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

Yong Xie

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

ResearcherUnclaimed source record
Computation and LanguageArtificial IntelligenceMachine LearningComputer Visioncond-mat.mes-hallCryptography and Securityhep-exhep-phphysics.app-phq-fin.ST

Catalog footprint

What is connected

6works
10topics
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

6 published item(s)

preprint2026arXiv

Controlled Automatic Task-Specific Synthetic Data Generation for Hallucination Detection

We present a novel approach to automatically generate non-trivial task-specific synthetic datasets for hallucination detection. Our approach features a two-step generation-selection pipeline, using hallucination pattern guidance and a language style alignment during generation. Hallucination pattern guidance leverages the most important task-specific hallucination patterns while language style alignment aligns the style of the synthetic dataset with benchmark text. To obtain robust supervised detectors from synthetic datasets, we also adopt a data mixture strategy to improve performance robustness and generalization. Our results on three datasets show that our generated hallucination text is more closely aligned with non-hallucinated text versus baselines, to train hallucination detectors with better generalization. Our hallucination detectors trained on synthetic datasets outperform in-context-learning (ICL)-based detectors by a large margin of 32%. Our extensive experiments confirm the benefits of our approach with cross-task and cross-generator generalization. Our data-mixture-based training further improves the generalization and robustness of hallucination detection.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Efficient Continual Pre-training for Building Domain Specific Large Language Models

Large language models (LLMs) have demonstrated remarkable open-domain capabilities. LLMs tailored for a domain are typically trained entirely on domain corpus to excel at handling domain-specific tasks. In this work, we explore an alternative strategy of continual pre-training as a means to develop domain-specific LLMs over an existing open-domain LLM. We introduce FinPythia-6.9B, developed through domain-adaptive continual pre-training on the financial domain. Continual pre-trained FinPythia showcases consistent improvements on financial tasks over the original foundational model. We further explore simple but effective data selection strategies for continual pre-training. Our data selection strategies outperform vanilla continual pre-training's performance with just 10% of corpus size and cost, without any degradation on open-domain standard tasks. Our work proposes an alternative solution to building domain-specific LLMs cost-effectively.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

MAD-OPD: Breaking the Ceiling in On-Policy Distillation via Multi-Agent Debate

On-policy distillation (OPD) trains a student on its own trajectories under token-level teacher supervision, but existing methods are capped by a single-teacher capability ceiling: when the teacher errs, the student inherits the error. OPD also remains largely unexplored in agentic tasks, where per-step errors compound across long trajectories and destabilize training. We propose MAD-OPD (Multi-Agent Debate-driven On-Policy Distillation), which breaks this ceiling by recasting the distillation teacher as a deliberative collective of teachers that debate over the student's on-policy state; the debate produces an emergent collective intelligence that supplies token-level supervision, with each teacher's contribution weighted by its post-debate confidence. To extend OPD to agentic tasks, we also introduce On-Policy Agentic Distillation (OPAD), which adds step-level sampling to stabilize training under multi-step error compounding. We additionally derive a task-adaptive divergence principle, selecting JSD (Jensen-Shannon divergence) for agentic stability and reverse KL (Kullback-Leibler) divergence for code generation, and verify it both theoretically and empirically. Across six teacher-student configurations (Qwen3 and Qwen3.5; 1.7B-14B students, 8B-32B teachers) and five agentic and code benchmarks, MAD-OPD ranks first across all six configurations; on the 14B+8B$\to$4B setting it lifts the agentic average by $+2.4\%$ and the code average by $+3.7\%$ over the stronger single-teacher OPD.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A Word is Worth A Thousand Dollars: Adversarial Attack on Tweets Fools Stock Predictions

More and more investors and machine learning models rely on social media (e.g., Twitter and Reddit) to gather real-time information and sentiment to predict stock price movements. Although text-based models are known to be vulnerable to adversarial attacks, whether stock prediction models have similar vulnerability is underexplored. In this paper, we experiment with a variety of adversarial attack configurations to fool three stock prediction victim models. We address the task of adversarial generation by solving combinatorial optimization problems with semantics and budget constraints. Our results show that the proposed attack method can achieve consistent success rates and cause significant monetary loss in trading simulation by simply concatenating a perturbed but semantically similar tweet.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Exploring the $e^+e^-\to Λ_c^+Λ_c^-$ cross sections

A simultaneous fit is performed to the $e^+e^-\to Λ_c^+Λ_c^-$ cross section data measured by Belle and BESIII from threshold up to 5.4 GeV. In order to accommodate both the BESIII measurement near threshold and the Belle observation of a resonance $Y(4630)$, we build a composite PDF with a Breit-Wigner resonance and a continuum contribution to model the full cross section line shape of $e^+e^-\to Λ_c^+Λ_c^-$. The fit gives a mass of $M=[4636.1_{-7.2}^{+9.8} ($stat$)\pm 8.0($syst$)]$~MeV/$c^2$, a width of $Γ_{\rm tot}=[34.5_{-16.2}^{+21.0} ($stat$)\pm 5.6($syst$)]$~MeV, and $Γ_{e^+ e^-}\mathcal{B}[Y(4630)\toΛ_c^+Λ_c^-]=[18.3_{-6.1}^{+8.8} ($stat$)\pm 1.1($syst$)]$~eV/$c^2$ for the resonance. The width of $Y(4630)$ from our study is narrower than the previous Belle fit. The mass and width of $Y(4630)$ also show good agreement with a vector resonance $Y(4626)$ recently observed in $D_s^+D_{s1}(2536)^-$ by Belle.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Hexagonal Boron Nitride Phononic Crystal Waveguides

Hexagonal boron nitride (h-BN), one of the hallmark van der Waals (vdW) layered crystals with an ensemble of attractive physical properties, is playing increasingly important roles in exploring two-dimensional (2D) electronics, photonics, mechanics, and emerging quantum engineering. Here, we report on the demonstration of h-BN phononic crystal waveguides with designed pass and stop bands in the radio frequency (RF) range and controllable wave propagation and transmission, by harnessing arrays of coupled h-BN nanomechanical resonators with engineerable coupling strength. Experimental measurements validate that these phononic crystal waveguides confine and support 15 to 24 megahertz (MHz) wave propagation over 1.2 millimeters. Analogous to solid-state atomic crystal lattices, phononic bandgaps and dispersive behaviors have been observed and systematically investigated in the h-BN phononic waveguides. Guiding and manipulating acoustic waves on such additively integratable h-BN platform may facilitate multiphysical coupling and information transduction, and open up new opportunities for coherent on-chip signal processing and communication via emerging h-BN photonic and phononic devices.

0 citations0 reviewsNo code linkedOpen access