned
Deciding whether a Sokoban puzzle is solvable is PSPACE-complete (Culberson, 1997): solutions can be exponenti
Generative search engines (GSEs) answer user queries directly from crawled web content. The capture of value f
An invariant behavioral profile is the defining vulnerability of traditional honeypot installations: a skilled
Many real-world interactions among self-interested parties can be modeled by game theory, and the rapid advanc
Evolutionary feature construction has shown strong promise in symbolic regression by automatically discovering
At a finite public-chance cut, counterfactual regret minimization (CFR) must choose how many outcomes to evalu
解決策候補の多様化を向上させる進化戦略を開発するために、LLMを用いて解決策候補を生成するアプローチを提案している。
Spiking Neural Networks (SNNs) encode information through binary spikes and compute in an event-driven manner,
この研究では、半推測状態の状況でアダプティブ行動を示すために、反射的な組織がどのように機能するかを調査する。これでは、現時点の観察だけを基に、内部状態が情報を保持できる計算プロパティを開発します。
成功した突然変異戦略の中には、単一の実行で利用可能な知識が存在し、その知識は複数のタスク間で移行することができる。しかし、既存のLLM-ベースの進化的フレームワークでは、再利用可能な知識は捨てられ、同じアイディアの再発見
この研究では、モンゴメリーカット形式のサブモデル問題の最適化に関して研究が行われている。特に、与えられた制約で問題を解くことができる方法論が提案されている。研究では、多タスク問題を同時に解くことで、問題を解く時間が短縮さ
Full justified representation (FJR) is among the strongest known satisfiable proportionality axioms for approv
As the adoption of satellite-enabled Internet of Things (IoT) continues to rise, its intricate multidomain arc
Competitive artificial-life systems can rank trained controllers differently under training and ecological eva
Learning dynamics in zero-sum games are typically analyzed under algorithmic symmetry: both agents use the sam
理論オブミンドの評価基準「Avalon-ToM-Bench」を提案。社会的認識を評価するための基準を提供する。
ポーカーの対局戦略を最適化する研究です。この研究では、独立チップモデル(ICM)を超える戦略的継続性最適化(SCO)アルゴリズムを提案しました。
This note aims to serve as an entry point to the literature on learning in games, a topic with significant the
We prove that every fair-division instance with four agents, additive valuations over the non-negative reals,
Training controllers that are safe and robust in simulation, and systematically assessing their readiness for
この論文では、自律的な生成モデルの内部表現を分析し、未知のデータ分類に基づいて順序性が生じていることを示します。
回帰ニューラルネットワークの接続を簡素化する方法がいくつか提案されてきたが、生物学的背景に基づいた方法は少ない。研究者たちは、これまでノイズから切断するノイズ-プルーンという方法を提案しており、これは再起動後の機能を最も
Particle swarm optimization (PSO) has been widely applied to solve complex optimization problems from real-wor
この研究では、記憶を維持し更新する能力を強化するため、繰り返し神経ネットワークを用いて記憶の特性を分析しました。記憶を維持するための神経計算の一種として、ダイショビュールノーマリゼーション（Divisive Normal
Spiking Neural Networks (SNNs) enable event-driven computation with sparse activations, but building multimoda
The rapid development of Large Language Models (LLMs) has opened new avenues for Automated Heuristic Design (A
この論文では、スワームと進化アルゴリズムを分析するための新しいアプローチを提案しています。このアプローチでは、候補生成を分離し、目標によらず変化する部分と、目標に関係する部分を同定します。
An LLM agent's capability depends not only on model weights but on its harness: prompts, tools, skills, and co
Automated heuristic design (AHD) with large language models (LLMs) has produced strong heuristics for combinat
ニュロモーティックコンピューティングの研究を目的としたフレームワーク
Symbolic regression provides analytical expressions, but it is usually applied one output at a time. This is l
Particle swarm optimization (PSO) is a widely used metaheuristic, prized for its simplicity and small paramete
Gene regulatory network modeling often requires balancing predictive accuracy and mechanistic interpretability