{"total":13,"items":[{"citing_arxiv_id":"2605.22097","ref_index":56,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"Q-PhotoNAS: Hybrid Quantum Neural Architecture Search Framework on Photonic Devices","primary_cat":"quant-ph","submitted_at":"2026-05-21T07:35:59+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Q-PhotoNAS applies genetic algorithm search to jointly optimize classical preprocessing, phase encoding, and photonic circuit structure for hybrid quantum-classical models, reporting 99.44% and 98.78% accuracy on Digits and MNIST with projected photonic QPU inference times.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.16184","ref_index":18,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"Runtime-Orchestrated Second-Order Optimization for Scalable LLM Training","primary_cat":"cs.DC","submitted_at":"2026-05-15T17:03:55+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Asteria is a runtime system that enables second-order optimization for LLMs by dynamically distributing optimizer state across GPU, CPU, and NVMe while using asynchronous inverse-root computations and bounded-staleness synchronization.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12628","ref_index":53,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"Multistep Belief Space Dynamics Learning For Risk-Aware Control","primary_cat":"cs.RO","submitted_at":"2026-05-12T18:11:12+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A structured learning approach for multistep distributional dynamics in belief space enables real-time risk-aware MPC, validated via ablation on real off-road data and deployment on a full-sized vehicle.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12387","ref_index":53,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"A Semi-Supervised Framework for Speech Confidence Detection using Whisper","primary_cat":"cs.SD","submitted_at":"2026-05-12T16:50:54+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A hybrid semi-supervised framework fusing Whisper embeddings with acoustic and prosodic features achieves 0.751 Macro-F1 for speaker confidence detection and outperforms baselines including WavLM, HuBERT, and Wav2Vec 2.0.","context_count":1,"top_context_role":"baseline","top_context_polarity":"baseline","context_text":"high-quality, diverse data is essential for capturing the subtle acoustic variations of uncertainty that are underrepresented in the small ground truth set. 3) Impact of Model Architecture:To isolate the contribu- tion of the underlying representation, the proposed Whisper- Base encoder was benchmarked against three state-of-the-art Self-Supervised Learning models (Wav2Vec 2.0 [15], Hu- BERT [53], WavLM [54]) and a smaller architectural variant (Whisper-Tiny). Table VII summarises the results. Wav2Vec 2.0 yields the lowest performance (0.661Macro-F1), but HuBERT and WavLM show notable gains, with WavLM achieving a strong score of0.737. Notably, WavLM matches the proposed model in the Medium confidence class (0.672). However, the proposed Whisper-Base architecture achieves"},{"citing_arxiv_id":"2605.11111","ref_index":32,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"ShardTensor: Domain Parallelism for Scientific Machine Learning","primary_cat":"cs.DC","submitted_at":"2026-05-11T18:20:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ShardTensor is a domain-parallelism system for SciML that enables flexible scaling of extreme-resolution spatial datasets by removing the constraint of batch size one per device.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"for high resolution data, deeper models often require more memory in training primarily due to the increased activations saved, and not because of the increased number of parameters - that is a secondary effect. Typical LLM models useN in andN out in the range of O(10,000) or higher, while frequently scientific operator- learning AI models like FNOs [31], Transolver [32], DoMINO [33], and others work at lower dimensional latent spaces below 1,000. Table I summarizes the impacts of parameters, opti- mizer states, and intermediate activations on memory usage, as the number of features or number of input points vary. As seen in Table I, despite being a contrived example, higher resolution data quickly outpaces the memory usage of model"},{"citing_arxiv_id":"2605.09746","ref_index":13,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"Sequential Feature Selection for Efficient Landslide Segmentation from Multi-Spectral Data","primary_cat":"cs.LG","submitted_at":"2026-05-10T20:46:30+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"Sequential Forward Floating Selection with a U-Net++ proxy identifies an 8-channel subset from multi-spectral and terrain data that matches or exceeds F1 scores of full 30-channel configurations for landslide segmentation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.07860","ref_index":56,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"On the Tradeoffs of On-Device Generative Models in Federated Predictive Maintenance Systems","primary_cat":"cs.LG","submitted_at":"2026-05-08T15:20:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Experiments on real industrial time series show that partial model sharing improves diffusion model performance in bandwidth-limited non-IID settings, while full sharing stabilizes GAN training but offers less robustness than VAE or DDPM alternatives.","context_count":1,"top_context_role":"dataset","top_context_polarity":"use_dataset","context_text":"complementary benchmark for the AD setting in this work. Among the candidate benchmarks, SWaT was selected be- cause it is the closest to the anomaly-detection protocol studied here: it supports unsupervised multivariate TSAD, contains temporally extended anomalous intervals, and allows client- wise validation and threshold selection under a federated par- tition. By contrast, C-MAPSS/N-CMAPSS [56], [57] primarily target prognostics and remaining-useful-life estimation, while SMAP/MSL [58] are relevant AD benchmarks but are less representative of the industrial PdM scenario emphasized in this paper. To emulate the same cross-silo federation setting, we set C= 5clients and partition the SWaT sequence chronologi- cally. The normal-operation portion of the dataset is split into"},{"citing_arxiv_id":"2605.03434","ref_index":51,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"Quantum Hierarchical Reinforcement Learning via Variational Quantum Circuits","primary_cat":"cs.LG","submitted_at":"2026-05-05T07:14:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Hybrid agent with variational quantum circuits for feature extraction in hierarchical RL outperforms classical baselines with 66% parameter savings, but quantum value estimation degrades results.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.25613","ref_index":8,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"One Coordinate at a Time: Convergence Guarantees for Rotosolve in Variational Quantum Algorithms","primary_cat":"quant-ph","submitted_at":"2026-04-28T13:23:15+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Rotosolve converges to ε-stationary points for smooth non-convex objectives and ε-suboptimal points under PL, with explicit worst-case rates in the finite-shot regime, outperforming or matching RCD in nuanced ways.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.00874","ref_index":52,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"Latent Space Probing for Adult Content Detection in Video Generative Models","primary_cat":"cs.CV","submitted_at":"2026-04-25T01:01:02+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Latent space probing on CogVideoX achieves 97.29% F1 for adult content detection on a new 11k-clip dataset with 4-6ms overhead.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.10199","ref_index":48,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"FatigueFusion: Latent Space Fusion for Fatigue-Driven Motion Synthesis","primary_cat":"cs.GR","submitted_at":"2026-04-11T13:12:20+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"FatigueFusion fuses fatigue features in latent space using algorithmic, data-driven, and PINN modules to synthesize novel fatigued motions from non-fatigued joint sequences in an end-to-end pipeline.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2510.23317","ref_index":45,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"Equivariance2Inverse: A Practical Self-Supervised CT Reconstruction Method Benchmarked on Real, Limited-Angle, and Blurred Data","primary_cat":"eess.IV","submitted_at":"2025-10-27T13:29:08+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Equivariance2Inverse merges equivariant imaging and sparse reconstruction into a self-supervised CT method that remains effective under scintillator blurring and limited-angle geometries, outperforming prior methods on real 2DeteCT data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2502.19056","ref_index":66,"ref_count":1,"confidence":0.35,"is_internal_anchor":false,"paper_title":"Fatigue-PINN: Physics-Informed Fatigue-Driven Motion Modulation and Synthesis","primary_cat":"cs.GR","submitted_at":"2025-02-26T11:14:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Fatigue-PINN applies physics-informed neural networks to simulate fatigue effects on human motion using a three-compartment muscle model for joint torque modulation in motion synthesis.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}