arXiv subjects

Ruijie He

Publications and source records attributed to Ruijie He.

8 recordsLinked to original sources

arXiv2026

RoboColosseum: Evaluating Physical Intelligence at Scale In-The-Wild

General-purpose robot policies must work across robots, tasks, and environments, yet the variation that makes physical trials informative also makes them difficult to reproduce. We introduce RoboColosseum, a distributed evaluation framework that combines unified policy serving and standardized protocols across a network of locally operated robots in diverse environments. Our initial study evaluates five vision-language-action (VLA) policies and three frontier AI agents across four embodiments (SO-101, Franka, Bimanual-YAM, and G1 Humanoid). The Open Track compares models without task-specific adaptation in blinded, pairwise trials under matched local conditions. The Fine-tune Track measures how well VLA policies adapt to 20 tasks across all four embodiments when each policy receives the same 100 demonstrations per task. Across 5,699 pairwise Open Track comparisons at 124 sites, frontier agents rival specialized VLA policies, while all systems struggle to complete open-world tasks reliably. After fine-tuning, GR00T N1.7 leads on SO-101 and pi0.5 leads on Franka and Bimanual-YAM. Policy rankings show weak agreement across embodiment-specific task suites (Kendall W = 0.24). Open Track rankings poorly predict relative performance after fine-tuning, with substantial rank reversals on all three embodiments shared between tracks. Measuring partial progress reveals differences that task success alone misses. On the G1 Humanoid, no policy completes any of the five tasks, yet average progress ranges from 13.3% to 23.2% across policies. Open Track rankings stabilize as trials accumulate, and active matchmaking achieves comparable ranking agreement with substantially fewer physical trials than uniform sampling. These results show how distributed physical evaluation can reveal where robot policies generalize, how well they adapt to new tasks, and how far they progress on tasks they fail to complete.

cs.RO↗

arXiv2026

EvoMaestro: Toward Interpretable and Steerable LLM-Driven Program Evolution

Recent program evolution systems use large language models (LLMs) to generate and iteratively improve populations of programs, producing striking advances in mathematics, algorithm design, and scientific computing. Yet these systems largely operate as fully automated black boxes. As populations grow, domain experts must make sense of the scores, code changes, reasoning, and algorithmic ideas across many programs, while current interfaces provide limited support for understanding or redirecting the evolution. We characterize this need to understand and steer populations of evolving algorithmic ideas as semantic oversight. A formative study with 8 domain experts yields six design requirements for this emerging human-computer interaction problem. We then propose a steerable program evolution framework that lets expert judgments shape subsequent evolution. Built on this framework, EvoMaestro is an interactive visual analytics system that organizes evolution information from population overview to source code. It helps experts locate and compare noteworthy programs, guide evolution with natural language, combine promising ideas, and prune unproductive directions. A seven-day system demonstration illustrates how an expert applied these capabilities throughout a long-running evolution process. A within-subjects study with 12 participants shows that EvoMaestro improves users' understanding of evolution processes, reduces cognitive workload, and supports expert steering. These findings suggest that preserving expert agency in open-ended LLM-driven search requires support for both informed judgment and the ability to shape subsequent automation.

cs.HC↗

arXiv2025

The fractions of reflection effect systems detected in different subclasses of hot subdwarfs

Hot subdwarfs with close low-mass M-type or brown dwarf companions usually show the reflection effect and often also eclipses. Through inspecting the light variations, we estimate the fractions of reflection effect systems detected in different subclasses of hot subdwarfs and constrain their possible formation channels. We noticed that none of the helium-rich hot subdwarfs were found with the reflection effect. Most of them might form through the merger channel. About 13\% of the He-poor hot subdwarfs on the extreme horizontal branch (EHB) with $T_{\rm eff}$ $\sim$ 25,000 -- 33,000 K are discovered with the reflection effect. Nevertheless, the cooler hot subdwarfs located on the top of the EHB, those grouped on the bottom of the EHB, and those located above the EHB with $T_{\rm eff}$ $\sim$ 45,000 -- 75,000 K exhibit lower fractions of reflection effect systems of about 2\%. These three subclasses of hot subdwarfs might contain higher fractions of single stars. Hot subdwarfs below the EHB also show a low fraction of reflection effect systems. They might have a higher fraction of stars with close WD companions. A similar fraction of reflection effect systems to those on the EHB with $T_{\rm eff}$ $\sim$ 25,000 -- 33,000 K is found in hot subdwarfs located above the EHB with $T_{\rm eff}$ $\sim$ 35,000 -- 45,000 K. We also discovered that hot subdwarfs close to the Galactic Plane might have a higher fraction of stars with the reflection effect than those at higher latitudes.

astro-ph.SR↗

arXiv2024

Radial velocity variability fractions of different types of hot subdwarf stars

Different types of hot subdwarfs may have different origins, which will cause them to present different radial velocity (RV) variability properties. Only 6$\pm$4% of our single-lined He-rich hot subdwarfs that only show spectroscopic features of hot subdwarfs are found to be RV variable, which is lower than the fraction of single-lined He-poor sdB stars (31$\pm$3%). Single-lined sdB stars with effective temperatures ($T_{\rm eff}$) $\sim$ 25,000 $-$ 33,000 K show an RV-variability fraction of 34$\pm$5%, while lower RV-variability fractions are observed for single-lined sdB stars cooler than about 25,000 K (11$\pm$4%), single-lined sdB/OB stars with $T_{\rm eff}$ $\sim$ 33,000 $-$ 40,000 K and surface gravities about 5.7 $-$ 6.0 (13$\pm$3%), as well as single-lined sdO/B stars with $T_{\rm eff}$ $\sim$ 45,000 $-$ 70,000 K (10$\pm$7%). Single-lined hot subdwarfs with $T_{\rm eff}$ $\sim$ 35,000 $-$ 45,000 K located above the extreme horizontal branch (EHB) show a similar RV-variability fraction of 34$\pm$9% as single-lined sdB stars at about 25,000 $-$ 33,000 K. The largest RV-variability fraction of 51$\pm$8% is found in single-lined hot subdwarfs below the canonical EHB. The detected RV-variability fraction of our composite hot subdwarfs with an infrared excess in their spectral energy distributions is 9$\pm$3%, which is lower than that fraction of single-lined hot subdwarfs. Since the average RV uncertainty we measured in the LAMOST spectra is about 7.0 km/s, the lower detected RV-variability fraction for composite hot subdwarfs is expected because the RV amplitudes associated with long-period systems are lower.

astro-ph.SR↗

arXiv2023

Mass distribution for single-lined hot subdwarf stars in LAMOST

Masses for 664 single-lined hot subdwarf stars identified in LAMOST were calculated by comparing synthetic fluxes from spectral energy distribution (SED) with observed fluxes from virtual observatory service. Three groups of hot subdwarf stars were selected from the whole sample according to their parallax precision to study the mass distributions. We found, that He-poor sdB/sdOB stars present a wide mass distribution from 0.1 to 1.0 $\mathrm{M}_{\odot}$ with a sharp mass peak around at 0.46 $\rm{M}_{\odot}$, which is consistent with canonical binary model prediction. He-rich sdB/sdOB/sdO stars present a much flatter mass distribution than He-poor sdB/sdOB stars and with a mass peak around 0.42 $\mathrm{M}_{\odot}$. By comparing the observed mass distributions to the predictions of different formation scenarios, we concluded that the binary merger channel, including two helium white dwarfs (He-WDs) and He-WD + main sequence (MS) merger, cannot be the only main formation channel for He-rich hot subdwarfs, and other formation channels such as the surviving companions from type Ia supernovae (SNe Ia) could also make impacts on producing this special population, especially for He-rich hot subdwarfs with masses less than 0.44 $\mathrm{M}_{\odot}$. He-poor sdO stars also present a flatter mass distribution with an inconspicuous peak mass at 0.18 $\mathrm{M}_{\odot}$. The similar mass - $ΔRV_\mathrm{max}$ distribution between He-poor sdB/sdOB and sdO stars supports the scenario that He-poor sdO stars could be the subsequent evolution stage of He-poor sdB/sdOB stars.

astro-ph.SR↗

arXiv2023

HOFA: Twitter Bot Detection with Homophily-Oriented Augmentation and Frequency Adaptive Attention

Twitter bot detection has become an increasingly important and challenging task to combat online misinformation, facilitate social content moderation, and safeguard the integrity of social platforms. Though existing graph-based Twitter bot detection methods achieved state-of-the-art performance, they are all based on the homophily assumption, which assumes users with the same label are more likely to be connected, making it easy for Twitter bots to disguise themselves by following a large number of genuine users. To address this issue, we proposed HOFA, a novel graph-based Twitter bot detection framework that combats the heterophilous disguise challenge with a homophily-oriented graph augmentation module (Homo-Aug) and a frequency adaptive attention module (FaAt). Specifically, the Homo-Aug extracts user representations and computes a k-NN graph using an MLP and improves Twitter's homophily by injecting the k-NN graph. For the FaAt, we propose an attention mechanism that adaptively serves as a low-pass filter along a homophilic edge and a high-pass filter along a heterophilic edge, preventing user features from being over-smoothed by their neighborhood. We also introduce a weight guidance loss to guide the frequency adaptive attention module. Our experiments demonstrate that HOFA achieves state-of-the-art performance on three widely-acknowledged Twitter bot detection benchmarks, which significantly outperforms vanilla graph-based bot detection techniques and strong heterophilic baselines. Furthermore, extensive studies confirm the effectiveness of our Homo-Aug and FaAt module, and HOFA's ability to demystify the heterophilous disguise challenge.

cs.SI↗

arXiv2022

Hot subdwarf stars identified in LAMOST DR8 with single-lined and composite spectra

222 hot subdwarf stars were identified with LAMOST DR8 spectra, among which 131 stars show composite spectra and have been decomposed, while 91 stars present single-lined spectra. Atmospheric parameters of all sample stars were obtained by fitting Hydrogen (H) and Helium (He) line profiles with synthetic spectra. Two long-period composite sdB binaries were newly discovered by combining our sample with the non-single star data from Gaia DR3. One of the new systems presents the highest eccentricity (i.e., 0.5 +/- 0.09) among known wide sdB binaries, which is beyond model predictions. 15 composite sdB stars fall in the high probability binary region of RUWE-AEN plane, and deserve priority follow-up observations to further study their binary nature. A distinct gap is clearly presented among temperatures of cool companions for our composite-spectra sample. But we could not come to a conclusion whether this feature is connected to the formation history of hot subdwarf stars before their binary natures are confirmed.

astro-ph.SR↗

arXiv2014

Efficient Planning under Uncertainty with Macro-actions

Deciding how to act in partially observable environments remains an active area of research. Identifying good sequences of decisions is particularly challenging when good control performance requires planning multiple steps into the future in domains with many states. Towards addressing this challenge, we present an online, forward-search algorithm called the Posterior Belief Distribution (PBD). PBD leverages a novel method for calculating the posterior distribution over beliefs that result after a sequence of actions is taken, given the set of observation sequences that could be received during this process. This method allows us to efficiently evaluate the expected reward of a sequence of primitive actions, which we refer to as macro-actions. We present a formal analysis of our approach, and examine its performance on two very large simulation experiments: scientific exploration and a target monitoring domain. We also demonstrate our algorithm being used to control a real robotic helicopter in a target monitoring experiment, which suggests that our approach has practical potential for planning in real-world, large partially observable domains where a multi-step lookahead is required to achieve good performance.

cs.AI↗