arXiv subjects

Huan Yao

Publications and source records attributed to Huan Yao.

3 recordsLinked to original sources

arXiv2026

Evaluating Deep-Search Agents under Hierarchical Web Evidence Poisoning

Search-augmented LLM agents are increasingly used for consumer decisions, making them vulnerable to Generative Engine Optimization (GEO) poisoning. Existing benchmarks largely measure whether manipulated content is retrieved or endorsed, but do not track whether an agent verifies suspicious evidence, revises adopted claims, or recovers before producing its final recommendation. We introduce HAE-GEO, a benchmark that tracks the full trajectory from exposure to recovery under progressively more persuasive Web poisoning. Agents interact via a multi-turn Search-Scrape interface across three attack levels (L1 direct assertion, L2 contextual camouflage, and L3 apparent corroboration), supported by a controlled corpus of 72,039 clean pages and 770 poisoned pages per level spanning 8 product categories and 154 brands. Evaluation combines deterministic behavioral measures with six semantic rubric dimensions. Evaluating 10 agents, we find three recurring patterns: evidence recognition degrades under the corroboration trap; agentic search improves final resistance without improving evidence recognition or utility; and defense prompting increases verification, yet rarely converts verification into recovery.

cs.CR↗

arXiv2021

Estimate Metabolite Taxonomy and Structure with a Fragment-Centered Database and Fragment Network

Metabolite structure identification has become the major bottleneck of the mass spectrometry based metabolomics research. Till now, number of mass spectra databases and search algorithms have been developed to address this issue. However, two critical problems still exist: the low chemical component record coverage in databases and significant MS/MS spectra variations related to experiment equipment and parameter settings. In this work, we considered the molecule fragment as basic building blocks of the metabolic components which had relatively consistent signatures in MS/MS spectra. And from a bottom-up point of view, we built a fragment centered database, MSFragDB, by reorganizing the data from the Human Metabolome Database (HMDB) and developed an intensity-free searching algorithm to search and rank the most relative metabolite according to the users' input. We also proposed the concept of fragment network, a graph structure that encoded the relationship between the molecule fragments to find close motif that indicated a specific chemical structure. Although based on the same dataset as the HMDB, validation results implied that the MSFragDB had a higher hit ratio and furthermore, estimated possible taxonomy that a query spectrum belongs to when the corresponding chemical component was missing in the database. Aid by the Fragment Network, the MSFragDB was also proved to be able to estimate the right structure while the MS/MS spectrum suffers from the precursor-contamination. The strategy proposed is general and can be adopted in existing databases. We believe MSFragDB and Fragment Network can improve the performance of structure identification with existing data. The beta version of the database is freely available at www.xrzhanglab.com/msfragdb/.

q-bio.QM↗

arXiv2010

NaI (Tl) Calorimeter Calibration and Simulation for Coulomb Sum Rule Experiment in Hall-A at Jefferson Lab

A precision measurment of inclusive electron scattering cross sections was carried out at Jefferson Lab in the quasi-elastic region for ${^4}$He, ${^{12}}$C, ${^{56}}$Fe and ${^{208}}$Pb targets. Longitudinal ($R_{L}$) and transverse ($R_{T}$) response functions of nucleon were extracted in the momentum transfer range 0.55 GeV/c$\le$$|q|$$\le$1.0 GeV/c. To achieve the above goal, a NaI (Tl) calorimeter was used to distinguish good electrons from background including pions and low energy electrons rescattered from walls of the spectrometer magnets. Due to a large set of kinematics and changes in HV settings, a number of calibrations were performed for the NaI (Tl) detector. Corrections for a few blocks of NaI (Tl) with bad or no signal were applied. The resolution of NaI (Tl) detector after calibration reached $\frac{δE}{\sqrt{E}} \approx 3\%$ at E=1 GeV. The performance of NaI (Tl) detector was compared with a simulation.

nucl-ex↗