L-layer transformers under Log-ICoT curriculum provably learn k-parity with poly(n) samples and log k stages, matching explicit CoT efficiency without inference overhead.
Integer quantization for deep learning inference: Principles and empirical evaluation
10 Pith papers cite this work, alongside 218 external citations. Polarity classification is still indexing.
10
Pith papers citing it
218
external citations · external index
representative citing papers
A simplified one-layer transformer provably learns parities first with explicit CoT supervision then internalizes to direct computation as CoT tokens are removed.
Dynamic quantization creates side channels allowing partial or full recovery of other users' batched data in at least four popular ML frameworks.
DharmaOCR models reach 0.925 and 0.911 extraction scores with 0.40% and 0.20% degeneration rates on a new benchmark covering printed, handwritten, and legal documents, outperforming open-source and commercial baselines via SFT plus DPO.
LLM.int8() performs 8-bit inference for transformers up to 175B parameters with no accuracy loss by combining vector-wise quantization for most features with 16-bit mixed-precision handling of systematic outlier dimensions.
QuantSR+ introduces RBD, QSA, and SFD techniques to achieve state-of-the-art accuracy-efficiency trade-offs in 2-4 bit quantized image super-resolution networks, with reported PSNR gains like 0.29 dB on Urban100 for SwinIR-S.
QuIDE defines the Intelligence Index I = (C × P) / log₂(T+1) as a unified score for the compression-accuracy-latency trade-off in quantized neural networks, with experiments showing task-dependent optimal bit widths.
FP8 formats E4M3 and E5M2 match 16-bit training accuracy on CNNs, RNNs, and Transformers up to 175B parameters without hyperparameter changes.
Knowledge distillation trains a 3.9x smaller YOLO student to retain 14.5% higher precision than direct training under INT8 quantization on BDD100K, exceeding the large teacher's FP32 precision while cutting false alarms.
Adaptive bit-length schedulers plus Laplacian DP in non-IID FL reduce communicated data by up to 52.64% on MNIST and 45% on CIFAR-10 while keeping competitive accuracy and privacy.
citing papers explorer
-
Transformers Provably Learn to Internalize Chain-of-Thought
L-layer transformers under Log-ICoT curriculum provably learn k-parity with poly(n) samples and log k stages, matching explicit CoT efficiency without inference overhead.
-
Learning through Internalization
A simplified one-layer transformer provably learns parities first with explicit CoT supervision then internalizes to direct computation as CoT tokens are removed.
-
Quantamination: Dynamic Quantization Leaks Your Data Across the Batch
Dynamic quantization creates side channels allowing partial or full recovery of other users' batched data in at least four popular ML frameworks.
-
DharmaOCR: Specialized Small Language Models for Structured OCR that outperform Open-Source and Commercial Baselines
DharmaOCR models reach 0.925 and 0.911 extraction scores with 0.40% and 0.20% degeneration rates on a new benchmark covering printed, handwritten, and legal documents, outperforming open-source and commercial baselines via SFT plus DPO.
-
LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale
LLM.int8() performs 8-bit inference for transformers up to 175B parameters with no accuracy loss by combining vector-wise quantization for most features with 16-bit mixed-precision handling of systematic outlier dimensions.
-
QuantSR+: Pushing the Limit of Quantized Image Super-Resolution Networks
QuantSR+ introduces RBD, QSA, and SFD techniques to achieve state-of-the-art accuracy-efficiency trade-offs in 2-4 bit quantized image super-resolution networks, with reported PSNR gains like 0.29 dB on Urban100 for SwinIR-S.
-
QuIDE: Mastering the Quantized Intelligence Trade-off via Active Optimization
QuIDE defines the Intelligence Index I = (C × P) / log₂(T+1) as a unified score for the compression-accuracy-latency trade-off in quantized neural networks, with experiments showing task-dependent optimal bit widths.
-
FP8 Formats for Deep Learning
FP8 formats E4M3 and E5M2 match 16-bit training accuracy on CNNs, RNNs, and Transformers up to 175B parameters without hyperparameter changes.
-
Edge AI for Automotive Vulnerable Road User Safety: Deployable Detection via Knowledge Distillation
Knowledge distillation trains a 3.9x smaller YOLO student to retain 14.5% higher precision than direct training under INT8 quantization on BDD100K, exceeding the large teacher's FP32 precision while cutting false alarms.
-
Enhanced Privacy and Communication Efficiency in Non-IID Federated Learning with Adaptive Quantization and Differential Privacy
Adaptive bit-length schedulers plus Laplacian DP in non-IID FL reduce communicated data by up to 52.64% on MNIST and 45% on CIFAR-10 while keeping competitive accuracy and privacy.