gpt-생성-텍스트-탐지-벤치마크-데이터셋-및-텐서-기반-탐지-방법

문제-중심-다중-전문가-대조-학습-프레임워크를-통한-심층-순차-지식-추적-모델의-정확성-및-해석-가능성-향상

최적-해-보장-없이-공정성-매트로이드-전역-제약-조건-하에서-최대-커버리지-달성

신경망-압축에서-격자-변환-부호화를-통한-율-왜곡-한계-접근

maximierung-der-abdeckung-unter-berücksichtigung-von-fairness-matroid-und-globalen-beschränkungen

ein-frage-zentriertes-multi-experten-kontrastives-lernframework-zur-verbesserung-der-genauigkeit-und-interpretierbarkeit-von-tiefen-sequenziellen-wissens-verfolgungsmodellen

erkennung-von-durch-gpt-generiertem-text-benchmark-datensatz-und-tensorbasierte-erkennungsmethode

modular-end-to-end-network-with-interpretable-sensorimotor-learning-for-autonomous-navigation


再帰的に生成されたデータでAIモデルを訓練すると、モデルが不可逆的な欠陥に陥る可能性がある。


coremsg

&nbsp;Analysis shows that indiscriminately training generative artificial intelligence on real and generated content, usually done by scraping data from&nbsp;the Internet, can lead to a collapse in the ability of the models to generate diverse high-quality output.

AI models collapse when trained on recursively generated data - Nature

再帰的に生成されたデータで訓練されたaiモデルの崩壊


再帰的に生成されたデータで訓練されたAIモデルの崩壊


title_rewrite


オープンAIが開発した新しい推論モデル「プロジェクト ストロベリー」は、現在の大規模言語モデルを大きく上回る性能を持ち、AIの真のモートを実現する可能性がある。


Project Strawberry, OpenAI’s Leaked Breakthrough

オープンaiの最新の画期的な成果-プロジェクト-ストロベリー-の詳細


オープンAIの最新の画期的な成果「プロジェクト ストロベリー」の詳細



グーグルDeepMindの研究者は、言語モデルのパフォーマンスと学習効率を大幅に向上させるためのアーキテクチャの改良を見出した。


A deep dive into the the Mixture-of-a-Million-Experts (MoME) architecture, which outperforms traditional LLMs like never before

Here Is Google DeepMind’s New Research To Build Massive LLMs With A Mixture Of Million Experts

グーグルdeepmindの新しい研究-数百万の専門家を組み合わせた巨大llmの構築


グーグルDeepMindの新しい研究:数百万の専門家を組み合わせた巨大LLMの構築



自然画像の中で、人間の脳と人工ニューラルネットワークは物体の実世界サイズを独立して表現する。


Remarkably, human brains have the ability to accurately perceive and process the real-world size of objects, despite vast differences in distance and perspective. While previous studies have delved into this phenomenon, distinguishing this ability from other visual perceptions, like depth, has been challenging. Using the THINGS EEG2 dataset with high time-resolution human brain recordings and more ecologically valid naturalistic stimuli, our study uses an innovative approach to disentangle neural representations of object real-world size from retinal size and perceived real-world depth in a way that was not previously possible. Leveraging this state-of-the-art dataset, our EEG representational similarity results reveal a pure representation of object real-world size in human brains. We report a representational timeline of visual object processing: object real-world depth appeared first, then retinal size, and finally, real-world size. Additionally, we input both these naturalistic images and object-only images without natural background into artificial neural networks. Consistent with the human EEG findings, we also successfully disentangled representation of object real-world size from retinal size and real-world depth in all three types of artificial neural networks (visual-only ResNet, visual-language CLIP, and language-only Word2Vec). Moreover, our multi-modal representational comparison framework across human EEG and artificial neural networks reveals real-world size as a stable and higher-level dimension in object space incorporating both visual and semantic information. Our research provides a detailed and clear characterization of the object processing process, which offers further advances and insights into our understanding of object space and the construction of more brain-like visual models.

### Competing Interest Statement

The authors have declared no competing interest.

Human EEG and artificial neural networks reveal disentangled representations of object real-world size in natural images

自然画像における人間の脳波と人工ニューラルネットワークが物体の実世界サイズの表現を明らかにする


AlphaFold2を使用して、タンパク質間相互作用を正確に識別するための機械学習モデルPPIscreenMLを開発した。


Protein-protein interactions underlie nearly all cellular processes. With the advent of protein structure prediction methods such as AlphaFold2 (AF2), models of specific protein pairs can be built extremely accurately in most cases. However, determining the relevance of a given protein pair remains an open question. It is presently unclear how to use best structure-based tools to infer whether a pair of candidate proteins indeed interact with one another: ideally, one might even use such information to screen amongst candidate pairings to build up protein interaction networks. Whereas methods for evaluating quality of modeled protein complexes have been co-opted for determining which pairings interact (e.g., pDockQ and iPTM), there have been no rigorously benchmarked methods for this task. Here we introduce PPIscreenML, a classification model trained to distinguish AF2 models of interacting protein pairs from AF2 models of compelling decoy pairings. We find that PPIscreenML out-performs methods such as pDockQ and iPTM for this task, and further that PPIscreenML exhibits impressive performance when identifying which ligand/receptor pairings engage one another across the structurally conserved tumor necrosis factor superfamily (TNFSF). Analysis of benchmark results using complexes not seen in PPIscreenML development strongly suggest that the model generalizes beyond training data, making it broadly applicable for identifying new protein complexes based on structural models built with AF2.

### Competing Interest Statement

The authors have declared no competing interest.

PPIscreenML: Structure-based screening for protein-protein interactions using AlphaFold

alphafoldを使用したタンパク質間相互作用のスクリーニング


AlphaFoldを使用したタンパク質間相互作用のスクリーニング



AIの内部メカニズムは人間には理解できない


No One Knows How AI Works

aiの仕組みがわからない


記憶の統合は、短期記憶システムの信頼性の高い記憶を優先的に長期記憶システムに統合することで、長期記憶の信号対雑音比を大幅に向上させることができる。


In a variety of species and behavioral contexts, learning and memory formation recruits two neural systems, with initial plasticity in one system being consolidated into the other over time. Moreover, consolidation is known to be selective; that is, some experiences are more likely to be consolidated into long-term memory than others. Here, we propose and analyze a model that captures common computational principles underlying such phenomena. The key component of this model is a mechanism by which a long-term learning and memory system prioritizes the storage of synaptic changes that are consistent with prior updates to the short-term system. This mechanism, which we refer to as recall-gated consolidation, has the effect of shielding long-term memory from spurious synaptic changes, enabling it to focus on reliable signals in the environment. We describe neural circuit implementations of this model for different types of learning problems, including supervised learning, reinforcement learning, and autoassociative memory storage. These implementations involve synaptic plasticity rules modulated by factors such as prediction accuracy, decision confidence, or familiarity. We then develop an analytical theory of the learning and memory performance of the model, in comparison to alternatives relying only on synapse-local consolidation mechanisms. We find that recall-gated consolidation provides significant advantages, substantially amplifying the signal-to-noise ratio with which memories can be stored in noisy environments. We show that recall-gated consolidation gives rise to a number of phenomena that are present in behavioral learning paradigms, including spaced learning effects, task-dependent rates of consolidation, and differing neural representations in short- and long-term pathways.

### Competing Interest Statement

The authors have declared no competing interest.

Selective consolidation of learning and memory via recall-gated plasticity

信頼性の高い記憶の選択的な統合による記憶システムの理論


視覚的手がかりの不確実性の増大に伴い、知覚誤差が暗黙的運動適応の過剰補償と飽和を引き起こす。


The sensorimotor system can recalibrate itself without our conscious awareness, a type of procedural learning whose computational mechanism remains undefined. Recent findings on implicit motor adaptation, such as over-learning from small perturbations and fast saturation for increasing perturbation size, challenge existing theories based on sensory errors. We argue that perceptual error, arising from the optimal combination of movement-related cues, is the primary driver of implicit adaptation. Central to our theory is the increasing sensory uncertainty of visual cues with increasing perturbations, which was validated through perceptual psychophysics (Experiment 1). Our theory predicts the learning dynamics of implicit adaptation across a spectrum of perturbation sizes on a trial-by-trial basis (Experiment 2). It explains proprioception changes and their relation to visual perturbation (Experiment 3). By modulating visual uncertainty in perturbation, we induced unique adaptation responses in line with our model predictions (Experiment 4). Overall, our perceptual error framework outperforms existing models based on sensory errors, suggesting that perceptual error in locating one’s effector, supported by Bayesian cue integration, underpins the sensorimotor system’s implicit adaptation.

### Competing Interest Statement

The authors have declared no competing interest.

Perceptual error based on Bayesian cue combination drives implicit motor adaptation

視覚的誤差に基づくベイズ的手がかり統合が暗黙的運動適応を駆動する


少数言語を含む200言語に対応する単一の大規模多言語モデルを開発し、従来の最先端モデルに比べて平均44%の翻訳品質向上を実現した。


Scaling neural machine translation to 200 languages is achieved by No Language Left Behind, a single massively multilingual model that leverages transfer learning across languages.

Scaling neural machine translation to 200 languages - Nature

200言語への神経機械翻訳の拡張


過去5年以上にわたり、トランスフォーマーアーキテクチャが主流となっているが、新しい基盤モデルYOCOがその地位を脅かす可能性がある。


YOCO, A New Foundation Model to Eliminate Transformers?

トランスフォーマーを排除する新しい基盤モデルyocoについて


トランスフォーマーを排除する新しい基盤モデルYOCOについて


再帰的に生成されたデータで訓練されたAIモデルの崩壊

オープンAIの最新の画期的な成果「プロジェクトストロベリー」の詳細

グーグルDeepMindの新しい研究:数百万の専門家を組み合わせた巨大LLMの構築

自然画像における人間の脳波と人工ニューラルネットワークが物体の実世界サイズの表現を明らかにする

AlphaFoldを使用したタンパク質間相互作用のスクリーニング

AIの仕組みがわからない

信頼性の高い記憶の選択的な統合による記憶システムの理論

視覚的誤差に基づくベイズ的手がかり統合が暗黙的運動適応を駆動する

200言語への神経機械翻訳の拡張

トランスフォーマーを排除する新しい基盤モデルYOCOについて

再帰的に生成されたデータで訓練されたAIモデルの崩壊

オープンAIの最新の画期的な成果「プロジェクト ストロベリー」の詳細

グーグルDeepMindの新しい研究:数百万の専門家を組み合わせた巨大LLMの構築

自然画像における人間の脳波と人工ニューラルネットワークが物体の実世界サイズの表現を明らかにする

AlphaFoldを使用したタンパク質間相互作用のスクリーニング

AIの仕組みがわからない

信頼性の高い記憶の選択的な統合による記憶システムの理論

視覚的誤差に基づくベイズ的手がかり統合が暗黙的運動適応を駆動する

200言語への神経機械翻訳の拡張

トランスフォーマーを排除する新しい基盤モデルYOCOについて

オープンAIの最新の画期的な成果「プロジェクトストロベリー」の詳細