Derives non-asymptotic 2-norm and infinity-norm error bounds for deterministic and stochastic variants of OPTQ and Qronos PTQ algorithms.
hub
Quip#: Even better llm quantization with hadamard incoherence and lattice codebooks.arXiv preprint arXiv:2402.04396
19 Pith papers cite this work. Polarity classification is still indexing.
19
Pith papers citing it
hub tools
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
LlamaWeb is a WebGPU backend for llama.cpp that uses static memory planning, tunable kernels, and templated multi-precision support to cut memory use by 29-33% and raise decode throughput by 45-69% versus prior browser frameworks on tested hardware.
A single fused int4 KV cache kernel on Apple Silicon outperforms fp16 in latency with 3x memory compression and near-zero quality loss on tested models.
High-variance activation directions are uncorrelated with predictions, transformer blocks grow more linear with depth, and single-block linear replacement yields 34x compression on Mistral's final block at a 1.71 perplexity cost.
FlashAttention-3 achieves 1.5-2x speedup on H100 GPUs for attention, reaching 740 TFLOPs/s (75% utilization) in FP16 and near 1.2 PFLOPs/s in FP8 while cutting numerical error by 2.6x versus baseline FP8 attention.
BitNet b1.58 shows that ternary 1.58-bit LLMs can match full-precision performance at substantially lower inference cost.
GEMQ applies global LP-based expert importance estimation and router fine-tuning within progressive quantization to cut memory and speed inference in MoE LLMs with little accuracy loss.
Waterfilling rate allocation makes quantized matrix multiplication for LLMs near information-theoretically optimal, with WaterSIC being basis-free and within 0.25 bits per entry of the limit.
ScaleSearch optimizes block floating point scales via fine-grained search to cut quantization error by 27% for NVFP4, improving PTQ by up to 15 points on MATH500 for Qwen3-8B and attention PPL by 0.77 on Llama 3.1 70B.
The paper introduces the Flatness metric, derives a theory-optimal quantization solution, and presents BDQ that uses bidirectional diagonal transformations to reduce outlier impact, achieving under 1% drop at W4A4 on LLaMA-3-8B.
BitRL enables on-device RL agents via 1-bit quantized language models, delivering 10-16x memory reduction and 3-5x energy efficiency gains with 85-98% retained performance.
LLM 2-bit quantization fails via either cumulative signal degradation or early computation collapse in key components.
GSQ uses Gumbel-Softmax to optimize scalar quantization grids for LLMs, closing most of the accuracy gap to vector methods like QTIP at 2-3 bits per parameter while using symmetric scalar grids compatible with existing kernels.
Redefining residual errors to include compensation-aware discrepancies and realigning calibration to full-precision outputs improves GPTQ and GPTAQ performance on LLMs.
BTC-LLM uses a binary codebook for pattern clustering and a learnable transformation to achieve 0.7-1.11 bit LLM quantization while limiting accuracy loss to a few percent on LLaMA and Qwen models.
GAMMA is a post-training framework that learns stable module sensitivity rankings for mixed-precision LLM quantization and projects them to exact bit budgets via integer programming, enabling reuse across arbitrary memory targets.
High-rate quantization theory yields accurate approximations for the distortion of absmax INT and FP schemes in generic weight-plus-activation matrix multiplication.
DuQuant++ adapts outlier-aware fine-grained rotation to MXFP4 by matching block size to the 32-element microscaling group, enabling a single rotation that smooths distributions and achieves SOTA performance on LLaMA-3 with lower cost.
The paper surveys techniques to speed up and reduce the resource needs of LLM inference, organized by data-level, model-level, and system-level changes, with comparative experiments on representative methods.
citing papers explorer
-
Provable Post-Training Quantization: Theoretical Analysis of OPTQ and Qronos
Derives non-asymptotic 2-norm and infinity-norm error bounds for deterministic and stochastic variants of OPTQ and Qronos PTQ algorithms.
-
Llamas on the Web: Memory-Efficient, Performance-Portable, and Multi-Precision LLM Inference with WebGPU
LlamaWeb is a WebGPU backend for llama.cpp that uses static memory planning, tunable kernels, and templated multi-precision support to cut memory use by 29-33% and raise decode throughput by 45-69% versus prior browser frameworks on tested hardware.
-
When Quantization Is Free: An int4 KV Cache That Outruns fp16 on Apple Silicon
A single fused int4 KV cache kernel on Apple Silicon outperforms fp16 in latency with 3x memory compression and near-zero quality loss on tested models.
-
Variance Is Not Importance: Structural Analysis of Transformer Compressibility Across Model Scales
High-variance activation directions are uncorrelated with predictions, transformer blocks grow more linear with depth, and single-block linear replacement yields 34x compression on Mistral's final block at a 1.71 perplexity cost.
-
FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision
FlashAttention-3 achieves 1.5-2x speedup on H100 GPUs for attention, reaching 740 TFLOPs/s (75% utilization) in FP16 and near 1.2 PFLOPs/s in FP8 while cutting numerical error by 2.6x versus baseline FP8 attention.
-
The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits
BitNet b1.58 shows that ternary 1.58-bit LLMs can match full-precision performance at substantially lower inference cost.
-
GEMQ: Global Expert-Level Mixed-Precision Quantization for MoE LLMs
GEMQ applies global LP-based expert importance estimation and router fine-tuning within progressive quantization to cut memory and speed inference in MoE LLMs with little accuracy loss.
-
High-Rate Quantized Matrix Multiplication II
Waterfilling rate allocation makes quantized matrix multiplication for LLMs near information-theoretically optimal, with WaterSIC being basis-free and within 0.25 bits per entry of the limit.
-
Search Your Block Floating Point Scales!
ScaleSearch optimizes block floating point scales via fine-grained search to cut quantization error by 27% for NVFP4, improving PTQ by up to 15 points on MATH500 for Qwen3-8B and attention PPL by 0.77 on Llama 3.1 70B.
-
Theory-optimal Quantization Based on Flatness
The paper introduces the Flatness metric, derives a theory-optimal quantization solution, and presents BDQ that uses bidirectional diagonal transformations to reduce outlier impact, achieving under 1% drop at W4A4 on LLaMA-3-8B.
-
BitRL: Reinforcement Learning with 1-bit Quantized Language Models for Resource-Constrained Edge Deployment
BitRL enables on-device RL agents via 1-bit quantized language models, delivering 10-16x memory reduction and 3-5x energy efficiency gains with 85-98% retained performance.
-
From Signal Degradation to Computation Collapse: Uncovering the Two Failure Modes of LLM Quantization
LLM 2-bit quantization fails via either cumulative signal degradation or early computation collapse in key components.
-
GSQ: Highly-Accurate Low-Precision Scalar Quantization for LLMs via Gumbel-Softmax Sampling
GSQ uses Gumbel-Softmax to optimize scalar quantization grids for LLMs, closing most of the accuracy gap to vector methods like QTIP at 2-3 bits per parameter while using symmetric scalar grids compatible with existing kernels.
-
Rethinking Residual Errors in Compensation-based LLM Quantization
Redefining residual errors to include compensation-aware discrepancies and realigning calibration to full-precision outputs improves GPTQ and GPTAQ performance on LLMs.
-
BTC-LLM: Efficient Sub-1-Bit LLM Quantization via Learnable Transformation and Binary Codebook
BTC-LLM uses a binary codebook for pattern clustering and a learnable transformation to achieve 0.7-1.11 bit LLM quantization while limiting accuracy loss to a few percent on LLaMA and Qwen models.
-
GAMMA: Global Bit Allocation for Mixed-Precision Models under Arbitrary Budgets
GAMMA is a post-training framework that learns stable module sensitivity rankings for mixed-precision LLM quantization and projects them to exact bit budgets via integer programming, enabling reuse across arbitrary memory targets.
-
High-Rate Quantized Matrix Multiplication I
High-rate quantization theory yields accurate approximations for the distortion of absmax INT and FP schemes in generic weight-plus-activation matrix multiplication.
-
DuQuant++: Fine-grained Rotation Enhances Microscaling FP4 Quantization
DuQuant++ adapts outlier-aware fine-grained rotation to MXFP4 by matching block size to the 32-element microscaling group, enabling a single rotation that smooths distributions and achieves SOTA performance on LLaMA-3 with lower cost.
-
A Survey on Efficient Inference for Large Language Models
The paper surveys techniques to speed up and reduce the resource needs of LLM inference, organized by data-level, model-level, and system-level changes, with comparative experiments on representative methods.