arXiv subjects

Pengjun Lu

Publications and source records attributed to Pengjun Lu.

6 recordsLinked to original sources

arXiv2026

The 1.6 $μ\mathrm{m}$ $\rm{H^{-}}$ bump as a diagnostic of TP-AGB contributions and recent quenching. I. Insights from stellar population models

The 1.6 $μ\mathrm{m}$ $\rm{H^{-}}$ bump arises from a minimum in the opacity of the negative hydrogen ion in cool stellar atmospheres. Although it is recognized as a common feature in galaxy spectra, its potential for studying stellar populations and recent quenching remains little explored. Its enhancement at intermediate ages is strong in some stellar population models with substantial thermally pulsing asymptotic giant branch (TP-AGB) contributions but weak or absent in others. This variation makes the feature both a potential age diagnostic and a source of model uncertainty. We use models based on the X-shooter Spectral Library (XSL) and Flexible Stellar Population Synthesis (FSPS) to examine what drives this behavior and what it can reveal about quenching. Controlled FSPS experiments show that the $\rm{H^{-}}$ bump depends on both the amount of TP-AGB light and the stellar template used to represent it. The adopted XSL and TP-AGB-on FSPS models predict a transient enhancement following rapid quenching, with a different age response from optical spectral features. Combining these diagnostics can therefore provide additional information about recent quenching, although the interpretation depends on the population model and star-formation history. An illustrative comparison with existing JWST spectra shows that models including TP-AGB contributions can describe the observed $\rm{H^{-}}$ region reasonably well. These results motivate further use of the $\rm{H^{-}}$ bump to test TP-AGB prescriptions and investigate quenching with current and forthcoming observations.

astro-ph.GA↗

arXiv2026

Rethinking the Evaluation of Efficiency Methods for Multi-Agent Systems

Efficiency is increasingly important for Large Language Model (LLM)-based multi-agent systems (MAS), as larger models and more agents introduce substantial execution costs. Recent methods aim to make MAS cheaper by pruning agents, removing communication edges, or searching for compact structures. However, we argue that existing evaluations may overestimate their true ability to improve MAS efficiency. Reported gains are often measured under method-specific prompts and starting topologies, making them difficult to attribute to the proposed structural changes. Moreover, many reported successes appear in non-MAS-demanding settings, where a single agent or a randomly pruned system can already preserve strong performance. To study these issues, we introduce a controlled and MAS-demanding diagnostic benchmark for representative MAS efficiency methods. We evaluate methods under a shared backbone model, agent registry, and runtime, across controlled variations in topology, scale, depth, and tool use. Our analysis shows that many reported gains are setup-dependent and may arise from structural collapse, disabled tool pathways, or starting systems where random pruning already preserves accuracy, rather than robust improvements in MAS efficiency.

cs.LG↗

arXiv2025

Assessing LLMs for Serendipity Discovery in Knowledge Graphs: A Case for Drug Repurposing

Large Language Models (LLMs) have greatly advanced knowledge graph question answering (KGQA), yet existing systems are typically optimized for returning highly relevant but predictable answers. A missing yet desired capacity is to exploit LLMs to suggest surprise and novel ("serendipitious") answers. In this paper, we formally define the serendipity-aware KGQA task and propose the SerenQA framework to evaluate LLMs' ability to uncover unexpected insights in scientific KGQA tasks. SerenQA includes a rigorous serendipity metric based on relevance, novelty, and surprise, along with an expert-annotated benchmark derived from the Clinical Knowledge Graph, focused on drug repurposing. Additionally, it features a structured evaluation pipeline encompassing three subtasks: knowledge retrieval, subgraph reasoning, and serendipity exploration. Our experiments reveal that while state-of-the-art LLMs perform well on retrieval, they still struggle to identify genuinely surprising and valuable discoveries, underscoring a significant room for future improvements. Our curated resources and extended version are released at: https://cwru-db-group.github.io/serenQA.

cs.CL↗

arXiv2025

Oxyster: A Circumgalactic Low-ionization Oxygen Nebula next to a Starburst Galaxy at $z\sim1$

Extended emission line nebulae around galaxies or active galactic nuclei (AGNs) provide a unique window to investigate the galactic ecosystem through the circumgalactic medium (CGM). Using Subaru Hyper-Suprime Cam narrow-band imaging and spectroscopic follow-up, we serendipitously discover "Oxyster" - a large ionized nebula next to an interacting starburst galaxy at $z=0.924$. The nebula is traced by extended [OII]3726,3729 ($\sim 30$ kpc) and [OIII]5007 ($\sim 20$ kpc) emission lines. On the nebula luminosity-size plane, Oxyster surpasses the extended narrow-line regions around low-$z$ AGNs, resembling a higher-$z$ analog of "Hanny's Voorwerp". However, its uniformly low [OIII]/[OII] ratio (O32) sets it apart from typical AGN light echoes. For the host galaxy, HST and JWST images reveal a disturbed red disk galaxy with a single blue spiral "arm". Spectral energy distribution (SED) fitting suggests the $2-6\times 10^{10} ~\rm M_{\odot}$ host galaxy sits above the star-forming main sequence with an ongoing starburst, especially in the "arm", and have $<5\%$ luminosity contribution from AGN, consistent with X-ray non-detection and radio continuum. Standard photoionization and shock models struggle to explain simultaneously Oxyster's emission line luminosities, low O32 ratio, and the non-detection of H$β$ line. A plausible explanation could involve the combination of a recent ($<10^8$ yrs) starburst and a low-luminosity AGN ($L_{\rm{bol}} \sim 1\times10^{42}$ erg/s). While Oxyster's nature awaits future investigation, its discovery highlights the potential of ground-based narrow-band imaging to uncover extended emission line nebulae around non-AGN systems, opening new avenues for studying the CGM of normal galaxies in the early Universe.

astro-ph.GA↗

arXiv2025

Discovery of a Pair of Galaxies with Both Hosting X-ray Binary Candidates at $z=2.544$

Among high-redshift galaxies, aside from active galactic nuclei (AGNs), X-ray binaries (XRBs) can be significant sources of X-ray emission. XRBs play a crucial role in galaxy evolution, reflecting the stellar populations of galaxies and regulating star formation through feedback, thereby shaping galaxy structure. In this study, we report a spectroscopically confirmed X-ray emitting galaxy pair (UDF3 and UDF3-2) at $z = 2.544$. By combining multi-wavelength observations from JWST/NIRSpec MSA spectra, JWST/NIRCam and MIRI imaging, Chandra, HST, VLT, ALMA, and VLA, we analyze the ionized emission lines, which are primarily driven by H II region-like processes. Additionally, we find that the mid-infrared radiation can be fully attributed to dust emission from galaxy themselves. Our results indicate that the X-ray emission from these two galaxies is dominated by high-mass XRBs, with luminosities of $L_X= (1.43\pm0.40) \times 10^{42} \, \text{erg} \, \text{s}^{-1}$ for UDF3, and $(0.40\pm0.12) \times 10^{42} \, \text{erg} \, \text{s}^{-1}$ for UDF3-2. Furthermore, we measure the star formation rate (SFR) of $529_{-88}^{+64}$ $M_\odot$ yr$^{-1}$ for UDF3, placing it $\approx$ 0.5 dex below the $L_X$/SFR-$z$ relation. This offset reflects the redshift-dependent enhancement of $L_X$/SFR-$z$ relation, which is influenced by metallicity and serves as a key observable for XRB evolution. In contrast, UDF3-2, with the SFR of $34_{-6}^{+6}$ $M_\odot$ yr$^{-1}$, aligns well with the $L_X$/SFR-$z$ relation. This galaxy pair represents the highest-redshift non-AGN-dominated galaxies with individual X-ray detections reported to date. This finding suggests that the contribution of XRBs to galaxy X-ray emission at high redshift may be underestimated.

astro-ph.GA↗

arXiv2017

Reinforcement Mechanism Design, with Applications to Dynamic Pricing in Sponsored Search Auctions

In this study, we apply reinforcement learning techniques and propose what we call reinforcement mechanism design to tackle the dynamic pricing problem in sponsored search auctions. In contrast to previous game-theoretical approaches that heavily rely on rationality and common knowledge among the bidders, we take a data-driven approach, and try to learn, over repeated interactions, the set of optimal reserve prices. We implement our approach within the current sponsored search framework of a major search engine: we first train a buyer behavior model, via a real bidding data set, that accurately predicts bids given information that bidders are aware of, including the game parameters disclosed by the search engine, as well as the bidders' KPI data from previous rounds. We then put forward a reinforcement/MDP (Markov Decision Process) based algorithm that optimizes reserve prices over time, in a GSP-like auction. Our simulations demonstrate that our framework outperforms static optimization strategies including the ones that are currently in use, as well as several other dynamic ones.

cs.GT↗