2-bit quantized reasoning models exhibit process failures like loops and delayed commitment that degrade end-to-end performance, but FP16 planning and loop rescue recover accuracy on MATH-500 from 17.2% to 74.2% for Qwen3-8B while retaining speed gains.
arXiv preprint arXiv:2408.08554 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
LBLLM achieves better accuracy than prior binarization methods for LLMs by decoupling weight and activation quantization through initialization, layer-wise distillation, and learnable activation scaling.
citing papers explorer
-
Extreme Low-Bit Inference in Reasoning Models: Failure Modes and Targeted Recovery
2-bit quantized reasoning models exhibit process failures like loops and delayed commitment that degrade end-to-end performance, but FP16 planning and loop rescue recover accuracy on MATH-500 from 17.2% to 74.2% for Qwen3-8B while retaining speed gains.
-
LBLLM: Lightweight Binarization of Large Language Models via Three-Stage Distillation
LBLLM achieves better accuracy than prior binarization methods for LLMs by decoupling weight and activation quantization through initialization, layer-wise distillation, and learnable activation scaling.