arXiv subjects

Yongmin Li

Publications and source records attributed to Yongmin Li.

At least 19 recordsLinked to original sources

arXiv2026

Research and simulation of analytical polarization control enabled by optical computing on an integrated photonics chip

Dynamic polarization controllers are key devices with broad applications in many fields. However, most on-chip polarization controllers still rely on traditional blind-search methods, whereas analytical optical-computing approaches remain insufficiently explored, particularly with respect to calibration and endless polarization control. With the accurate relative phase of Mach-Zehnder interferometer (MZI) being fully controllable on an integrated photonics chip, we present an analytical polarization control (APC) method using four phase shifters and optical computing, eliminating the need for the traditional inefficient blind-search procedure. The basic structures and operations of APC are clarified. The proposed calibration method and endless control method enable continuous APC while compensating for phase differences within the MZI structures. We simulate the influence of the endless control unit on polarization control and quantify the effect of the fourth phase difference on the output extinction ratio. With the fourth phase shifter, the phase difference encountered during Stokes vector measurement can be effectively compensated, and rotations around all three axes on the Poincar\'e sphere can be realized. These results establish a practical APC architecture based on optical computing for photonics chips. The proposed APC methods, combined with a FPGA-based hardware acceleration, will enable high speed on-chip polarization controllers.

physics.optics↗

arXiv2026

Does Fine-Tuning Undo Activation Steering? Behavioural Recovery Without Weight-Edit Reversal

Activation steering can be embedded directly into a language model's weights, shaping behaviour without inference-time intervention and offering a way to encode alignment prior to release. However, models are routinely fine-tuned after deployment, and it is unknown whether embedded interventions survive this. We study the stability of embedded steering for refusal suppression and brevity induction across five instruction-tuned models (3B-14B) under non-adversarial SFT and RLHF. Behaviourally, preservation tracks the training data: steering degrades when optimisation pressure contradicts the targeted behaviour and persists otherwise, with refusal ablation losing 64% of its effect on average under SFT. Mechanistically, however, the weight edit survives almost untouched even where behaviour reverts: mean vector recovery is $\rho = 0.004$, and the fine-tuning update along the steering direction is near-orthogonal to its pre-edit weight pattern (mean $\cos\theta = 0.074$). When steered behaviour degrades, fine-tuning does not achieve it by dismantling or reversing the steering mechanism itself. Embedded steering is therefore mechanistically durable but functionally vulnerable, and requires behavioural re-validation after downstream training.

cs.CL↗

arXiv2026

Efficient Grammar-Constrained Decoding via Parser Stack Classification

LLMs are widely used to generate structured output like source code or JSON. Grammar-constrained decoding (GCD) can guarantee the syntactic validity of the generated output, by masking out tokens that violate rules specified by a context-free grammar. However, the online computational overhead of existing GCD methods, with latency typically scaling linearly with vocabulary size, limits the throughput of LLMs, especially for models with large vocabularies. To address this issue, we propose PSC, a novel grammar-constrained decoding method. By combining acceptance conditions of all vocabulary tokens into a single classifier of the parser stack during preprocessing, PSC can compute the complete vocabulary mask by checking the parser stack exactly once per decoding step, with time complexity independent of the vocabulary size. Experiments show that PSC computes masks up to 700$\times$ faster than baselines on complex programming language grammars, and up to 30$\times$ faster for schema-conformant JSON; end-to-end LLM throughput with PSC approaches that of unconstrained decoding. We analyze the preprocessing overhead for preprocessing providers and decoding users, and provide a break-even point analysis to help users decide whether to do preprocessing by themselves.

cs.SE↗

arXiv2026

ClarifyCodeBench: Evaluating LLMs on Clarifying Ambiguous Requirements for Code Generation

Large Language Models have emerged as programming assistants. However, the efficacy of code generation is constrained by the quality of input requirements, which are frequently ambiguous, incomplete, or underspecified. While LLMs excel at one-shot code synthesis, their ability to proactively clarify intent remains underexplored, as a critical trait for robust software engineering. Existing benchmarks largely overlook this interactive bottleneck, assuming perfectly specified prompts that do not reflect the iterative nature of requirement elicitation. To bridge this gap, we introduce ClarifyCodeBench, a novel interactive benchmark for evaluating LLMs' capability in resolving requirement ambiguity. Constructed from real-world programming tasks, ClarifyCodeBench features high-quality manual annotations, including N unique ambiguity types, associated clarification questions, and corresponding ground-truth answers. Furthermore, we formalize two rigorous metrics to assess the interaction quality: Turn-discounted Key Question Rate, which penalizes inefficient questioning, and Optimal Round Adherence, which measures the precision of the elicitation process. We conduct a systematic evaluation of six state-of-the-art LLMs using ClarifyCodeBench. Our empirical results yield three critical insights: 1) Capability Decoupling: Strong code generation performance does not inherently translate to effective requirement clarification; 2) The Reasoning Paradox: While increased computational thinking enhances code correctness, it yields marginal gains in identifying ambiguities; 3) The Multi-ambiguity Ceiling: LLMs' clarification performance degrades sharply as the density of ambiguities increases, revealing a significant bottleneck in handling complex, real-world specifications. Our work underscores the necessity for future AI4SE research to transition from static synthesis to interactive elicitation.

cs.SE↗

arXiv2026

Optomechanical system with tunable dissipative and dispersive couplings

We demonstrate an optomechanical system with tunable dissipative and dispersive couplings using a Fabry-Perot cavity and a string mechanical resonator. By varying the diameter and material of the mechanical resonator, and the relative location between the mechanical resonator and the cavity, the relative strengths of dissipative and dispersive coupling could be tuned continuously from dissipation-dominated regime to dispersion-dominated regime. In our experiments, the dissipative-to-dispersive coupling ratios of 1.3 and 0.6 are achieved by using two different mechanical resonators, corresponding to a transition from dissipation-dominated to dispersion-dominated optomechanical system. Theoretically, the coupling ratio could be tuned from 25 to 0.02 by optimizing the mechanical resonator, spanning over three orders of magnitude. These two distinct coupling regimes are achieved with the same experimental platform. The capability to freely adjust the coupling ratio provides a versatile platform for exploring quantum effects of massive mechanical resonators and quantum-limited measurements.

quant-ph↗

arXiv2026

Saber: An Efficient Sampling with Adaptive Acceleration and Backtracking Enhanced Remasking for Diffusion Language Model

Diffusion language models (DLMs) are emerging as a compelling alternative to the dominant autoregressive paradigm, offering inherent advantages in parallel generation and bidirectional context modeling. However, for the tasks with strict structural constraints such as code generation, DLMs face a critical trade-off between inference speed and output quality, where accelerating generation by reducing sampling steps often leads to catastrophic performance collapse. We find that the fundamental reasons are: 1) the generation difficulty is non-uniform in the structured sequence decoding steps, making DLM's static acceleration strategy suboptimal; 2) the context of tokens generated by DLM evolves continuously, causing early high-confidence predictions to turn into irreversible errors. In this paper, we introduce efficient Sampling with Adaptive acceleration and Backtracking Enhanced Remasking (i.e., Saber), a novel training-free sampling algorithm for DLMs that first achieves both better inference speed and output quality in code generation. Saber dynamically adjusts the number of tokens unmasked per step based on the model's evolving confidence, and utilizes a backtracking mechanism to revert tokens whose confidence drops as new context emerges, with its effectiveness supported by theoretical analysis. Extensive experiments on multiple mainstream code generation benchmarks show that Saber boosts Pass@1 accuracy by an average of 1.9\% over mainstream DLM sampling methods, while achieving an average 251.4\% inference speedup. By leveraging the inherent advantages of DLMs, our work significantly narrows the performance gap with autoregressive models in code generation.

cs.AI↗

arXiv2026

Semi-Device-Independent Quantum Random Number Generator Resistant to General Attacks

Quantum random number generators (QRNGs) produce true random numbers based on the inherent randomness of quantum theory, rendering them a foundational segment of quantum cryptography. Distinguished from trusted-device QRNGs whose security depends on characterized devices, semi-device-independent (semi-DI) QRNGs permit partial devices to be defective or even maliciously manipulated, which achieves a good trade-off between generation rate and security. In this paper, we propose a semi-DI QRNG that resists general attacks while accounting for finite-size effects. The protocol requires no rigorous characterization of the source and measurement devices other than limiting the energy of the emitted states, significantly reducing the demands on practical QRNG systems. Leveraging the tight Kato inequality for correlated variables, we show that our protocol generates more randomness than it consumes. Furthermore, we demonstrate the scheme on a continuous-variable system with ternary inputs of states. Heterodyne detection is employed to enable phase compensation through data postprocessing, alleviating the stringent requirement on system stability. The system operates at 100 MHz, achieving a net random number generation rate of 1.165 Mbps at 5.3x10^9 rounds. Our work offers a promising approach to achieve both the robust security and high generation rate with a simple experimental setup.

quant-ph↗

arXiv2026

Lookahead-then-Verify: Reliable Constrained Decoding for Diffusion LLMs under Context-Free Grammars

Diffusion Large Language Models (dLLMs) have demonstrated promising generative capabilities and are increasingly used to produce formal languages defined by context-free grammars, such as source code and chemical expressions. However, as probabilistic models, they still struggle to generate syntactically valid outputs reliably. A natural and promising direction to address this issue is to adapt constrained decoding techniques to enforce grammatical correctness during generation. However, applying these techniques faces two primary obstacles. On the one hand, the non-autoregressive nature of dLLMs renders most existing constrained decoding approaches inapplicable. On the other hand, current approaches specifically designed for dLLMs may allow intermediate outputs that are impossible to complete into valid sentences, which significantly limits their reliability in practice. To address these challenges, we present LAVE, a constrained decoding approach specifically designed for dLLMs. Our approach leverages a key property of dLLMs, namely their ability to predict token distributions for all positions in parallel during each forward pass. Whenever a new token is proposed by model, LAVE performs lookahead using these distributions to efficiently and reliably verify the validity of the proposed token. This design ensures reliable constraints by reliably preserving the potential for intermediate outputs to be extended into valid sentences. Extensive experiments across four widely used dLLMs and three representative benchmarks demonstrate that LAVE consistently outperforms existing baselines and achieves substantial improvements in syntactic correctness, while incurring negligible runtime overhead.

cs.CL↗

arXiv2025

AdapTrack: Constrained Decoding without Distorting LLM's Output Intent

Language model-based code generation and completion tools have been widely adopted, but they may sometimes produce code that does not meet necessary constraints, such as syntactic correctness or API existence. Constrained decoding techniques are developed to help the model generate code adhering to the constraints by greedily eliminating generation options that violate constraints at each step of the generation process. However, there is a severe limitation of constrained decoding, that it distorts the model's output intent, forcing it to produce code that may satisfy the constraint but does not match the development intent and is therefore incorrect. In response to this challenge, we propose AdapTrack. By incorporating backtracking into the generation process, AdapTrack avoids distorting the output intent of the model, thereby producing results that are not only constraint-compliant but also more semantically aligned with model's output intent. On our synthetic API completion dataset, AdapTrack can achieve up to 360.87% improvement compared to constrained decoding; on the real-world API completion dataset we collect that exhibits similar issues, AdapTrack can achieve up to 38.93% improvement over constrained decoding; in general code genration benchmarks, compared to constrained decoding, AdapTrack can achieve up to 7.84% improvement on HumanEval, and up to 6.42% improvement on MBPP. This indicates that, simply by better adhering to the model's output intent, AdapTrack can achieve significant improvements. We provide a theoretical proof that the distribution produced by AdapTrack aligns with the model's distribution given the generated tokens, thereby ensuring that the model's output intent is not distorted. Experiments on DSL problems show that, compared to existing methods, our approach can provide generation results that are more consistent with the language model's distribution.

cs.SE↗

arXiv2025

A Fully Transformer Based Multimodal Framework for Explainable Cancer Image Segmentation Using Radiology Reports

We introduce Med-CTX, a fully transformer based multimodal framework for explainable breast cancer ultrasound segmentation. We integrate clinical radiology reports to boost both performance and interpretability. Med-CTX achieves exact lesion delineation by using a dual-branch visual encoder that combines ViT and Swin transformers, as well as uncertainty aware fusion. Clinical language structured with BI-RADS semantics is encoded by BioClinicalBERT and combined with visual features utilising cross-modal attention, allowing the model to provide clinically grounded, model generated explanations. Our methodology generates segmentation masks, uncertainty maps, and diagnostic rationales all at once, increasing confidence and transparency in computer assisted diagnosis. On the BUS-BRA dataset, Med-CTX achieves a Dice score of 99% and an IoU of 95%, beating existing baselines U-Net, ViT, and Swin. Clinical text plays a key role in segmentation accuracy and explanation quality, as evidenced by ablation studies that show a -5.4% decline in Dice score and -31% in CIDEr. Med-CTX achieves good multimodal alignment (CLIP score: 85%) and increased confi dence calibration (ECE: 3.2%), setting a new bar for trustworthy, multimodal medical architecture.

cs.CV↗

arXiv2025

MMIS-Net for Retinal Fluid Segmentation and Detection

Purpose: Deep learning methods have shown promising results in the segmentation, and detection of diseases in medical images. However, most methods are trained and tested on data from a single source, modality, organ, or disease type, overlooking the combined potential of other available annotated data. Numerous small annotated medical image datasets from various modalities, organs, and diseases are publicly available. In this work, we aim to leverage the synergistic potential of these datasets to improve performance on unseen data. Approach: To this end, we propose a novel algorithm called MMIS-Net (MultiModal Medical Image Segmentation Network), which features Similarity Fusion blocks that utilize supervision and pixel-wise similarity knowledge selection for feature map fusion. Additionally, to address inconsistent class definitions and label contradictions, we created a one-hot label space to handle classes absent in one dataset but annotated in another. MMIS-Net was trained on 10 datasets encompassing 19 organs across 2 modalities to build a single model. Results: The algorithm was evaluated on the RETOUCH grand challenge hidden test set, outperforming large foundation models for medical image segmentation and other state-of-the-art algorithms. We achieved the best mean Dice score of 0.83 and an absolute volume difference of 0.035 for the fluids segmentation task, as well as a perfect Area Under the Curve of 1 for the fluid detection task. Conclusion: The quantitative results highlight the effectiveness of our proposed model due to the incorporation of Similarity Fusion blocks into the network's backbone for supervision and similarity knowledge selection, and the use of a one-hot label space to address label class inconsistencies and contradictions.

eess.IV↗

arXiv2025

Continuous-variable quantum key distribution over 50.4 km fiber using integrated silicon photonic transmitter and receiver

Quantum key distribution (QKD) is the fastest-growing and relatively mature technology in the field of quantum information, enabling information-theoretically secure key distribution between two remote users. Although QKD based on off-the-shelf telecom components has been validated in both laboratory and field tests, its high cost and large volume remain major obstacles to large-scale deployment. Photonic integration, featured by its compact size and low cost, offers an effective approach to addressing the above challenges faced by QKD. Here, we implement a high-performance, integrated local local oscillator continuous-variable (CV) QKD system based on an integrated silicon photonic transmitter and receiver. By employing a high-speed silicon photonic integrated in-phase and quadrature modulator, a low-noise and high bandwidth silicon photonic integrated heterodyne detector, and digital signal processing, our CV-QKD system achieves a symbol rate of up to 1.5625 GBaud. Furthermore, the system achieves asymptotic secret key rates of 31.05 and 5.05 Mbps over 25.8 and 50.4 km standard single-mode fiber, respectively, using an 8-phase-shift keying discrete modulation. Our integrated CV-QKD system with high symbol rate and long transmission distance pays the way for the quantum secure communication network at metropolitan area.

quant-ph↗

arXiv2025

High-Performance Fully Passive Discrete-State Continuous-Variable Quantum Key Distribution With Local Local Oscillator

We propose and demonstrate a fully passive discrete-state continuous-variable quantum key distribution (CV-QKD), which can eliminate all modulator side channels on the source side, using a local local oscillator (LLO). The CV-QKD system achieves a maximum transmission length of 100 km with a repetition rate of 1 GHz using specially designed phase rotation and discretization methods, and the corresponding secret key bit rate is 127 kbps, as estimated based on the amplitude of prepared states at the transmitter, as well as the first- and second-order moments of quadratures at the receiver by employing the convex optimization without imposing any assumptions on the quantum channel. The performance of the protocol is similar to that of modulated CV LLO protocols and better than those of passive discrete-variable and CV protocols. Our protocol is expected to play an important role in the quantum metropolitan area networks and quantum access networks with high realistic security.

quant-ph↗

arXiv2025

Highly integrated broadband entropy source for quantum random number generators based on vacuum fluctuations

In this work, we designed and experimentally verified a highly integrated broadband entropy source for a quantum random number generator (QRNG) based on vacuum fluctuations. The core of the entropy source is a hybrid laser-and-silicon-photonics chip, which is only 6.3 $ \times $ 2.6 $ \times $ 1.5 mm$^{3}$ in size. A balanced homodyne detector based on cascaded radio-frequency amplifiers in the entropy source achieves a 3 dB bandwidth of 2.4 GHz and a common-mode rejection ratio above 25 dB. The quantum-to-classical-noise ratio is 9.51 dB at a photoelectron current of 1 mA. The noise equivalent power and equivalent transimpedance are 8.85$\,\text{pW}/\sqrt{\text{Hz}}$ , and 22.8 k$Ω$, respectively. After optimization using equalizer technology that eliminates the dependence of adjacent samples, the quantum random number generation rate reaches 67.9 Gbps under average conditional minimum entropy and 61.9 Gbps under the worst-case conditional minimum entropy. The developed hybrid chip enhances the integrability and speed of QRNG entropy sources based on vacuum fluctuations.

quant-ph↗

arXiv2025

Automated UX Insights from User Research Videos by Integrating Facial Emotion and Text Sentiment

Emotion recognition technology has been studied from the past decade. With its growing importance and applications such as customer service, medical, education, etc., this research study aims to explore its potential and importance in the field of User experience evaluation. Recognizing and keeping track of user emotions in user research video is important to understand user needs and expectations from a service/product. Little research has been done that focuses on automating emotion extraction from a video where more than one modality has been incorporated in the field of UX. The study aims at implementing different modalities such as facial emotion recognition, speech-to-text and text-based emotion recognition for capturing emotional nuances from a user research video and extract meaningful actionable insights. For selection of facial emotion recognition model, 10 pre-trained models were evaluated on three benchmark datasets i.e. FER-2013, AffectNet and CK+, selecting the model with most generalization ability. To extract speech and convert to text, OpenAI's Whisper model was implemented and finally the emotions from text were recognized using a pre-trained model available at HuggingFace website having an evaluation accuracy more than 95%. The study also integrates the gathered data using temporal alignment and fusion for deeper and contextual insights. The study further demonstrates a way of automating data analysis through PandasAI Python library where OpenAI's GPT-4o model was implemented along with a discussion on other possible solutions. This study is an attempt to demonstrate a proof of concept where automated meaningful insights are extracted from a video based on user emotions.

cs.HC↗

arXiv2025

Advances in Continuous Variable Measurement-Device-Independent Quantum Key Distribution

Continuous variable quantum key distribution (CV-QKD), utilizes continuous variables encoding such as the quadra-ture components of the quantized electromagnetic field and coherent detection decoding, offering good compatibility with the existing telecommunications technology and components. Continuous variable measurement-device-independent QKD (CV-MDI-QKD) can eliminate all the security threats arising from the receiver effectively, the crucial security loophole of CV-QKD implementations. Recently, CV-MDI-QKD has attracted extensive attentions and witnessed rapid progress. Here, we review the achievements that have been made in the field of CV-MDI-QKD, including the basic principle, advancements in theoretical protocols and experimental demonstrations. Finally, we discuss the challenges faced in practical applications and future research directions.

quant-ph↗

arXiv2024

Phononically shielded multi-wavelength photonic-crystal membrane for cavity quantum optomechanics

We propose and design a stoichiometric silicon-nitride membrane resonator featuring highly reflective at multi-wavelengths and high mechanical quality factor. The membrane resonator has a thickness of 100 nm and 2D-photonic and phononic crystal patterns. By designing concentric holes of suitable radius on both sides of the membrane, high reflectivity at multi-wavelengths can be achieved. In particularly, the simulation shows that high reflectivity can be realized at telecommunications wavelength and alkaline atoms absorption lines, with reflectivity of 99.76% at 852 nm, and 99.98% at 1054 nm, and 99.96% at 1566 nm, respectively. The designed device can find useful applications in cavity optomechanical system to realize quantum frequency conversion and precise quantum measurement, and other field of quantum information processing tasks.

physics.optics↗

arXiv2024

Combining Genre Classification and Harmonic-Percussive Features with Diffusion Models for Music-Video Generation

This study presents a novel method for generating music visualisers using diffusion models, combining audio input with user-selected artwork. The process involves two main stages: image generation and video creation. First, music captioning and genre classification are performed, followed by the retrieval of artistic style descriptions. A diffusion model then generates images based on the user's input image and the derived artistic style descriptions. The video generation stage utilises the same diffusion model to interpolate frames, controlled by audio energy vectors derived from key musical features of harmonics and percussives. The method demonstrates promising results across various genres, and a new metric, Audio-Visual Synchrony (AVS), is introduced to quantitatively evaluate the synchronisation between visual and audio elements. Comparative analysis shows significantly higher AVS values for videos generated using the proposed method with audio energy vectors, compared to linear interpolation. This approach has potential applications in diverse fields, including independent music video creation, film production, live music events, and enhancing audio-visual experiences in public spaces.

cs.MM↗