arXiv subjects

Hanling Yi

Publications and source records attributed to Hanling Yi.

13 recordsLinked to original sources

arXiv2026

OCGQuant: Outlier-Companion Grouping for NVFP4 Quantization

NVFP4 is an efficient microscaling format for low-bit inference, but activation outliers can still degrade quantization accuracy within NVFP4 blocks. Within each quantization block, large activations can dominate the block scale, increasing the quantization error of the remaining values sharing the same scale. Existing post-training quantization (PTQ) methods mitigate outlier errors through strategies such as mixed precision, rotation, or residual compensation, but these approaches are either not specifically tailored to NVFP4 or introduce additional computation. In this work, we revisit NVFP4 from a channel-grouping perspective and define the reducible error incurred by remaining block values under the scale set by the block maximum as Collateral Quantization Error. Based on this insight, we propose OCGQuant, a post-training quantization method centered on Outlier-Companion Grouping (OCG), which adaptively pairs outlier channels with low-magnitude companion channels to improve NVFP4 activation block composition. Experiments on Llama3 and Qwen3 show that OCGQuant achieves the lowest WikiText-2 perplexity and highest average downstream accuracy among evaluated PTQ methods, while maintaining prefill speedup close to RTN and matching its peak decoding memory. Code is available at https://github.com/Eshamont/OCGQuant.

cs.CL↗

arXiv2026

Agentic Kernel Optimization: Generating State-of-the-Art GPU Kernels Without Hand-Written CUDA

We study whether general-purpose code agents can produce state-of-the-art GPU kernels without any manually written CUDA code. We investigate this question using representative workloads from FlashInfer-Bench, focusing on the Fused MoE, DSA TopK Indexer, and DSA Sparse Attention, and evaluate all generated kernels under the correctness-gated FlashInfer-Bench protocol on NVIDIA B200 GPUs. Starting from the PyTorch implementations, workload definitions, benchmark commands, and a compact set of CUDA optimization skills, we build a kernel optimization workflow in Houmao, a multi-agent orchestration framework for heterogeneous coding agents, to generate, debug, profile, and optimize the kernels. Humans remain strictly in an orchestration role: defining the workflow, enforcing correctness and anti-hacking constraints, supplying key references, and redirecting the search when progress stalls, without reviewing or editing the kernel code itself. Across roughly 1.9 billion agent tokens, the resulting kernels achieve speedups of 92.68x on Fused MoE, 1101.02x on DSA TopK Indexer, and 181.35x on DSA Sparse Attention relative to the PyTorch reference implementations, while also significantly outperforming the corresponding FlashInfer baselines. In the official evaluation of the MLSys 2026 FlashInfer AI Kernel Generation Contest, our generated Fused MoE kernel achieves a 1.71x speedup over the FlashInfer baseline, exceeding the top result of the Fused MoE agent-assisted track, which reports a 1.68x speedup. These results suggest that, under a disciplined correctness-first workflow, code agents can serve as effective autonomous optimizers for modern GPU kernel development.

cs.DC↗

arXiv2026

Bridging Coarse and Fine Recognition: A Hybrid Approach for Open-Ended Multi-Granularity Object Recognition in Interactive Educational Games

Recent advances in Multimodal Large Language Models (MLLMs) have enabled open-ended object recognition, yet they struggle with fine-grained tasks. In contrast, CLIP-style models excel at fine-grained recognition but lack broad coverage of general object categories. To bridge this gap, we propose \textbf{HyMOR}, a \textbf{Hy}brid \textbf{M}ulti-granularity open-ended \textbf{O}bject \textbf{R}ecognition framework that integrates an MLLM with a CLIP model. In HyMOR, the MLLM performs open-ended and coarse-grained object recognition, while the CLIP model specializes in fine-grained identification of domain-specific objects such as animals and plants. This hybrid design enables accurate object understanding across multiple semantic granularities, serving as a robust perceptual foundation for downstream multi-modal content generation and interactive gameplay. To support evaluation in content-rich and educational scenarios, we introduce TBO (TextBook Objects), a dataset containing 20,942 images annotated with 8,816 object categories extracted from textbooks. Extensive experiments demonstrate that HyMOR narrows the fine-grained recognition gap with CLIP to 0.2\% while improving general object recognition by 2.5\% over a baseline MLLM, measured by average Sentence-BERT (SBert) similarity. Overall, HyMOR achieves a 23.2\% improvement in average SBert across all evaluated datasets, highlighting its effectiveness in enabling accurate perception for multi-modal game content generation and interactive learning applications.

cs.CV↗

arXiv2024

Generation Meets Verification: Accelerating Large Language Model Inference with Smart Parallel Auto-Correct Decoding

This research aims to accelerate the inference speed of large language models (LLMs) with billions of parameters. We propose \textbf{S}mart \textbf{P}arallel \textbf{A}uto-\textbf{C}orrect d\textbf{E}coding (SPACE), an innovative approach designed for achieving lossless acceleration of LLMs. By integrating semi-autoregressive inference and speculative decoding capabilities, SPACE uniquely enables autoregressive LLMs to parallelize token generation and verification. This is realized through a specialized semi-autoregressive supervised fine-tuning process that equips existing LLMs with the ability to simultaneously predict multiple tokens. Additionally, an auto-correct decoding algorithm facilitates the simultaneous generation and verification of token sequences within a single model invocation. Through extensive experiments on a range of LLMs, SPACE has demonstrated inference speedup ranging from 2.7x-4.0x on HumanEval-X while maintaining output quality.

cs.CL↗

arXiv2024

BiTA: Bi-Directional Tuning for Lossless Acceleration in Large Language Models

Large language models (LLMs) commonly employ autoregressive generation during inference, leading to high memory bandwidth demand and consequently extended latency. To mitigate this inefficiency, we present Bi-directional Tuning for lossless Acceleration (BiTA), an innovative method expediting LLMs via streamlined semi-autoregressive generation and draft verification. Inspired by the concept of prompt tuning, we enhance LLMs with a parameter-efficient design called bi-directional tuning for the capability in semi-autoregressive generation. Employing efficient tree-based decoding, the models perform draft candidate generation and verification in parallel, ensuring outputs identical to their autoregressive counterparts under greedy sampling. BiTA serves as a lightweight plug-in module, seamlessly boosting the inference efficiency of existing LLMs without requiring additional assistance models or incurring significant extra memory costs. Applying the proposed BiTA, LLaMA-2-70B-Chat achieves a 2.7$\times$ speedup on the MT-Bench benchmark. Extensive experiments confirm our method surpasses state-of-the-art acceleration techniques.

cs.CL↗

arXiv2022

FaceMap: Towards Unsupervised Face Clustering via Map Equation

Face clustering is an essential task in computer vision due to the explosion of related applications such as augmented reality or photo album management. The main challenge of this task lies in the imperfectness of similarities among image feature representations. Given an existing feature extraction model, it is still an unresolved problem that how can the inherent characteristics of similarities of unlabelled images be leveraged to improve the clustering performance. Motivated by answering the question, we develop an effective unsupervised method, named as FaceMap, by formulating face clustering as a process of non-overlapping community detection, and minimizing the entropy of information flows on a network of images. The entropy is denoted by the map equation and its minimum represents the least description of paths among images in expectation. Inspired by observations on the ranked transition probabilities in the affinity graph constructed from facial images, we develop an outlier detection strategy to adaptively adjust transition probabilities among images. Experiments with ablation studies demonstrate that FaceMap significantly outperforms existing methods and achieves new state-of-the-arts on three popular large-scale datasets for face clustering, e.g., an absolute improvement of more than $10\%$ and $4\%$ comparing with prior unsupervised and supervised methods respectively in terms of average of Pairwise F-score. Our code is publicly available on github.

cs.CV↗

arXiv2021

Enhancing Social Relation Inference with Concise Interaction Graph and Discriminative Scene Representation

There has been a recent surge of research interest in attacking the problem of social relation inference based on images. Existing works classify social relations mainly by creating complicated graphs of human interactions, or learning the foreground and/or background information of persons and objects, but ignore holistic scene context. The holistic scene refers to the functionality of a place in images, such as dinning room, playground and office. In this paper, by mimicking human understanding on images, we propose an approach of \textbf{PR}actical \textbf{I}nference in \textbf{S}ocial r\textbf{E}lation (PRISE), which concisely learns interactive features of persons and discriminative features of holistic scenes. Technically, we develop a simple and fast relational graph convolutional network to capture interactive features of all persons in one image. To learn the holistic scene feature, we elaborately design a contrastive learning task based on image scene classification. To further boost the performance in social relation inference, we collect and distribute a new large-scale dataset, which consists of about 240 thousand unlabeled images. The extensive experimental results show that our novel learning framework significantly beats the state-of-the-art methods, e.g., PRISE achieves 6.8$\%$ improvement for domain classification in PIPA dataset.

cs.CV↗

arXiv2020

Spatio-Temporal Hybrid Graph Convolutional Network for Traffic Forecasting in Telecommunication Networks

Telecommunication networks play a critical role in modern society. With the arrival of 5G networks, these systems are becoming even more diversified, integrated, and intelligent. Traffic forecasting is one of the key components in such a system, however, it is particularly challenging due to the complex spatial-temporal dependency. In this work, we consider this problem from the aspect of a cellular network and the interactions among its base stations. We thoroughly investigate the characteristics of cellular network traffic and shed light on the dependency complexities based on data collected from a densely populated metropolis area. Specifically, we observe that the traffic shows both dynamic and static spatial dependencies as well as diverse cyclic temporal patterns. To address these complexities, we propose an effective deep-learning-based approach, namely, Spatio-Temporal Hybrid Graph Convolutional Network (STHGCN). It employs GRUs to model the temporal dependency, while capturing the complex spatial dependency through a hybrid-GCN from three perspectives: spatial proximity, functional similarity, and recent trend similarity. We conduct extensive experiments on real-world traffic datasets collected from telecommunication networks. Our experimental results demonstrate the superiority of the proposed model in that it consistently outperforms both classical methods and state-of-the-art deep learning models, while being more robust and stable.

cs.LG↗

arXiv2019

Efficient and Robust Equilibrium Strategies of Utilities in Day-ahead Market with Load Uncertainty

We consider the scenario where $N$ utilities strategically bid for electricity in the day-ahead market and balance the mismatch between the committed supply and actual demand in the real-time market, with uncertainty in demand and local renewable generation in consideration. We model the interactions among utilities as a non-cooperative game, in which each utility aims at minimizing its per-unit electricity cost. We investigate utilities' optimal bidding strategies and show that all utilities bidding according to (net load) prediction is a unique pure strategy Nash Equilibrium with two salient properties. First, it incurs no loss of efficiency; hence, competition among utilities does not increase the social cost. Second, it is robust and (0, $N-1$) fault immune. That is, fault behaviors of irrational utilities only help to reduce other rational utilities' costs. The expected market supply-demand mismatch is minimized simultaneously, which improves the planning and supply-and-demand matching efficiency of the electricity supply chain. We prove the results hold under the settings of correlated prediction errors and a general class of real-time spot pricing models, which capture the relationship between the spot price, the day-ahead clearing price, and the market-level mismatch. Simulations based on real-world traces corroborate our theoretical findings. Our study adds new insights to market mechanism design. In particular, we derive a set of fairly general sufficient conditions for the market operator to design real-time pricing schemes so that the interactions among utilities admit the desired equilibrium.

eess.SY↗

arXiv2019

Competitive Online Optimization under Inventory Constraints

This paper studies online optimization under inventory (budget) constraints. While online optimization is a well-studied topic, versions with inventory constraints have proven difficult. We consider a formulation of inventory-constrained optimization that is a generalization of the classic one-way trading problem and has a wide range of applications. We present a new algorithmic framework, \textsf{CR-Pursuit}, and prove that it achieves the minimal competitive ratio among all deterministic algorithms (up to a problem-dependent constant factor) for inventory-constrained online optimization. Our algorithm and its analysis not only simplify and unify the state-of-the-art results for the standard one-way trading problem, but they also establish novel bounds for generalizations including concave revenue functions. For example, for one-way trading with price elasticity, the \textsf{CR-Pursuit} algorithm achieves a competitive ratio that is within a small additive constant (i.e., 1/3) to the lower bound of $\ln \theta+1$, where $\theta$ is the ratio between the maximum and minimum base prices.

cs.PF↗

arXiv2019

Balancing Cost and Dissatisfaction in Online EV Charging under Real-time Pricing

We consider an increasingly popular demand-response scenario where a user schedules the flexible electric vehicle (EV) charging load in response to real-time electricity prices. The objective is to minimize the total charging cost with user dissatisfaction taken into account. We focus on the online setting where neither accurate prediction nor distribution of future real-time prices is available to the user when making irrevocable charging decision in each time slot. The emphasis on considering user dissatisfaction and achieving optimal competitive ratio differentiates our work from existing ones and makes our study uniquely challenging. Our key contribution is two simple online algorithms with the best possible competitive ratio among all deterministic algorithms. The optimal competitive ratio is upper-bounded by \min\left\{ \sqrt{\alpha/p_{\min}},p_{\max}/p_{\min}\right\} and the bound is asymptotically tight with respect to \alpha, where p_{\max} and p_{\min} are the upper and lower bounds of real-time prices and \alpha\geq p_{\min} captures the consideration of user dissatisfaction. The bound under small and large values of \alpha suggests the fundamental difference of the problems with considering user dissatisfaction (\alpha takes small values) and without (\alpha takes large values). Along the way we also develop a general technique for designing online algorithms from an idea hinted in One-Way-Trading Problem, which can be of independent interest. Simulation results based on real-world traces corroborate our theoretical findings and show that our algorithms achieve substantial performance gain as compared to conceivable alternatives. The results also suggest that increasing EV charging rate limit decreases overall cost almost linearly.

eess.SY↗

arXiv2017

A Parallelizable Acceleration Framework for Packing Linear Programs

This paper presents an acceleration framework for packing linear programming problems where the amount of data available is limited, i.e., where the number of constraints m is small compared to the variable dimension n. The framework can be used as a black box to speed up linear programming solvers dramatically, by two orders of magnitude in our experiments. We present worst-case guarantees on the quality of the solution and the speedup provided by the algorithm, showing that the framework provides an approximately optimal solution while running the original solver on a much smaller problem. The framework can be used to accelerate exact solvers, approximate solvers, and parallel/distributed solvers. Further, it can be used for both linear programs and integer linear programs.

math.OC↗

arXiv2016

Online Offering Strategies for Storage-Assisted Renewable Power Producer in Hour-Ahead Market

A promising approach to hedge against the inherent uncertainty of renewable generation is to equip the renewable plants with energy storage systems. This paper focuses on designing profit maximization offering strategies, i.e., the strategies that determine the offering price and volume, for a storage-assisted renewable power producer that participates in hour-ahead electricity market. Designing the strategies is challenging since (i) the underlying problem is coupled across time due to the evolution of the storage level, and (ii) inputs to the problem including the renewable output and market clearing price are unknown when submitting offers. Following the competitive online algorithm design approach, we first study a basic setting where the renewable output and the clearing price are known for the next hour. We propose sOffer, a simple online offering strategy that achieves the best possible competitive ratio of O(log θ), where $θ$ is the ratio between the maximum and the minimum clearing prices. Then, we consider the case where the clearing price is unknown. By exploiting the idea of submitting multiple offers to combat price uncertainty, we propose mOffer, and demonstrate that the competitive ratio of mOffer converges to that of sOffer as the number of offers grows. Finally, we extend our approach to the scenario where the renewable output has forecasting error. We propose gOffer as the generalized offering strategy and characterize its competitive ratio as a function of the forecasting error. Our trace-driven experiments demonstrate that our algorithms achieve performance close to the offline optimal and outperform a baseline alternative significantly.

cs.GT↗