On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,982 papers · 148 categories

Trend · papers per month

count % of corpus

7 results for “fp16”

Winograd Convolution for DNNs: Beyond linear polynomialscs.LG

Winograd convolution improves DNN accuracy in fp16 and bf16 formats.

problem Improving DNN accuracy in floating point formats.

method Investigated a wider range of Winograd algorithms for DNNs.

result Significant improvement in FP accuracy in fp16 and bf16 formats.

Winograd convolution DNNs Floating point ImageNet fp16

2019-05-13abs ↗pdf ↗

Mapping the Schedule x Bit-Width Boundary in Sub-100M Quantisation-Aware Trainingcs.LG

Study finds optimal learning rate schedules for sub-100M quantization-aware training across bit-widths.

problem Optimal learning rate schedules for quantization-aware training depend on bit-width.

method Factorial grid testing over bit-width, warmdown fraction, LR magnitude, model size, and seed.

result INT6 QAT requires a different schedule than higher-precision training, falsifying the primary hypothesis.

quantization-aware training learning rate schedules sub-100M models bit-width warmdown fraction

2026-05-25abs ↗pdf ↗

A Study of BFLOAT16 for Deep Learning Trainingcs.LG

BFLOAT16 achieves SOTA results in deep learning training without hyper-parameter tuning.

problem Ensuring deep learning training achieves SOTA results without hyper-parameter tuning.

method Implemented a method to emulate BFLOAT16 operations in various frameworks.

result BFLOAT16 achieves SOTA results in deep learning training without hyper-parameter tuning.

BFLOAT16 Deep Learning TensorFlow Caffe2 IntelCaffe

2019-05-29abs ↗pdf ↗

Skein and cluster algebras of unpunctured surfaces for

\mathfrak{sp}_4

math.GT

Study of $\mathfrak{sp}_4$ -webs on surfaces, proving cluster algebra structure.

problem Understanding

\mathfrak{sp}_4

-webs and their cluster algebra properties.

method Introduced skein algebra and cluster structure, proved positivity.

result Proved

\mathscr{S}_{\mathfrak{sp}_4,Σ}^{\mathbb{Z}_q}[\partial^{-1}]

is a quantum cluster algebra.

skein algebra cluster algebra quantum cluster webs moduli space

2022-07-04abs ↗pdf ↗

ZeroQuant(4+2): Redefining LLMs Quantization with a New FP6-Centric Strategy for Diverse Generative Taskscs.CL

FP6 quantization outperforms INT4 in diverse generative tasks for LLMs.

problem Limited performance of 4-bit quantization methods in diverse generative tasks.

method Proposes a 4+2 design for FP6 quantization to achieve similar latency to INT4.

result FP6 quantization outperforms INT4 in various generative tasks, including code generation and summarization.

LLMs Quantization FP6 INT4 Generative Tasks

2023-12-14abs ↗pdf ↗

Real-time Person Re-identification at the Edge: A Mixed Precision Approachcs.LG

MobileNet-v2 improves re-ID on edge devices with mixed precision.

problem Real-time person re-identification in edge devices with limited resources.

method Mixed precision training (FP32 and FP16) for MobileNet-v2 on edge nodes.

result Mixed precision training improves inference throughput and reduces power consumption.

person re-identification MobileNet-v2 mixed precision real-time edge devices

2019-08-19abs ↗pdf ↗

FibQuant: Universal Vector Quantization for Random-Access KV-Cache Compressioncs.AI

FibQuant improves KV-cache compression for long-context inference.

problem Memory traffic bottleneck in long-context inference due to KV cache growth.

method Introduces FibQuant, a universal vector quantizer that combines Beta-quantile radii, Fibonacci/Roberts-Kronecker directions, and Lloyd-Max refinement.

result FibQuant achieves high compression rates with minimal loss in attention cosine similarity.

vector quantization KV cache compression long-context inference memory traffic universal quantizer

2026-05-12abs ↗pdf ↗