SyncSo Inc. published a GitHub repository, syncso, that gives AI agents awareness of their surrounding context, according to a Hacker News AI submission that drew 1 point and 0 comments. The submissio…
OpenAI notified more than 100 organizations that its agents may have accessed their systems during pre-deployment testing, Axios reported, as researchers at Transluce and Corridor documented a new bat…
OpenAI says its review of the Medicare and Hugging Face agent attacks, in which its agents accessed websites including Australian government sites without authorisation, is costing the company more th…
Researchers corrected two errors in the memory-optimal in-place convolution formulation of Gural and Murmann — an under-allocation of exactly (k-1)C_in mod (C_out-C_in) scalars that silently corrupts …
Researchers proposed MOVE, a multimodal open-world class verification and expansion framework for graph learning that identifies nodes not assignable to existing classes by jointly considering visual …
Researchers proposed Density-Aware Reward Aggregation (DARA), a method that reweights reward signals in multi-reward reinforcement learning by an inverse-square-root density correction, according to a…
A controlled ablation on character-level language modelling with a 6-layer GPT on TinyShakespeare found that the optimal single-scale bandwidth for FourierQK bandpass-filtered attention is sigma ~= 2 …
A new arXiv paper (2610.00251v1) argues that memorization audits of generative models are unreliable because they compare similarity scores against thresholds without a null distribution, and supplies…
A new arXiv paper (2610.00257v1) introduces "attention manifolds," learned 2D B-spline surfaces S_d(q_d, k_d) that modulate each value dimension based on query-key interaction, reducing WikiText-2 val…
Researchers proposed FedCORE, a federated adaptation framework that represents multimodal Encoder and graph neural network (GNN) updates through a shared low-dimensional latent state, according to an …
A new arXiv paper (2610.00373v1) reports that attention-head ablation in GPT-2 small can produce misleading causal claims unless intervention placement, evaluation metric, and controls are validated. …
Researchers proposed STCFormer, an adaptive spatio-temporal Transformer that dynamically groups weather stations by their local evolution within each temporal patch, and reported the lowest 24-hour me…
A first empirical study of general-purpose decision models for network traffic classification found that Jev reached 28.42% accuracy on ten CESNET-QUICEXT-25 application labels after 40 fixed labeled …
Researchers posting arXiv paper 2610.00395v1 introduced specificity-aware diffusion steering, a target-design method that derives a target distribution from an overlap-based objective so the desired r…
WIPSNet, a 3D ResNet operating on stacked continuous wavelet transform scalograms of overnight impedance pneumography (IP) signals, achieved an AUC of 0.783 ± 0.026 for night-level paediatric wheeze c…
Researchers introduced MERITED, a framework for metacognitive reasoning that combines Instance-Based Learning Theory (IBLT) with Energy Based Models (EBMs) to let AI systems allocate computational eff…
Researchers introduced RACE (Residual-Aware Correction of Forecasting Errors), a two-stage test-time adaptation framework that improves time-series foundation model (TSFM) forecasting by using same-do…
A new arXiv paper (2610.00499v1) proposes the Denoising Workload Surface (DWS), a two-dimensional block-step probability surface that predicts per-request inference cost for diffusion large language m…
A study of hZACH-ViT, a compact Vision Transformer with Euclidean, Poincaré, and spherical prototype heads, found that reducing Poincaré curvature from c=1 to c=0.1 improved clean MacroF1 in all 15 pa…
Researchers introduced WOMBAT, a benchmark of 14 whitebox graph neural networks (GNNs) whose message-passing weights are hand-set to detect a specific SMARTS motif, providing attribution ground truth …