arXiv subjects

Zining Zhang

Publications and source records attributed to Zining Zhang.

15 recordsLinked to original sources

arXiv2026

TherMosaic: Accelerating Perceived Thermal Transitions Through Spatiotemporal Thermal Feedback

Thermal feedback can enrich immersive interaction, but thermoelectric devices often change temperature too slowly to match interactive timing. We present TherMosaic, a spatiotemporal thermal feedback approach that accelerates perceived temperature transitions by leveraging two perceptual mechanisms: spatial summation and thermal adaptation. Focusing on the fingertip, we first investigate this approach using a custom 2*2 array of independently controlled Peltier modules. Across three controlled perceptual studies, we show that distributed thermal stimulation can preserve stable hot and cold percepts despite local deviations, that adaptation helps maintain these percepts during changing stimulation, and that combining these effects reduces perceived transition time by about 30%-40% for transitions originating from hot or cold states. We then translate the same design principles into a standalone wearable implementation of TherMosaic and evaluate it in virtual reality. Our results show that this approach reduces perceived thermal lag and improves temporal alignment between thermal and visual events in interactive use.

cs.HC↗

arXiv2026

MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing

Memory is a fundamental component for long-context LLM agents, supporting persistent state across interactions through a continuous serve-and-update lifecycle. Despite substantial prior work, many stateful systems retain sequential autoregressive extraction or state-dependent maintenance on the write path, delaying when new evidence becomes queryable. To address these challenges, we present MemForest, a memory framework that reformulates agent memory as a write-efficient temporal data-management problem. MemForest breaks the sequential bottleneck via parallel extraction, decoupling memory construction into concurrent, independent operations. We further introduce MemTree, a hierarchical temporal index that organizes memory as time-ordered trees and replaces global rewrites with localized dirty-path refresh. Dirty summaries can be refreshed in parallel across nodes and trees. End-to-end work remains proportional to incoming content; the logarithmic bound applies only to structural insertion and level-dependent refresh depth in balanced trees. We evaluate MemForest on two long-context benchmarks, LongMemEval-S and LoCoMo. Experiments use Qwen3-4B, Qwen3-30B, and Gemma-4-12B-IT. With Qwen3-30B, MemForest reaches 81.8 percent pass at 1 on LongMemEval-S, while its input-normalized build rate is 6.0 times that of EverMemOS. On LoCoMo categories 1 to 4, it reaches 84.09 percent, within 0.13 percentage points of EverMemOS; on a matched conversation, its build rate is 9.5 times higher. These results show that MemForest reduces memory-freshness latency while retaining strong answer quality.

cs.DB↗

arXiv2026

As Content and Layout Co-Evolve: TangibleSite for Scaffolding Blind People's Webpage Design through Multimodal Interaction

Creating webpages requires generating content and arranging layout while iteratively refining both to achieve a coherent design, a process that can be challenging for blind individuals. To understand how blind designers navigate this process, we conducted two rounds of co-design sessions with blind participants, using design probes to elicit their strategies and support needs. Our findings reveal a preference for content and layout to co-evolve, but this process requires external support through cues that situate local elements within the broader page structure as well as multimodal interactions. Building on these insights, we developed TangibleSite, an accessible web design tool that provides real-time multimodal feedback through tangible, auditory, and speech-based interactions. TangibleSite enables blind individuals to create, edit, and reposition webpage elements while integrating content and layout decisions. A formative evaluation with six blind participants demonstrated that TangibleSite enabled independent webpage creation, supported refinement across content and layout, and reduced barriers to achieving visually consistent designs.

cs.HC↗

arXiv2025

XtraGPT: Context-Aware and Controllable Academic Paper Revision via Human-AI Collaboration

Despite the growing adoption of large language models (LLMs) in academic workflows, their capabilities remain limited in supporting high-quality scientific writing. Most existing systems are designed for general-purpose scientific text generation and fail to meet the sophisticated demands of research communication beyond surface-level polishing, for example, maintaining conceptual coherence across sections. Furthermore, academic writing is inherently iterative and revision-driven, a process that is not well supported by direct prompting-based paradigms. To address these scenarios, we propose a human-AI collaboration framework for academic paper revision, centered on criteria-guided intent alignment and context-aware modeling. To validate the framework, we curate a dataset of 7,000 research papers from top-tier venues, annotated with 140,000 instruction--response pairs that reflect realistic, section-level scientific revisions. We instantiate the framework in XtraGPT, the first suite of open-source LLMs (1.5B to 14B parameters) specifically fine-tuned for context-aware academic paper revision. Extensive experiments show that XtraGPT significantly outperforms same-scale baselines and rivals the quality of proprietary counterparts. Both automated preference assessments and human evaluations confirm the effectiveness of XtraGPT in improving scientific drafts. Our code and models are available at https://github.com/Xtra-Computing/XtraGPT and https://huggingface.co/collections/Xtra-Computing/xtragpt.

cs.CL↗

arXiv2025

LogQuant: Log-Distributed 2-Bit Quantization of KV Cache with Superior Accuracy Preservation

We introduce LogQuant, a groundbreaking 2-bit quantization technique for KV Cache in large language model (LLM) inference, delivering substantial memory savings while preserving superior performance. Previous methods either assume that later tokens are more important or attempt to predict important tokens based on earlier attention patterns. Both approaches, however, can result in performance bottlenecks or frequent mispredictions. LogQuant takes a different approach. By applying a log-based filtering mechanism, it selectively compresses the KV Cache across the entire context, achieving better performance with the same or even reduced memory footprint compared to existing methods. In benchmark tests, it enhances throughput by 25% and boosts batch size by 60% without increasing memory consumption. For challenging tasks such as Math and Code Completion, LogQuant improves accuracy by 40% to 200% at the same compression ratio, outperforming comparable techniques.LogQuant integrates effortlessly with popular inference frameworks like Python's transformers library. Implementation can be available in https://github.com/Concyclics/LogQuantKV.

cs.LG↗

arXiv2025

Feather-SQL: A Lightweight NL2SQL Framework with Dual-Model Collaboration Paradigm for Small Language Models

Natural Language to SQL (NL2SQL) has seen significant advancements with large language models (LLMs). However, these models often depend on closed-source systems and high computational resources, posing challenges in data privacy and deployment. In contrast, small language models (SLMs) struggle with NL2SQL tasks, exhibiting poor performance and incompatibility with existing frameworks. To address these issues, we introduce Feather-SQL, a new lightweight framework tailored for SLMs. Feather-SQL improves SQL executability and accuracy through 1) schema pruning and linking, 2) multi-path and multi-candidate generation. Additionally, we introduce the 1+1 Model Collaboration Paradigm, which pairs a strong general-purpose chat model with a fine-tuned SQL specialist, combining strong analytical reasoning with high-precision SQL generation. Experimental results on BIRD demonstrate that Feather-SQL improves NL2SQL performance on SLMs, with around 10% boost for models without fine-tuning. The proposed paradigm raises the accuracy ceiling of SLMs to 54.76%, highlighting its effectiveness.

cs.CL↗

arXiv2025

PCB Renewal: Iterative Reuse of PCB Substrates for Sustainable Electronic Making

PCB (printed circuit board) substrates are often single-use, leading to material waste in electronics making. We introduce PCB Renewal, a novel technique that "erases" and "reconfigures" PCB traces by selectively depositing conductive epoxy onto outdated areas, transforming isolated paths into conductive planes that support new traces. We present the PCB Renewal workflow, evaluate its electrical performance and mechanical durability, and model its sustainability impact, including material usage, cost, energy consumption, and time savings. We develop a software plug-in that guides epoxy deposition, generates updated PCB profiles, and calculates resource usage. To demonstrate PCB Renewal's effectiveness and versatility, we repurpose a single PCB across four design iterations spanning three projects: a camera roller, a WiFi radio, and an ESPboy game console. We also show how an outsourced double-layer PCB can be reconfigured, transforming it from an LED watch to an interactive cat toy. The paper concludes with limitations and future directions.

cs.HC↗

arXiv2024

Aggressive Post-Training Compression on Extremely Large Language Models

The increasing size and complexity of Large Language Models (LLMs) pose challenges for their deployment on personal computers and mobile devices. Aggressive post-training model compression is necessary to reduce the models' size, but it often results in significant accuracy loss. To address this challenge, we propose a novel network pruning technology that utilizes over 0.7 sparsity and less than 8 bits of quantization. Our approach enables the compression of prevailing LLMs within a couple of hours while maintaining a relatively small accuracy loss. In experimental evaluations, our method demonstrates effectiveness and potential for practical deployment. By making LLMs available on domestic devices, our work can facilitate a new era of natural language processing applications with wide-ranging impacts.

cs.CL↗

arXiv2024

Comparing Vibrotactile and Skin-Stretch Haptic Feedback for Conveying Spatial Information of Virtual Objects to Blind VR Users

Perceiving spatial information of a virtual object (e.g, direction, distance) is critical yet challenging for blind users seeking an immersive virtual reality (VR) experience. To facilitate VR accessibility for blind users, in this paper, we investigate the effectiveness of two types of haptic cues - vibrotactile and skin-stretch cues - in conveying the spatial information of a virtual object when applied to the dorsal side of a blind user's hand. We conducted a user study with 10 blind users to investigate how they perceive static and moving objects in VR with a custom-made haptic apparatus. Our results reveal that blind users can more accurately understand an object's location and movement when receiving skin-stretch cues, as opposed to vibrotactile cues. We discuss the pros and cons of both types of haptic cues and conclude with design recommendations for future haptic solutions for VR accessibility.

cs.HC↗

arXiv2024

JetUnit: Rendering Diverse Force Feedback in Virtual Reality Using Water Jets

We propose JetUnit, a water-based VR haptic system designed to produce force feedback with a wide spectrum of intensities and frequencies through water jets. The key challenge in designing this system lies in optimizing parameters to enable the haptic device to generate force feedback that closely replicates the most intense force produced by direct water jets while ensuring the user remains dry. In this paper, we present the key design parameters of the JetUnit wearable device determined through a set of quantitative experiments and a perception study. We further conducted a user study to assess the impact of integrating our haptic solutions into virtual reality experiences. The results revealed that, by adhering to the design principles of JetUnit, the water-based haptic system is capable of delivering diverse force feedback sensations, significantly enhancing the immersive experience in virtual reality.

cs.HC↗

arXiv2024

SolderlessPCB: Reusing Electronic Components in PCB Prototyping through Detachable 3D Printed Housings

The iterative prototyping process for printed circuit boards (PCBs) frequently employs surface-mounted device (SMD) components, which are often discarded rather than reused due to the challenges associated with desoldering, leading to unnecessary electronic waste. This paper introduces SolderlessPCB, a collection of techniques for solder-free PCB prototyping, specifically designed to promote the recycling and reuse of electronic components. Central to this approach are custom 3D-printable housings that allow SMD components to be mounted onto PCBs without soldering. We detail the design of SolderlessPCB and the experiments conducted to evaluate its design parameters, electrical performance, and durability. To illustrate the potential for reusing SMD components with SolderlessPCB, we discuss two scenarios: the reuse of components from earlier design iterations and from obsolete prototypes. We also provide examples demonstrating that SolderlessPCB can handle high-current applications and is suitable for high-speed data transmission. The paper concludes by discussing the limitations of our approach and suggesting future directions to overcome these challenges.

cs.HC↗

arXiv2022

HARL: Hierarchical Adaptive Reinforcement Learning Based Auto Scheduler for Neural Networks

To efficiently perform inference with neural networks, the underlying tensor programs require sufficient tuning efforts before being deployed into production environments. Usually, enormous tensor program candidates need to be sufficiently explored to find the one with the best performance. This is necessary to make the neural network products meet the high demand of real-world applications such as natural language processing, auto-driving, etc. Auto-schedulers are being developed to avoid the need for human intervention. However, due to the gigantic search space and lack of intelligent search guidance, current auto-schedulers require hours to days of tuning time to find the best-performing tensor program for the entire neural network. In this paper, we propose HARL, a reinforcement learning (RL) based auto-scheduler specifically designed for efficient tensor program exploration. HARL uses a hierarchical RL architecture in which learning-based decisions are made at all different levels of search granularity. It also automatically adjusts exploration configurations in real-time for faster performance convergence. As a result, HARL improves the tensor operator performance by 22% and the search speed by 4.3x compared to the state-of-the-art auto-scheduler. Inference performance and search speed are also significantly improved on end-to-end neural networks.

cs.LG↗

arXiv2021

TransMask: A Compact and Fast Speech Separation Model Based on Transformer

Speech separation is an important problem in speech processing, which targets to separate and generate clean speech from a mixed audio containing speech from different speakers. Empowered by the deep learning technologies over sequence-to-sequence domain, recent neural speech separation models are now capable of generating highly clean speech audios. To make these models more practical by reducing the model size and inference time while maintaining high separation quality, we propose a new transformer-based speech separation approach, called TransMask. By fully un-leashing the power of self-attention on long-term dependency exception, we demonstrate the size of TransMask is more than 60% smaller and the inference is more than 2 times faster than state-of-the-art solutions. TransMask fully utilizes the parallelism during inference, and achieves nearly linear inference time within reasonable input audio lengths. It also outperforms existing solutions on output speech audio quality, achieving SDR above 16 over Librimix benchmark.

cs.SD↗

arXiv2020

X-TaSNet: Robust and Accurate Time-Domain Speaker Extraction Network

Extracting the speech of a target speaker from mixed audios, based on a reference speech from the target speaker, is a challenging yet powerful technology in speech processing. Recent studies of speaker-independent speech separation, such as TasNet, have shown promising results by applying deep neural networks over the time-domain waveform. Such separation neural network does not directly generate reliable and accurate output when target speakers are specified, because of the necessary prior on the number of speakers and the lack of robustness when dealing with audios with absent speakers. In this paper, we break these limitations by introducing a new speaker-aware speech masking method, called X-TaSNet. Our proposal adopts new strategies, including a distortion-based loss and corresponding alternating training scheme, to better address the robustness issue. X-TaSNet significantly enhances the extracted speech quality, doubling SDRi and SI-SNRi of the output speech audio over state-of-the-art voice filtering approach. X-TaSNet also improves the reliability of the results by improving the accuracy of speaker identity in the output audio to 95.4%, such that it returns silent audios in most cases when the target speaker is absent. These results demonstrate X-TaSNet moves one solid step towards more practical applications of speaker extraction technology.

eess.AS↗

arXiv2020

GAZEV: GAN-Based Zero-Shot Voice Conversion over Non-parallel Speech Corpus

Non-parallel many-to-many voice conversion is recently attract-ing huge research efforts in the speech processing community. A voice conversion system transforms an utterance of a source speaker to another utterance of a target speaker by keeping the content in the original utterance and replacing by the vocal features from the target speaker. Existing solutions, e.g., StarGAN-VC2, present promising results, only when speech corpus of the engaged speakers is available during model training. AUTOVCis able to perform voice conversion on unseen speakers, but it needs an external pretrained speaker verification model. In this paper, we present our new GAN-based zero-shot voice conversion solution, called GAZEV, which targets to support unseen speakers on both source and target utterances. Our key technical contribution is the adoption of speaker embedding loss on top of the GAN framework, as well as adaptive instance normalization strategy, in order to address the limitations of speaker identity transfer in existing solutions. Our empirical evaluations demonstrate significant performance improvement on output speech quality and comparable speaker similarity to AUTOVC.

cs.SD↗