Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.
arXiv:2609.37147v1 Announce Type: cross Abstract: Distributional Diffusion Models (DDMs) replace the standard mean-prediction denoiser with a \emph{distributional} denoiser trained via a scoring rule...
arXiv:2609.37226v1 Announce Type: cross Abstract: Answering questions and completing tasks over large document collections often requires connecting evidence spread across multiple documents, such as...
arXiv:2609.37294v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used for software engineering tasks, yet their stochastic behavior challenges the validity, reproducibi...
arXiv:2609.37304v1 Announce Type: new Abstract: Large reasoning models improve performance on challenging problems by allocating additional computation before answering, but longer reasoning does not...
arXiv:2609.37359v1 Announce Type: cross Abstract: Coding agents can now write, run, and debug programs with little human help. Robot tasks, however, are usually specified by a sentence that leaves ou...
arXiv:2609.37469v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) grounds large language models in external sources, but retrieved passages often name the right entities without...
arXiv:2609.37624v1 Announce Type: cross Abstract: Fine-tuning a language model on a narrow set of harmful demonstrations, such as bad medical advice, can make it broadly misaligned on unrelated quest...
arXiv:2609.37824v1 Announce Type: cross Abstract: Transformer language models build predictions through successive residual updates, but how their representations become specific to an eventual outco...
arXiv:2609.37882v1 Announce Type: cross Abstract: Every text classifier for an African language begins with a budgeting question: how many labelled examples are needed, and can labels from other Afri...
arXiv:2609.37889v1 Announce Type: cross Abstract: Multimodal continual instruction tuning (MCIT) aims to enable multimodal large language models to acquire new capabilities from sequential tasks whil...
arXiv:2609.37891v1 Announce Type: cross Abstract: Current pre-training datasets are derived from web crawls, with all their issues, and were not designed to support mid- and post-training pipelines--...
arXiv:2609.38058v1 Announce Type: cross Abstract: Time Series Foundation Models (TSFMs) currently provide state-of-the-art results in forecasting tasks. They are available out-of-the-box and rely on...
arXiv:2609.38165v1 Announce Type: cross Abstract: The landscape of satellite imagery time series datasets and boundary-pushing architectures for cropland segmentation has never been richer. However,...
The paper introduces Global Relative Kinetic Utility (Global RKU), a label‑free method for calibrating cross‑layer credit in global structured pruning of large language models. Global RKU estimates channel importance via a final‑hidden‑state activation‑gradient signal and applies block‑relative normalization to remove block‑common scale while preserving within‑block ordering, enabling a single‑stage static pruning topology. Experiments on Qwen‑2.5‑7B show significant performance gains at various sparsity levels, and ablation studies confirm the effectiveness of the relative‑normalization step.
arXiv:2605.12070v3 Announce Type: replace-cross Abstract: Asynchronous reinforcement learning improves rollout throughput for large language model agents by decoupling sample generation from policy o...
The paper investigates the impact of low‑resolution data on machine learning models, especially when high‑resolution data is scarce. It provides a theoretical analysis using Kullback‑Leibler divergence to quantify how data resolution affects a datapoint’s influence and derives bounds on information loss from downsampling. Empirical results with a vision transformer and a convolutional neural network show that incorporating low‑resolution data consistently improves performance in such settings.
The paper investigates how the choice of graph tokenization affects transformer expressivity. It analyzes three tokenization families—spectral, random‑walk, and adjacency—showing that each induces different depth requirements and that some tokenizations are inherently lossy or ill‑conditioned for certain tasks. The authors prove lower bounds and impossibility results for converting between tokenizations and validate these findings with experiments on synthetic and real‑world data.
arXiv:2605.23200v2 Announce Type: replace-cross Abstract: The linear growth of the Key-Value (KV) cache is a critical bottleneck in long-form LLM inference. Existing KV compression methods mitigate t...
The paper introduces RAISE, a diagnostic framework that tests whether a costly large language model (LLM) signal provides enough pre-call information to justify selective use. It identifies the failure mode of acquisition collapse, where an LLM appears useful overall but lacks actionable evidence for individual decisions. The authors demonstrate RAISE with Structured Hypothesis Embeddings (SHE) and evaluate it across multiple study designs, showing that predictable incremental benefit, rather than average lift, indicates recoverable selective value.
The paper presents a Physics‑Conditioned Neural Operator (PCNO), a transformer‑based model that predicts mitral and tricuspid valve leaflet displacement, strain, and stress fields conditioned on systolic blood pressure and tissue properties. Trained on FEBio finite‑element simulations, PCNO achieves a displacement error of 4.48 % and maintains mean errors of unsupervised geometric measures within 3.5 % even when pressure and material parameters lie outside the training set, demonstrating a conditioned solution operator rather than simple interpolation. Compared to graph neural network baselines, PCNO shows superior accuracy, particularly in stress predictions.