HodgeCover isolates the harmonic kernel of a simplicial Laplacian on an expert 2-complex to identify irreducible merge cycles and selects experts for aggressive compression, matching or exceeding baselines on open-weight MoE models.
hub
Awq: Activation-aware weight quantization for on-device llm compression and acceleration.Proceedings of machine learning and systems, 6:87–100
14 Pith papers cite this work. Polarity classification is still indexing.
14
Pith papers citing it
hub tools
citation-role summary
background 1
other 1
citation-polarity summary
years
2026 14representative citing papers
AS-LoRA adaptively chooses which LoRA factor to update per layer and round using a curvature-aware second-order score, eliminating reconstruction error floors and improving performance in DP federated learning.
BitsMoE uses SVD decomposition and activation-aware ILP bit allocation to quantize MoE LLMs at ultra-low bits with reduced accuracy degradation compared to GPTQ.
OScaR mitigates token norm imbalance via canalized rotation and omni-token scaling to enable near-lossless INT2 KV cache quantization with up to 3x decoding speedup and 5.3x memory reduction.
Nonlinear Bipolar Compensation with Bipolar Logarithmic Transformation reduces outlier effects in post-training quantization by performing compensation in a compressed transformed space.
PARSE trains a prompt-aware linear router on dense-model outputs to select dynamic SVD ranks, improving accuracy up to 10% at 0.6 compression ratio on LLaMA-7B while delivering 2.5x prefill and 2.4x decode speedups.
FP16 KV caching in transformers causes deterministic token divergence versus cache-free inference due to non-associative floating-point accumulation orderings.
AdaHOP applies pattern-aware Hadamard transforms and selective outlier extraction to enable from-scratch MXFP4 training of LLMs at BF16 quality with up to 3.6X memory compression and 1.46X speedup.
Mix-Quant quantizes prefilling to NVFP4 and keeps BF16 for decoding in agentic LLMs, achieving up to 3x prefilling speedup while largely preserving task performance on long-context and agentic benchmarks.
Quant.npu provides a fully static quantization pipeline for on-device LLMs on NPUs by combining rotation matrices, bit-width-aware initialization, two-stage selective optimization, and adaptive mixed precision.
HCInfer recovers up to 5.2% accuracy over compressed LLMs and delivers 10.4x speedup versus full-precision models by offloading compensation parameters to CPU with async execution on resource-limited hardware.
SnapMLA achieves up to 1.91x higher throughput in long-output MLA decoding using FP8 quantization and specialized kernels while keeping benchmark quality near the BF16 baseline.
citing papers explorer
-
HodgeCover: Higher-Order Topological Coverage Drives Compression of Sparse Mixture-of-Experts
HodgeCover isolates the harmonic kernel of a simplicial Laplacian on an expert 2-complex to identify irreducible merge cycles and selects experts for aggressive compression, matching or exceeding baselines on open-weight MoE models.
-
Adaptive Selection of LoRA Components in Privacy-Preserving Federated Learning
AS-LoRA adaptively chooses which LoRA factor to update per layer and round using a curvature-aware second-order score, eliminating reconstruction error floors and improving performance in DP federated learning.
-
BitsMoE: Efficient Spectral Energy-Guided Bit Allocation for MoE LLM Quantization
BitsMoE uses SVD decomposition and activation-aware ILP bit allocation to quantize MoE LLMs at ultra-low bits with reduced accuracy degradation compared to GPTQ.
-
OScaR: The Occam's Razor for Extreme KV Cache Quantization in LLMs and Beyond
OScaR mitigates token norm imbalance via canalized rotation and omni-token scaling to enable near-lossless INT2 KV cache quantization with up to 3x decoding speedup and 5.3x memory reduction.
-
Nonlinear Bipolar Compensation: Handling Outliers in Post-Training Quantization
Nonlinear Bipolar Compensation with Bipolar Logarithmic Transformation reduces outlier effects in post-training quantization by performing compensation in a compressed transformed space.
-
Different Prompts, Different Ranks: Prompt-aware Dynamic Rank Selection for SVD-based LLM Compression
PARSE trains a prompt-aware linear router on dense-model outputs to select dynamic SVD ranks, improving accuracy up to 10% at 0.6 compression ratio on LLaMA-7B while delivering 2.5x prefill and 2.4x decode speedups.
-
The Illusion of Equivalence: Systematic FP16 Divergence in KV-Cached Autoregressive Inference
FP16 KV caching in transformers causes deterministic token divergence versus cache-free inference due to non-associative floating-point accumulation orderings.
-
AdaHOP: Fast and Accurate Low-Precision Training via Outlier-Pattern-Aware Rotation
AdaHOP applies pattern-aware Hadamard transforms and selective outlier extraction to enable from-scratch MXFP4 training of LLMs at BF16 quality with up to 3.6X memory compression and 1.46X speedup.
-
Mix-Quant: Quantized Prefilling, Precise Decoding for Agentic LLMs
Mix-Quant quantizes prefilling to NVFP4 and keeps BF16 for decoding in agentic LLMs, achieving up to 3x prefilling speedup while largely preserving task performance on long-context and agentic benchmarks.
-
Quant.npu: Enabling Efficient Mobile NPU Inference for on-device LLMs via Fully Static Quantization
Quant.npu provides a fully static quantization pipeline for on-device LLMs on NPUs by combining rotation matrices, bit-width-aware initialization, two-stage selective optimization, and adaptive mixed precision.
-
HCInfer: An Efficient Inference System via Error Compensation for Resource-Constrained Devices
HCInfer recovers up to 5.2% accuracy over compressed LLMs and delivers 10.4x speedup versus full-precision models by offloading compensation parameters to CPU with async execution on resource-limited hardware.
-
SnapMLA: Efficient Long-Context MLA Decoding via Hardware-Aware FP8 Quantized Pipelining
SnapMLA achieves up to 1.91x higher throughput in long-output MLA decoding using FP8 quantization and specialized kernels while keeping benchmark quality near the BF16 baseline.
- ActQuant: Sub-4-bit Action-Guided Quantization for Vision-Language-Action Models
- SAFE-SVD: Sensitivity-Aware Fidelity-Enforcing SVD for Physics Foundation Models