Tootfinder

Optimised neural networks for online processing of ATLAS calorimeter data on FPGAs
Georges Aad, Raphael Bertrand, Lauri Laatu, Emmanuel Monnier, Arno Straessner, Nairit Sur, Johann C. Voigt
https://arxiv.org/abs/2510.11469

Optimised neural networks for online processing of ATLAS calorimeter data on FPGAs
A study of neural network architectures for the reconstruction of the energy deposited in the cells of the ATLAS liquid-argon calorimeters under high pile-up conditions expected at the HL-LHC is presented. These networks are designed to run on the FPGA-based readout hardware of the calorimeters under strict size and latency constraints. Several architectures, including Dense, Recurrent (RNN), and Convolutional (CNN) neural networks, are optimised using a Bayesian procedure that balances energy …

Google debuts Titans, an architecture combining RNN speed with transformer performance for real-time learning, able to scale effectively to a 2M context window (Google Research)
https://research.google/blog/titans-miras-helping-ai-have-long-term-memory/

Titans + MIRAS: Helping AI have long-term memory

GRNND: A GPU-Parallel Relative NN-Descent Algorithm for Efficient Approximate Nearest Neighbor Graph Construction
Xiang Li (Nanjing University), Qiong Chang (Institute of Science Tokyo), Yun Li (Nanjing University), Jun Miyazaki (Institute of Science Tokyo)
https://arxiv.org/abs/2510.02774

GRNND: A GPU-Parallel Relative NN-Descent Algorithm for Efficient Approximate Nearest Neighbor Graph Construction
Relative Nearest Neighbor Descent (RNN-Descent) is a state-of-the-art algorithm for constructing sparse approximate nearest neighbor (ANN) graphs by combining the iterative refinement of NN-Descent with the edge-pruning rules of the Relative Neighborhood Graph (RNG). It has demonstrated strong effectiveness in large-scale search tasks such as information retrieval and related tasks. However, as the amount and dimensionality of data increase, the complexity of graph construction in RNN-Descent r…

Dissecting Larval Zebrafish Hunting using Deep Reinforcement Learning Trained RNN Agents
Raaghav Malik, Satpreet H. Singh, Sonja Johnson-Yu, Nathan Wu, Roy Harpaz, Florian Engert, Kanaka Rajan
https://arxiv.org/abs/2510.03699

Dissecting Larval Zebrafish Hunting using Deep Reinforcement Learning Trained RNN Agents
Larval zebrafish hunting provides a tractable setting to study how ecological and energetic constraints shape adaptive behavior in both biological brains and artificial agents. Here we develop a minimal agent-based model, training recurrent policies with deep reinforcement learning in a bout-based zebrafish simulator. Despite its simplicity, the model reproduces hallmark hunting behaviors -- including eye vergence-linked pursuit, speed modulation, and stereotyped approach trajectories -- that c…

StateX: Enhancing RNN Recall via Post-training State Expansion
Xingyu Shen, Yingfa Chen, Zhen Leng Thai, Xu Han, Zhiyuan Liu, Maosong Sun
https://arxiv.org/abs/2509.22630 https:…

StateX: Enhancing RNN Recall via Post-training State Expansion
While Transformer-based models have demonstrated remarkable language modeling performance, their high complexities result in high costs when processing long contexts. In contrast, recurrent neural networks (RNNs) such as linear attention and state space models have gained popularity due to their constant per-token complexities. However, these recurrent models struggle with tasks that require accurate recall of contextual information from long contexts, because all contextual information is comp…

Native Hybrid Attention for Efficient Sequence Modeling
Jusen Du, Jiaxi Hu, Tao Zhang, Weigao Sun, Yu Cheng
https://arxiv.org/abs/2510.07019 https://arxiv.…

Native Hybrid Attention for Efficient Sequence Modeling
Transformers excel at sequence modeling but face quadratic complexity, while linear attention offers improved efficiency but often compromises recall accuracy over long contexts. In this work, we introduce Native Hybrid Attention (NHA), a novel hybrid architecture of linear and full attention that integrates both intra \& inter-layer hybridization into a unified layer design. NHA maintains long-term context in key-value slots updated by a linear RNN, and augments them with short-term tokens fro…

Lipschitz-Based Robustness Certification for Recurrent Neural Networks via Convex Relaxation
Paul Hamelbeck, Johannes Schiffer
https://arxiv.org/abs/2509.17898 https://

Lipschitz-Based Robustness Certification for Recurrent Neural Networks via Convex Relaxation
Robustness certification against bounded input noise or adversarial perturbations is increasingly important for deployment recurrent neural networks (RNNs) in safety-critical control applications. To address this challenge, we present RNN-SDP, a relaxation based method that models the RNN's layer interactions as a convex problem and computes a certified upper bound on the Lipschitz constant via semidefinite programming (SDP). We also explore an extension that incorporates known input constraint…

An Adversarial Robust Behavior Sequence Anomaly Detection Approach Based on Critical Behavior Unit Learning
Dongyang Zhan, Kai Tan, Lin Ye, Xiangzhan Yu, Hongli Zhang, Zheng He
https://arxiv.org/abs/2509.15756

An Adversarial Robust Behavior Sequence Anomaly Detection Approach Based on Critical Behavior Unit Learning
Sequential deep learning models (e.g., RNN and LSTM) can learn the sequence features of software behaviors, such as API or syscall sequences. However, recent studies have shown that these deep learning-based approaches are vulnerable to adversarial samples. Attackers can use adversarial samples to change the sequential characteristics of behavior sequences and mislead malware classifiers. In this paper, an adversarial robustness anomaly detection method based on the analysis of behavior units i…

Overcoming the curse of dimensionality: Enabling multi-layer photon transport with recurrent neural network
Daniel Carne, Ziqi Guo, Xiulin Ruan
https://arxiv.org/abs/2509.22890 …

Overcoming the curse of dimensionality: Enabling multi-layer photon transport with recurrent neural network
Monte Carlo simulations are commonly used to calculate photon reflectance, absorptance, and transmittance of multi-layer scattering and absorbing media, but they can quickly become prohibitively expensive as the number of layers increases. In this study, we show that although a plain neural network suffers from the curse of dimensionality and fails to yield acceptable predictions of multilayer media, we introduce a recurrent neural network (RNN) trained on the same Monte Carlo simulation datase…

🇺🇦 #NowPlaying on BBCRadio3's #Breakfast
Trad., Rune Tonsgaard SŸrensen, Frederik Øland, AsbjŸrn NŸrgaard, Fredrik Sjölin & Danish String Quartet:
🎵 As I Walked Out
#Trad #RuneTonsgaardSørensen #FrederikØland #AsbjørnNørgaard #FredrikSjölin #DanishStringQuartet