BZPEERPapers, people, evidence and research paths
Menu

Discover

SearchFind papers, people, topics, institutions and opportunities from one place.GraphSee how papers, authors, topics and evidence connect.

Research tools

HomeContinue from the papers, people and topics that matter now.CollectionsTurn saved papers into reading lists, evidence maps and shareable context.ResearchReconstruct the path that led you from search to paper to question.CollaborationMove from a paper, topic or expert request into a focused thread.PublishCreate full-text research articles with metadata, formulas and versions.

Network

ExpertsFind specialists whose work and reviews explain why they can help.CommunitiesJoin research groups organized around papers, topics and shared evidence.

Opportunities

ListingsBrowse roles, calls and collaborations connected to real research context.

Account

Log inReturn to your saved papers, graph views and active threads.Create accountStart saving papers, building a research trail and joining teams.
Log inCreate account

Source author record

Juntao Li

Juntao Li appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record
physics.opticsArtificial IntelligenceComputation and Languagequant-phnlin.PSphysics.app-phComputer Visioncond-mat.otherInformation Retrieval

Catalog footprint

What is connected

14works
9topics
4close collaborators

Actions

Connect this record

Log in to claim
Open graphBrowse works

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

14 published item(s)

preprint2026arXiv

EviRerank: Adaptive Evidence Construction for Long-Document LLM Reranking

Decoder-only LLM rerankers struggle with long documents: inference is costly and relevance signals can be diluted by irrelevant context. Motivated by an attention analysis indicating a consistent degradation trend when non-relevant text is appended, we propose EviRerank, an evidence-based long-document reranking framework for decoder-only LLMs. EviRerank (i) scores document blocks with a lightweight selector (BM25, bi-encoder, or cross-encoder), (ii) constructs a compact reranking context under a hard token cap by dynamically budgeting evidence blocks with Adaptive Evidence Budgeting (AEB) and adding a global summary cue via Summary Augmentation (SA), and (iii) reranks with a decoder-only LLM. Across TREC DL'19, DL'23, and MLDR-zh, EviRerank consistently outperforms full-document LLM reranking and strong block-selection baselines while substantially reducing the required input length. On TREC DL'19, EviRerank achieves 0.743 nDCG@10 and 0.307 MAP, establishing a new best result and improving over RankLLaMA (0.701/0.288) by +0.042 nDCG@10 (+6.0%) and +0.019 MAP (+6.6%).

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

MemReread: Enhancing Agentic Long-Context Reasoning via Memory-Guided Rereading

To tackle long-context reasoning tasks without the quadratic complexity of standard attention mechanisms, approaches based on agent memory have emerged, which typically maintain a dynamically updated memory when linearly processing document chunks. To mitigate the potential loss of latent evidence in this memorize-while-reading paradigm, recent works have integrated retrieval modules that allow agents to recall information previously discarded during memory overwriting. However, retrieval-based recall suffers from both evidence loss during memory formation and interference induced by invalid queries. To overcome these limitations, we propose MemReread. Built upon streaming reading, MemReread circumvents intermediate retrieval. It triggers question decomposition and rereading when the final memory is insufficient, enabling the recovery of indirect facts that were prematurely discarded. This design supports non-linear reasoning while preserving the inherent logical flow of document comprehension. To further enhance practicality, we introduce a reinforcement learning framework that enhances length extrapolation capability while dynamically determining the number of rereading passes based on task complexity, thereby flexibly controlling computational overhead. Extensive experiments demonstrate that MemReread consistently outperforms baseline frameworks on long-context reasoning tasks, while maintaining linear time complexity with respect to context length.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

OmniSelect: Dynamic Modality-Aware Token Compression for Efficient Omni-modal Large Language Models

Omnimodal large language models (OmniLLMs) have recently gained increasing attention for unified audio-video understanding. However, processing long multimodal token sequences introduces substantial computational overhead, making efficient token compression crucial. Existing methods typically rely on fixed, modality-specific guidance, which fails to account for the varying importance of modalities across different queries. To address this limitation, we propose $\textbf{OmniSelect}$, a training-free, modality-adaptive token pruning framework that dynamically selects appropriate compression strategies for multimodal inputs. Specifically, we leverage a lightweight AudioCLIP model to estimate cross-modal relevance and categorize each input into three pruning regimes: Audio-Centric, Video-Centric, and Uniform pruning. Based on these relevance scores, OmniSelect further performs fine-grained token pruning within each temporal group, adaptively allocating pruning ratios to preserve informative tokens across modalities. By explicitly modeling modality preference and enabling dynamic strategy selection, OmniSelect effectively avoids the pitfalls of one-size-fits-all compression. Extensive experiments demonstrate that our method achieves efficient multimodal token reduction while maintaining strong performance, without requiring any additional training.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

When to Vote, When to Rewrite: Disagreement-Guided Strategy Routing for Test-Time Scaling

Large Reasoning Models (LRMs) achieve strong performance on mathematical reasoning tasks but remain unreliable on challenging instances. Existing test-time scaling methods, such as repeated sampling, self-correction, and tree search, improve performance at the cost of increased computation, yet often exhibit diminishing returns on hard problems. We observe that output disagreement is strongly correlated with instance difficulty and prediction correctness, providing a useful signal for guiding instance-level strategy selection at test time. Based on this insight, we propose a training-free framework that formulates test-time scaling as an instance-level routing problem, rather than allocating more computation within a single strategy, dynamically selecting among different scaling strategies based on output disagreement. The framework applies lightweight resolution for consistent cases, majority voting for moderate disagreement, and rewriting-based reformulation for highly ambiguous instances. Experiments on seven mathematical benchmarks and three models show that our method improves accuracy by 3% - 7% while reducing sampling cost compared to existing approaches.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Chinese grammatical error correction based on knowledge distillation

In view of the poor robustness of existing Chinese grammatical error correction models on attack test sets and large model parameters, this paper uses the method of knowledge distillation to compress model parameters and improve the anti-attack ability of the model. In terms of data, the attack test set is constructed by integrating the disturbance into the standard evaluation data set, and the model robustness is evaluated by the attack test set. The experimental results show that the distilled small model can ensure the performance and improve the training speed under the condition of reducing the number of model parameters, and achieve the optimal effect on the attack test set, and the robustness is significantly improved. Code is available at https://github.com/Richard88888/KD-CGEC.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Correction of aberrations via polarization in single layer metalenses

The correction of multiple aberrations in an optical system requires different optical elements, which increases its cost and complexity. Metasurfaces hold great promise to providing new functionality for miniaturized and low-cost optical systems. A key advantage over their bulk counterparts is the metasurface's ability to respond to the polarization of light, which adds a new degree of freedom to the optical design. Here, we show that polarization control enables a form-birefringent metalens to correct for both spherical and off-axis aberrations using a single element only, which is not possible with bulk optics. The metalens encodes two phase profiles onto the same surface, thus allowing switching from high resolution to wide field of view operation. Such ability to obtain both high resolution and wide field of view in a single layer is an important step towards integration of miniaturized optical systems, which may find many applications, e.g., in microscopy and endoscopy.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Metalenses with polarization-independent adaptive nano-antennas

Metalens research has made major advances in recent years. These advances rely on the simple design principle of arranging meta-atoms in regular arrays to create an arbitrary phase and polarization profile. Unfortunately, the concept of equally spaced meta-atoms reaches its limit for high deflection angles where the deflection efficiency decreases. The efficiency can be increased using nano-antennas with multiple elements, but their polarization sensitivity hinders their application in metalenses. Here, we show that by designing polarization-insensitive dimer nano-antennas and abandoning the principle of equally spaced unit cells, polarization-independent ultrahigh numerical aperture (NA=1.48) oil-immersion operation with an efficiency of 43% can be demonstrated. This represents a significant improvement on other polarization-independent designs at visible wavelength. We also use this single layer metalens to replace a conventional objective lens and demonstrate the confocal scanning microscopic imaging of a grating with a period of 300 nm at 532 nm operating wavelength. Overall, our results experimentally demonstrate a novel design concept that further improves metalens performance.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Cross-Lingual Low-Resource Set-to-Description Retrieval for Global E-Commerce

With the prosperous of cross-border e-commerce, there is an urgent demand for designing intelligent approaches for assisting e-commerce sellers to offer local products for consumers from all over the world. In this paper, we explore a new task of cross-lingual information retrieval, i.e., cross-lingual set-to-description retrieval in cross-border e-commerce, which involves matching product attribute sets in the source language with persuasive product descriptions in the target language. We manually collect a new and high-quality paired dataset, where each pair contains an unordered product attribute set in the source language and an informative product description in the target language. As the dataset construction process is both time-consuming and costly, the new dataset only comprises of 13.5k pairs, which is a low-resource setting and can be viewed as a challenging testbed for model development and evaluation in cross-border e-commerce. To tackle this cross-lingual set-to-description retrieval task, we propose a novel cross-lingual matching network (CLMN) with the enhancement of context-dependent cross-lingual mapping upon the pre-trained monolingual BERT representations. Experimental results indicate that our proposed CLMN yields impressive results on the challenging task and the context-dependent cross-lingual mapping on BERT yields noticeable improvement over the pre-trained multi-lingual BERT model.

0 citations0 reviewsNo code linkedOpen access
preprint2014arXiv

Control of the stability and soliton formation of dipole moments in a nonlinear plasmonic finite nanoparticle array

We perform numerical analysis of a finite nanoparticle array, in which the transversal dipolar polarizations are excited by a homogenous optical field. Considering the linearly long-range dipole-dipole interaction and the cubic dipole nonlinearity of particle, the characteristics of stability of a finite number nanoparticle array should be revised, compared with that of an infinite number nanoparticle array. A critical point in the low branch of the bistable curve is found, beyond which the low branch becomes unstable for a finite number of nanoparticles. The influence of the external field intensities and detuning frequencies on this critical point are investigated in detail. When the total number of particles approaches infinity, our results become similar to that of an infinity number particle system [cf. Ref.32]. Notably, with appropriate external optical field, a dark dipole soliton is formed. Moreover, when the scaled detuning is set to an appropriate value, a double monopole dark soliton (DMDS) consisting of two particles is formed. The DMDS may have potential applications in the subwavelength highly precise detection because of its very small width.

0 citations0 reviewsNo code linkedOpen access
preprint2014arXiv

Ultra-compact optical auto-correlator based on slow-light enhanced third harmonic generation in a silicon photonic crystal waveguide

The ability to use coherent light for material science and applications is directly linked to our ability to measure short optical pulses. While free-space optical methods are well-established, achieving this on a chip would offer the greatest benefit in footprint, performance, flexibility and cost, and allow the integration with complementary signal processing devices. A key goal is to achieve operation at sub-Watt peak power levels and on sub-picosecond timescales. Previous integrated demonstrations require either a temporally synchronized reference pulse, an off-chip spectrometer, or long tunable delay lines. We report the first device capable of achieving single-shot time-domain measurements of near-infrared picosecond pulses based on an ultra-compact integrated CMOS compatible device, with the potential to be fully integrated without any external instrumentation. It relies on optical third-harmonic generation in a slow-light silicon waveguide. Our method can also serve as a powerful in-situ diagnostic tool to directly map, at visible wavelengths, the propagation dynamics of near-infrared pulses in photonic crystals.

0 citations0 reviewsNo code linkedOpen access
preprint2013arXiv

Buffering and Trapping Ultrashort Optical Pulses in Concatenated Bragg Gratings

Strong retardation of ultrashort optical pulses, including their deceleration and stoppage in the form of Bragg solitons in a cascaded Bragg-grating (BG) structure, is proposed. The manipulations of the pulses are carried out, using nonlinear effects, in a chirped BG segment which is linked, via a defect, to a uniform grating. The storage of the ultrashort pulses is shown to be very robust with respect to variations of the input field intensity, suggesting the feasibility of storing ultrafast optical pulses in such a structure. Physical estimates are produced for the BGs written in silicon.

0 citations0 reviewsNo code linkedOpen access
preprint2011arXiv

Characteristics of Correlated Photon Pairs Generated in Ultra-compact Silicon Slow-light Photonic Crystal Waveguides

We report the characterization of correlated photon pairs generated in dispersion-engineered silicon slow-light photonic crystal waveguides pumped by picosecond pulses. We found that taking advantage of the 15 nm flat-band slow-light window (vg ~ c/30) the bandwidth for correlated photon-pair generation in 96 and 196 \mum long waveguides was at least 11.2 nm; while a 396 \mum long waveguide reduced the bandwidth to 8 nm (only half of the slow-light bandwidth due to the increased impact of phase matching in a longer waveguide). The key metrics for a photon-pair source: coincidence to accidental ratio (CAR) and pair brightness were measured to be a maximum 33 at a pair generation rate of 0.004 pair per pulse in a 196 \mum long waveguide. Within the measurement errors the maximum CAR achieved in 96, 196 and 396 \mum long waveguides is constant. The noise analysis shows that detector dark counts, leaked pump light, linear and nonlinear losses, multiple pair generation and detector jitter are the limiting factors to the CAR performance of the sources.

0 citations0 reviewsNo code linkedOpen access
preprint2011arXiv

Slow-light Enhanced Correlated Photon-Pair Generation in Silicon

We report the generation of correlated photon pairs in the telecom C-band, at room temperature, from a dispersion-engineered silicon photonic crystal waveguide. The spontaneous four-wave mixing process producing the photon pairs is enhanced by slow-light propagation enabling an active device length of less than 100 μm. With a coincidence to accidental ratio of 12.8, at a pair generation rate of 0.006 per pulse, this ultra-compact photon pair source is immediately applicable towards scalable quantum information processing realized on-chip.

0 citations0 reviewsNo code linkedOpen access
preprint2007arXiv

Manipulation of the quantum state by Majorana transition in spinor Bose-Einstein condensates

Manipulation of the quantum state by the Majorana transition in spinor BEC system has been realized by altering the rotation frequency of the magnetic field's direction. This kind of manipulation method has no limitation on the transition speed in principle and the system is well closed, which provides a new and superior tool to manipulate quantum states. Using this methord on pulsed atom laser, multicomponent spinor atom laser is generated. We demonstrate that the experiment results are agreed with the theoretical predication.

0 citations0 reviewsNo code linkedOpen access