31 Oct 2026
Calling a tool through generated text adds decoding steps, even when the computation itself is short. We study an alternative: a small neural module that runs inside a language model's forward pass.
Trending research and the full catalog - each paper linked to the benchmarks, methods, and models it introduces.
Filtering here covers the 2,000 most recent papers, as much as one page can hold in memory. See the full index of 23,060 papers.
31 Oct 2026
Calling a tool through generated text adds decoding steps, even when the computation itself is short. We study an alternative: a small neural module that runs inside a language model's forward pass.
30 Oct 2026
Robot policy learning is dominated by zeroth-order methods such as PPO, which treat the simulator as a black box and pay for it with enormous sample counts. First-order policy learning differentiates through the dynamics, but on contact-rich whole-body control it has rarely…
30 Oct 2026
Distribution Matching Distillation (DMD) trains a few-step student using the difference between a teacher's score and an online critic's estimate of the student distribution's score.
30 Oct 2026
Large language models (LLMs) are increasingly used to assist spreadsheet users with tasks such as spreadsheet model construction, formula generation, and data analysis, yet their interaction with real-world spreadsheets remains limited by the mismatch between the structure of…
30 Oct 2026
We develop a finite-dimensional orbit framework for constructing data-adapted dictionaries, sparse realizations, and predictive extensions. The method uses wrapped weighted shifts to generate orbit dictionaries and local interpolation maps from independent orbit representatives,…
30 Oct 2026
Bias in transformer-based models is a growing concern as these architectures become more prevalent in automated decision-making systems, such as prediction models for hiring, lending, and clinical treatment recommendation.
30 Oct 2026
Fuzzy clustering provides a robust framework for handling ambiguous and overlapping data boundaries, yet its traditional formulation generates entirely dense membership assignments that limit noise resilience and prevent estimating the true number of clusters, when it is…
30 Oct 2026
Spline representations provide a compact way to approximate continuous trajectories using a finite number of parameters. When approximating trajectories generated by dynamical systems, the efficiency of a spline representation depends strongly on the placement of knot points.
30 Oct 2026
Why does length generalization (LG) depend so sharply on the positional encoding (PE)? Relative schemes extrapolate fixed offsets, absolute schemes fix positions, and neither handles growing offsets.
30 Oct 2026
We study how the layers in deep neural networks (NNs) converge, using Wilsonian Renormalization-Group (RG) ideas to organize the spectral picture. During training, these layers often develop heavy-tailed power-law (PL) spectra, and the strongest checkpoints often appear near the…
29 Oct 2026
Coding agents increasingly rely on automated approvers to decide whether a proposed shell command may run. Existing approvers, from rule-based checkers to guardrail models and deployed permission classifiers, base their decisions on the text of the request and command.
29 Oct 2026
In recent years, a growing number of online retail companies have embraced deep learning models for demand forecasting, leveraging their scalability in processing large-scale datasets.
29 Oct 2026
Large language models (LLMs) are increasingly used as computational models of human language processing, yet it remains unclear which components of their representations drive alignment with neural activity.
29 Oct 2026
Increasingly, users express their needs as natural-language intents rather than structured requests. These intents may be incomplete, high-level, or internally conflicting.
29 Oct 2026
Early detection of atrial fibrillation (AFIB), characterized by irregular heart rhythms, can potentially save thousands of lives annually [20]. Numerous studies have attempted to address this challenge using various algorithms and methodologies.
29 Oct 2026
In multi-output structured prediction tasks, while only one ground truth label may be provided in the training data, multiple equally valid outputs may be possible, making reliable evaluation a persistent challenge.
29 Oct 2026
Filtering has not yet had its foundation-model moment: state estimation is still dominated either by online Bayesian filtering algorithms that require known dy- namics as inputs, or by learning-based neural filters trained from scratch for a single system or parametric family.
29 Oct 2026
In this work, we provide a formal framework and mathematical definition for the subset of foundation models focused on enabling the property of "any-task inference" (ATI) (similar to zero-shot inference), rather than those designed for pre-training fine-tuning.
28 Oct 2026
Large language models (LLMs) frequently fail at multi-digit arithmetic, but the underlying mechanisms governing these failures remain poorly understood. While recent work demonstrates that addition errors follow a "noisy quantisation" model localised around carry…
28 Oct 2026
Language models increasingly create 3D content through code, with practical workflows often relying on editing existing programs rather than regenerating them, for speed and precise control.
28 Oct 2026
Frontier large language models write correct CUDA kernels for most single-operator benchmark tasks, yet their accuracy collapses on accelerators whose architecture and programming language they have rarely seen.
28 Oct 2026
Unlearnable data methods (ULD) protect public data from unauthorized model training by adding imperceptible perturbations that create a learning shortcut. A deep neural network can exploit the shortcut learning path to minimize training loss without learning generalizable…
28 Oct 2026
Physics-informed holomorphic neural networks (PIHNNs) exploit complex-analytic representations to satisfy homogeneous PDEs by construction, but their extension to nonhomogeneous problems is nontrivial.
28 Oct 2026
Adversarial attacks on graph neural networks (GNNs) have been studied as a model-centric optimization problem driven by gradients, learned policies, or meta-learning, with the graph treated as a passive substrate.