← Back to all news

Large language models

Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.

28,308 stories · RSS feed

arXiv AI
Oct 2

ShatterQuant: Breaking Uniform Precision with Block-Wise Mixed-Precision on a Systolic Transformer Hardware Accelerator

ShatterQuant is a hardware-software co-designed framework that enables mixed-precision quantization within individual tensors by assigning different bit-widths to blocks of a weight projection. It couples precision granularity with processing element configuration, allowing each precision to determine an effective block height. The framework includes a hardware-aware post-training method based on block-level standard deviation and weight sensitivity, a ShatterQuant Transformer Accelerator supporting 1/2/4/8-bit weight precision, precision-dependent PE configuration, block rescaling, and integrated softmax and piecewise-linear nonlinearities, and an evaluation showing 1.5 TOPS, 760 GOPS/$mm^2$ area efficiency, and 2.8 TOPS/W energy efficiency on a TSMC 16nm PDK implementation.

By Mikolaj Walczak, Edward Humes, Chao Fang, Marian Verhelst, Tinoosh Mohsenin
llmsefficiencybenchmarks
More like this →
arXiv AI
Oct 2

LENS-GRF: Permutation-Invariant Lesion Evidence Network with Gated Residual Fusion for Acne Severity Grading and Multi-Rater Clinical Oracle Analysis

LENS‑GRF is a permutation‑invariant lesion evidence network that uses a Set‑Transformer and gated residual fusion to combine global facial context with localized lesion patches for four‑class acne severity grading. The framework integrates adaptive facial skin segmentation, a Vision Transformer prior, and a lesion set transformer that encodes spatial geometry, with a gating mechanism that modulates local residual contributions. In experiments on ACNE04 and PLSBRACNE01, the fully automated model achieved 80.82% accuracy, while using ground‑truth lesion annotations raised accuracy to 95.89% and a Quadratic Weighted Kappa of 0.9753; zero‑shot evaluation on the full cohort yielded 35.00% accuracy versus 42.50% for a global baseline, and oracle analyses on a 148‑subject cohort showed improved accuracy and QWK up to 47.97% and 0.5799.

By Muhammad Muhtasim Shahriar, M. F. Mridha
llmscomputer-vision
More like this →
arXiv AI
Oct 2

The Weakest Link: Distilling LLM Reasoning with Worst-Case Constrained Reinforcement Learning

The paper introduces a new approach to distill reasoning abilities from large language models (LLMs) into smaller student models by framing the task as a constrained reinforcement learning problem. It enforces a worst‑case constraint on the teacher’s log‑likelihood for every prefix of the reasoning chain, avoiding reward hacking and excessive teacher regularization. Experiments on mathematical reasoning and code generation show that this method improves the balance between accuracy and fidelity, achieving the highest rigorous reasoning success rate among evaluated settings.

By Matthieu Zimmer, Xiaotong Ji, Tu Nguyen, Haitham Bou-Ammar
llmsreinforcement-learningefficiency
More like this →
arXiv Statistics ML
Oct 2

STCFormer: Adaptive Spatio-Temporal Modeling with Dynamic Cluster Transformer for Station-based Weather Forecasting

STCFormer is an adaptive spatio‑temporal Transformer that dynamically clusters weather stations within each temporal patch, combining fine‑grained local attention inside clusters with global attention over regional summaries. The model’s design is supported by a Lipschitz upper bound that suggests robustness benefits, and it achieves the lowest 24‑hour mean squared error across eight temperature and wind forecasting tasks on three real‑world datasets, ranking first or second in 47 of 48 comparisons. Ablation studies and case analyses confirm the advantages of locally adaptive grouping and complementary local‑global interactions.

By Rongwen Li, Haixin Xie, Mingyang Wang, Hongwu Liu, Kun Fang, Changjian Chen, Zhuo Tang, Kenli Li
llms
More like this →
arXiv Statistics ML
Oct 2

IrekoGPT: Turning Structured Pruning into Post-Hoc Slimmable LLMs

IrekoGPT is a post‑hoc technique that transforms pretrained large language models into slimmable versions, enabling dynamic width adjustment during inference. It builds on SliceGPT by keeping the original projection matrices intact, thereby exposing nested subnetworks at various widths. The method enhances robustness through layer‑wise calibration across multiple compression ratios and refines downstream linear layers using gradient‑free ridge regression, yielding better performance than naive PCA‑based slimming on Llama and Qwen models, especially at high compression levels.

By Pietro Moriello, Pietro Buzzega, Angelo Porrello, Simone Calderara
llmsefficiency
More like this →
arXiv AI
Oct 2

Misalignment of Low-Loss Regions Causes Grokking

arXiv:2610.00620v1 Announce Type: cross Abstract: Grokking refers to the delayed emergence of validation-set generalization after a model has already overfit the training set. Although first observed...

By Yongding Tian, Zaid Al-Ars, Maksim Kitsak, Peter Hofstee
llms
More like this →
arXiv AI
Oct 2

Auto-Formalizing Neuro-Symbolic Predictors

arXiv:2610.01519v1 Announce Type: cross Abstract: Neuro-Symbolic (NeSy) predictors incorporate prior knowledge into the prediction process of neural networks, ensuring that outputs satisfy specified...

By Samuele Bortolotti, Weixin Chen, Han Zhao, Andrea Passerini, Stefano Teso, Antonio Vergari
llmsbenchmarks
More like this →