LLM Digest

Story

arxiv_cs_cl · Nov 6, 2026 · paper

Source brief

CodeQuant2: Enhanced Outlier Mitigation and Adaptive Clustering for Low-Precision LRMs

arxiv.orgNov 6, 2026
original source linked

In brief

Large reasoning models (LRMs) improve task performance by generating longer intermediate reasoning traces, but this substantially increases inference costs and makes low-precision quantization increasingly important....

Read the original at arxiv.org →Open in live feed Read that day’s brief

CodeQuant2: Enhanced Outlier Mitigation and Adaptive Clustering for Low-Precision LRMs

Earlier in this thread 4 items

Query-Guided RoPE Modulation for Lost-in-the-Middle Mitigation in Long-Context LLMs

Adaptive Scheduling Pipeline for Multi-Instance Asynchronous Reinforcement Learning

Route to Remember: Adaptive Memory Selection for LLM Agents

VIGOR: Visual Gain Ordering for Hallucination Mitigation in Multimodal Discrete Diffusion Language Models