arXiv subjects

Janghwan Lee

Publications and source records attributed to Janghwan Lee.

11 recordsLinked to original sources

arXiv2026

ReQAT: Achieving Full-Precision Reasoning Accuracy with 4-bit Floating-Point Quantization-Aware Training

Large Reasoning Models (LRMs) achieve strong problem-solving through long chain-of-thought, but their deployment is constrained by the high cost of full-precision inference and growing KV cache footprints. Microscaled FP4 formats enable efficient FP4 deployment; however, fully quantizing weights, activations, and KV caches (W4A4KV4) causes severe reasoning degradation that existing PTQ and QAT fail to recover. We identify that FP4 failures concentrate on low-entropy tokens--precise symbolic commitments such as digits and operators--where quantization noise inflates sampling errors that cascade through reasoning traces. Based on this insight, we propose ReQAT, a reasoning-centric FP4 training framework with three components: (i) Trace-Aligned QAT (TAQ), which revisits identical reasoning traces to focus updates on critical low-entropy decisions; (ii) Selective Entropy Minimization (SEM), which reinforces confidence at low-entropy positions; and (iii) Q-FIT, a quantization-friendly initialization that jointly calibrates RoPE-consistent KV cache transformations to stabilize QAT. Under the same training budget, ReQAT not only recovers but surpasses BF16 fine-tuning accuracy, while delivering up to 3.9x throughput speedup on NVIDIA DGX Spark and 3.1x on B200.

cs.LG↗

arXiv2026

ReSET: Accurate Latency-Critical NVFP4 Reasoning via Step-Aware Temperature Scaling

Large reasoning models (LRMs) improve complex problem-solving by generating long intermediate reasoning traces, but this substantially increases inference costs. NVFP4 inference offers a promising approach to reduce both computational and memory costs through hardware-supported low-precision execution. However, directly applying NVFP4 to LRMs introduces two practical limitations: reasoning accuracy degrades under quantization, and existing NVFP4 kernels do not fully realize latency benefits in small-batch autoregressive decoding. In this work, we analyze the effect of NVFP4 quantization on token-level uncertainty during reasoning. We show that quantization increases incorrect sampling at low-entropy symbolic tokens, while causing over-concentration on a small set of tokens in high-uncertainty reasoning steps. Based on this observation, we propose \textbf{ReSET}, a reasoning-step entropy-based temperature-scaling method that estimates step-level uncertainty online and adapts the decoding temperature using both token-level and step-level entropy signals. To address the latency gap, we further design a CUDA-core small-$M$ NVFP4 kernel for latency-critical autoregressive decoding. Across reasoning benchmarks and model scales, ReSET improves NVFP4 reasoning accuracy by up to $\sim\!$2 points over the NVFP4 baseline. Our CUDA-core small-$M$ kernel further improves latency-critical decoding, delivering up to $2.5\!\times$ kernel-level speedup over NVFP4 vLLM and approximately $2\!\times$ end-to-end decoding speedup over BF16. Code is available at https://github.com/aiha-lab/ReSET.

cs.LG↗

arXiv2025

Patch-aware Vector Quantized Codebook Learning for Unsupervised Visual Defect Detection

Unsupervised visual defect detection is critical in industrial applications, requiring a representation space that captures normal data features while detecting deviations. Achieving a balance between expressiveness and compactness is challenging; an overly expressive space risks inefficiency and mode collapse, impairing detection accuracy. We propose a novel approach using an enhanced VQ-VAE framework optimized for unsupervised defect detection. Our model introduces a patch-aware dynamic code assignment scheme, enabling context-sensitive code allocation to optimize spatial representation. This strategy enhances normal-defect distinction and improves detection accuracy during inference. Experiments on MVTecAD, BTAD, and MTSD datasets show our method achieves state-of-the-art performance.

cs.CV↗

arXiv2024

RILQ: Rank-Insensitive LoRA-based Quantization Error Compensation for Boosting 2-bit Large Language Model Accuracy

Low-rank adaptation (LoRA) has become the dominant method for parameter-efficient LLM fine-tuning, with LoRA-based quantization error compensation (LQEC) emerging as a powerful tool for recovering accuracy in compressed LLMs. However, LQEC has underperformed in sub-4-bit scenarios, with no prior investigation into understanding this limitation. We propose RILQ (Rank-Insensitive LoRA-based Quantization Error Compensation) to understand fundamental limitation and boost 2-bit LLM accuracy. Based on rank analysis revealing model-wise activation discrepancy loss's rank-insensitive nature, RILQ employs this loss to adjust adapters cooperatively across layers, enabling robust error compensation with low-rank adapters. Evaluations on LLaMA-2 and LLaMA-3 demonstrate RILQ's consistent improvements in 2-bit quantized inference across various state-of-the-art quantizers and enhanced accuracy in task-specific fine-tuning. RILQ maintains computational efficiency comparable to existing LoRA methods, enabling adapter-merged weight-quantized LLM inference with significantly enhanced accuracy, making it a promising approach for boosting 2-bit LLM performance. Our code is available at https://github.com/aiha-lab/RILQ.

cs.LG↗

arXiv2024

AMXFP4: Taming Activation Outliers with Asymmetric Microscaling Floating-Point for 4-bit LLM Inference

As large language models (LLMs) grow in parameter size and context length, computation precision has been reduced from 16-bit to 4-bit to improve inference efficiency. However, this reduction causes accuracy degradation due to activation outliers. Rotation-based INT4 methods address this via matrix calibration, but they introduce multi-hour overheads and leave key computations in full precision. Microscaling (MX) floating-point (FP) formats offer fine-grained representation with a shared scale, enabling fully quantized matrix multiplications through direct casting without calibration. However, existing research shows unsatisfactory empirical results for MXFP4 inference, and the robustness of MX formats remains largely unexplored. In this work, we uncover the fundamental tradeoffs of the MX format: while it effectively suppresses activation outliers, it does so at the cost of increased group-wise asymmetry. To address this, we propose AMXFP4, a 4-bit asymmetric FP format that handles both issues using asymmetric shared scales, without requiring calibration. Our custom MAC engine adds negligible hardware cost while improving accuracy: AMXFP4 outperforms MXFP4 by 3% on VQA and exceeds rotation-based methods by 1.6% on CSQA. It also surpasses recently deployed commercial MXFP4 variants. Code: https://github.com/aiha-lab/MX-QLLM

cs.AI↗

arXiv2024

Efficient Generation of Molecular Clusters with Dual-Scale Equivariant Flow Matching

Amorphous molecular solids offer a promising alternative to inorganic semiconductors, owing to their mechanical flexibility and solution processability. The packing structure of these materials plays a crucial role in determining their electronic and transport properties, which are key to enhancing the efficiency of devices like organic solar cells (OSCs). However, obtaining these optoelectronic properties computationally requires molecular dynamics (MD) simulations to generate a conformational ensemble, a process that can be computationally expensive due to the large system sizes involved. Recent advances have focused on using generative models, particularly flow-based models as Boltzmann generators, to improve the efficiency of MD sampling. In this work, we developed a dual-scale flow matching method that separates training and inference into coarse-grained and all-atom stages and enhances both the accuracy and efficiency of standard flow matching samplers. We demonstrate the effectiveness of this method on a dataset of Y6 molecular clusters obtained through MD simulations, and we benchmark its efficiency and accuracy against single-scale flow matching methods.

cond-mat.mtrl-sci↗

arXiv2024

Enhancing Computation Efficiency in Large Language Models through Weight and Activation Quantization

Large Language Models (LLMs) are proficient in natural language processing tasks, but their deployment is often restricted by extensive parameter sizes and computational demands. This paper focuses on post-training quantization (PTQ) in LLMs, specifically 4-bit weight and 8-bit activation (W4A8) quantization, to enhance computational efficiency -- a topic less explored compared to weight-only quantization. We present two innovative techniques: activation-quantization-aware scaling (AQAS) and sequence-length-aware calibration (SLAC) to enhance PTQ by considering the combined effects on weights and activations and aligning calibration sequence lengths to target tasks. Moreover, we introduce dINT, a hybrid data format combining integer and denormal representations, to address the underflow issue in W4A8 quantization, where small values are rounded to zero. Through rigorous evaluations of LLMs, including OPT and LLaMA, we demonstrate that our techniques significantly boost task accuracies to levels comparable with full-precision models. By developing arithmetic units compatible with dINT, we further confirm that our methods yield a 2$\times$ hardware efficiency improvement compared to 8-bit integer MAC unit.

cs.CL↗

arXiv2024

Improving Conversational Abilities of Quantized Large Language Models via Direct Preference Alignment

The rapid advancement of large language models (LLMs) has facilitated their transformation into conversational chatbots that can grasp contextual nuances and generate pertinent sentences, closely mirroring human values through advanced techniques such as instruction tuning and reinforcement learning from human feedback (RLHF). However, the computational efficiency required for LLMs, achieved through techniques like post-training quantization (PTQ), presents challenges such as token-flipping that can impair chatbot performance. In response, we propose a novel preference alignment approach, quantization-aware direct preference optimization (QDPO), that aligns quantized LLMs with their full-precision counterparts, improving conversational abilities. Evaluated on two instruction-tuned LLMs in various languages, QDPO demonstrated superior performance in improving conversational abilities compared to established PTQ and knowledge-distillation fine-tuning techniques, marking a significant step forward in the development of efficient and effective conversational LLMs.

cs.CL↗

arXiv2024

SPADE: Sparse Pillar-based 3D Object Detection Accelerator for Autonomous Driving

3D object detection using point cloud (PC) data is essential for perception pipelines of autonomous driving, where efficient encoding is key to meeting stringent resource and latency requirements. PointPillars, a widely adopted bird's-eye view (BEV) encoding, aggregates 3D point cloud data into 2D pillars for fast and accurate 3D object detection. However, the state-of-the-art methods employing PointPillars overlook the inherent sparsity of pillar encoding where only a valid pillar is encoded with a vector of channel elements, missing opportunities for significant computational reduction. Meanwhile, current sparse convolution accelerators are designed to handle only element-wise activation sparsity and do not effectively address the vector sparsity imposed by pillar encoding. In this paper, we propose SPADE, an algorithm-hardware co-design strategy to maximize vector sparsity in pillar-based 3D object detection and accelerate vector-sparse convolution commensurate with the improved sparsity. SPADE consists of three components: (1) a dynamic vector pruning algorithm balancing accuracy and computation savings from vector sparsity, (2) a sparse coordinate management hardware transforming 2D systolic array into a vector-sparse convolution accelerator, and (3) sparsity-aware dataflow optimization tailoring sparse convolution schedules for hardware efficiency. Taped-out with a commercial technology, SPADE saves the amount of computation by 36.3--89.2\% for representative 3D object detection networks and benchmarks, leading to 1.3--10.9$\times$ speedup and 1.5--12.6$\times$ energy savings compared to the ideal dense accelerator design. These sparsity-proportional performance gains equate to 4.1--28.8$\times$ speedup and 90.2--372.3$\times$ energy savings compared to the counterpart server and edge platforms.

cs.AR↗

arXiv2023

Token-Scaled Logit Distillation for Ternary Weight Generative Language Models

Generative Language Models (GLMs) have shown impressive performance in tasks such as text generation, understanding, and reasoning. However, the large model size poses challenges for practical deployment. To solve this problem, Quantization-Aware Training (QAT) has become increasingly popular. However, current QAT methods for generative models have resulted in a noticeable loss of accuracy. To counteract this issue, we propose a novel knowledge distillation method specifically designed for GLMs. Our method, called token-scaled logit distillation, prevents overfitting and provides superior learning from the teacher model and ground truth. This research marks the first evaluation of ternary weight quantization-aware training of large-scale GLMs with less than 1.0 degradation in perplexity and achieves enhanced accuracy in tasks like common-sense QA and arithmetic reasoning as well as natural language understanding. Our code is available at https://github.com/aiha-lab/TSLD.

cs.CL↗

arXiv2023

SHAPNN: Shapley Value Regularized Tabular Neural Network

We present SHAPNN, a novel deep tabular data modeling architecture designed for supervised learning. Our approach leverages Shapley values, a well-established technique for explaining black-box models. Our neural network is trained using standard backward propagation optimization methods, and is regularized with realtime estimated Shapley values. Our method offers several advantages, including the ability to provide valid explanations with no computational overhead for data instances and datasets. Additionally, prediction with explanation serves as a regularizer, which improves the model's performance. Moreover, the regularized prediction enhances the model's capability for continual learning. We evaluate our method on various publicly available datasets and compare it with state-of-the-art deep neural network models, demonstrating the superior performance of SHAPNN in terms of AUROC, transparency, as well as robustness to streaming data.

cs.LG↗