The Archive

The Copilot usage metrics API now shows bring your own key (BYOK) usage across GitHub Copilot, so you can separate it from Copilot-hosted model usage. Reports cover two kinds of… The post BYOK and enterprise custom models in the Copilot usage API appeared first on The GitHub Blog.

A²ODA proposes adaptive optimization with p-norm preconditioning for large-scale convex optimization without line search.

HARP reframes LLM recommendations as agentic communication, enabling symmetric user-item profiles without retraining.

Novel regularization method (GSBR) for online RL that balances imitation-learning priors with policy improvement; relevant to RL practitioners but narrow domain.

Method using activation-space geometry (anon-vectors) to detect suppressed knowledge in instruction-tuned LLMs; advances interpretability and evaluation of model internals.

Examines non-collapsibility of AUC-based fairness metrics in clinical prediction models, questioning interpretability of eight existing fairness metrics.

Hybrid neural-process models with asymmetric optimization explore equifinality in evapotranspiration partitioning without constraining internal mechanism disagreement.

Ground-Wing CFD dataset benchmarks neural surrogate models for 3D aerodynamic prediction across 770 parameterized wing configurations.

Bidirectional Centered Score Matching improves discrete diffusion distillation by aligning logit geometry instead of token-wise KL under few-step settings.

Larridin raised $30M Series A to help enterprises measure AI ROI and optimize adoption metrics.

Nonparametric neural estimator for GARCH-type processes with minimax-optimal finite-sample variance bounds.

Anatomical guidance via prostate zonal masks improves PI-RADS scoring and cancer detection in ResNet and ViT architectures.

Sensorized human demonstrations offer scalable data for mobile manipulation, but embodiment differences complicate policy learning. We present Pixel-UMI, a pipeline for learning environment-aware whole-body policies from human data. We first introduce Environment-Aware Retargeting (EAR), which combines reconstructed scene geometry with demonstrated head and gripper trajectories to determine robot posture and generate collision-aware joint and base targets. Delay-calibrated action assignment converts these trajectories into actions. However, kinematic differences can leave residual pose errors...

AI agents are increasingly deployed in settings where they interact and coordinate with other agents, yet honesty is mostly measured in single-turn, single-agent benchmarks that pressure the model to deceive. We study deception that emerges in multi-agent interaction without any instruction to deceive. We introduce Kradle, a platform for evaluating AI models, and Four Bridges, an information asymmetry game in which four agents distribute across four rooms: one is the death room, the others hold food worth more if not shared. One agent, the informed agent, knows which room is the death room an...

We show that the causal structure of spacetime can emerge within the memory of a learning agent performing a quantum information task. We train a model-based reinforcement-learning agent to maximize the quantum mutual information between localized quantum probes interacting with a quantum field in Minkowski spacetime, without directly providing it with information regarding the spacetime metric or causal structure. In learning to accomplish this task, the reinforcement-learning agent constructs an internal predictive model whose planner develops the characteristic null structure of spacetime,...

Neural operators provide efficient surrogates for computational fluid dynamics (CFD), but most are designed for static geometries and are not equipped to model unsteady flows induced by continuously moving boundaries. We study dynamic-boundary 4D CFD prediction for articulated moving geometries and introduce RoboCFD-4D, a temporally resolved robotic CFD benchmark spanning diverse morphologies and flow conditions. To address the computational and autoregressive challenges of this setting, we propose DynaSolver, a neural operator that explicitly conditions on boundary evolution and causal flow ...

We consider nonparametric maximum likelihood estimation of the intensity function of a Poisson process, allowing the intensity to depend on observed covariates. Such models describe how event rates vary with external conditions, including conditions under which the intensity becomes arbitrarily small or vanishes. However, risk analysis based on the Kullback-Leibler (KL) divergence typically requires the true intensity to be bounded away from zero, a restrictive assumption in applications. We develop a unified approach that avoids this assumption and derive an oracle inequality for the covaria...

Contact-rich manipulation benefits from tactile feedback that captures both detailed contact geometry and interaction forces. When scooping grains, contact patterns help infer spoon orientation, while resistance transmitted through the grasp guides motion. To provide these complementary sensing capabilities, we present PanoTac, a low-cost, compliant tactile fingertip designed for dense three-axis tactile sensing. Each fingertip provides a $12 \times 32 \times 3$ tactile representation, with one normal ($z$) and two shear ($x$, $y$) response components at each taxel. The sensor operates at 60 ...

Parameter-efficient continual learning on pre-trained Vision Transformers largely relies on a costly premise: preventing catastrophic forgetting requires either spawning a dedicated LoRA branch per session---causing adapter parameters to scale linearly with task horizon---or applying sparse Mixture-of-Experts gating symmetrically to both forward and backward passes, which sacrifices representational rank at test time. We argue that reading consolidated knowledge and writing new task credit are fundamentally asymmetric operations. Because visual features in a shared low-rank space distribute a...

Graph forecasters for sensor networks are trained in one regime and deployed in others, where sensor coupling and signal frequency content both change. We split this spectral shift into an operator channel and a frequency channel, each with a dimensionless measure. ST-GFM (Spectral-Temporal Graph Forecasting Model) filters every input step with Chebyshev and bidirectional diffusion filters on one learned, non-symmetric operator and alternates them with per-node temporal attention. For the operator channel we prove Frobenius-norm perturbation bounds for both graph branches, show that the Cheby...

For effective resource management, infrastructure planning, and sustainable tourism, it is really important to predict accurate visitors. This paper estimates the daily visitation of Nebraska lakes using spatial, basic temporal, and interpretable temporal features. Moreover, we evaluate the base models with multiple metrics. Using the best three models, our method will ensemble hybrid models and choose the best hybrid technique. Further, we investigate the impact of lag features and how the models work with different outlier detection techniques such as Interquartile Range (IQR) and anomaly d...

LUCA synthetic benchmark evaluates LLM agents on experimental reasoning, diagnosis, and intervention across seven frontier models with outcome-independent metrics.

TSB-SEG unifies time-series segmentation benchmarks for change-point and state detection with standardized evaluation metrics.

SF dating culture adopts startup metrics; lifestyle commentary unrelated to AI technical development.

MUPA²S: multimodal stress assessment combining facial video and ECG with asymmetric attention architecture.

Hippocampal-inspired architecture enables continual learning without replay by indexing input geometry changes across tasks.

Framework integrating Riemannian geometry into deep learning for representation learning of geometric structures.

Distinguishes task depth (sequential dependencies) from length in long-horizon benchmarks; proposes metrics to measure and synthesize task difficulty.

LM6-CastBench dataset enables thermal prediction for aluminum sand casting using geometry-aware simulation.

AStockArena benchmarks financial trading agents on real-time China A-share data, auditing trade fidelity beyond profit metrics.