BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Chenxi Li

Chenxi Li contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Computation and Languagecond-mat.mtrl-sciArtificial IntelligenceComputer Science and Game TheoryComputer Visioncond-mat.mes-hallMethodologyMultiagent Systemsphysics.app-phphysics.comp-ph

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
8works
0followers
10topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

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

Published work

8 published item(s)

preprint2026arXiv

Achieving Gold-Medal-Level Olympiad Reasoning via Simple and Unified Scaling

Recent progress in reasoning models has substantially advanced long-horizon mathematical and scientific problem solving, with several systems now reaching gold-medal-level performance on International Mathematical Olympiad (IMO) and International Physics Olympiad (IPhO) problems. In this paper, we introduce a simple and unified recipe for converting a post-trained reasoning backbone into a rigorous olympiad-level solver. The recipe first uses a reverse-perplexity curriculum for SFT to instill rigorous proof-search and self-checking behaviors, then scales these behaviors through a two-stage RL pipeline that progresses from RL with verifiable rewards to more delicate proof-level RL, and finally boosts solving performance with test-time scaling. Applying this recipe, we train a 30B-A3B backbone with SFT on around 340K sub-8K-token trajectories followed by 200 RL steps. The resulting model, SU-01, supports stable reasoning on difficult problems with trajectories exceeding 100K tokens, while achieving gold-medal-level performance on mathematical and physical olympiad competitions, including IMO 2025/USAMO 2026 and IPhO 2024/2025. It also demonstrates strong generalization of scientific reasoning to domains beyond mathematics and physics.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

LabBuilder: Protocol-Grounded 3D Layout Generation for Interactable and Safe Laboratory

Automated laboratories hold the promise of accelerating scientific discovery, yet their deployment is bottlenecked by the difficulty of designing safe and executable environments. While simulator-based design offers scalability, existing 3D scene generation methods are primarily tailored for household settings, optimizing for visual plausibility while neglecting the rigorous functional semantics and safety constraints essential for scientific experimentation. We present LabBuilder, an end-to-end system that generates and verifies 3D laboratory layouts from concise textual specifications. It operates through three tightly coupled components: LabForge first curates a meta-dataset of annotated assets and chemical knowledge, translating natural language specifications into structured protocols; building on these protocols, LabGen synthesizes laboratory layouts via an iterative, constraint-aware optimization strategy; finally, LabTouchstone evaluates the resulting layouts as a unified benchmark. Extensive experiments demonstrate that LabBuilder significantly outperforms existing state-of-the-art methods, producing laboratory environments that are not only realistic but also functionally valid and safe for complex experimental workflows.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

OrthoGeoLoRA: Geometric Parameter-Efficient Fine-Tuning for Structured Social Science Concept Retrieval on theWeb

Large language models and text encoders increasingly power web-based information systems in the social sciences, including digital libraries, data catalogues, and search interfaces used by researchers, policymakers, and civil society. Full fine-tuning is often computationally and energy intensive, which can be prohibitive for smaller institutions and non-profit organizations in the Web4Good ecosystem. Parameter-Efficient Fine-Tuning (PEFT), especially Low-Rank Adaptation (LoRA), reduces this cost by updating only a small number of parameters. We show that the standard LoRA update $ΔW = BA^\top$ has geometric drawbacks: gauge freedom, scale ambiguity, and a tendency toward rank collapse. We introduce OrthoGeoLoRA, which enforces an SVD-like form $ΔW = BΣA^\top$ by constraining the low-rank factors to be orthogonal (Stiefel manifold). A geometric reparameterization implements this constraint while remaining compatible with standard optimizers such as Adam and existing fine-tuning pipelines. We also propose a benchmark for hierarchical concept retrieval over the European Language Social Science Thesaurus (ELSST), widely used to organize social science resources in digital repositories. Experiments with a multilingual sentence encoder show that OrthoGeoLoRA outperforms standard LoRA and several strong PEFT variants on ranking metrics under the same low-rank budget, offering a more compute- and parameter-efficient path to adapt foundation models in resource-constrained settings.

0 citations0 reviewsNo code linkedOpen access
preprint2025arXiv

BEDA: Belief Estimation as Probabilistic Constraints for Performing Strategic Dialogue Acts

Strategic dialogue requires agents to execute distinct dialogue acts, for which belief estimation is essential. While prior work often estimates beliefs accurately, it lacks a principled mechanism to use those beliefs during generation. We bridge this gap by first formalizing two core acts Adversarial and Alignment, and by operationalizing them via probabilistic constraints on what an agent may generate. We instantiate this idea in BEDA, a framework that consists of the world set, the belief estimator for belief estimation, and the conditional generator that selects acts and realizes utterances consistent with the inferred beliefs. Across three settings, Conditional Keeper Burglar (CKBG, adversarial), Mutual Friends (MF, cooperative), and CaSiNo (negotiation), BEDA consistently outperforms strong baselines: on CKBG it improves success rate by at least 5.0 points across backbones and by 20.6 points with GPT-4.1-nano; on Mutual Friends it achieves an average improvement of 9.3 points; and on CaSiNo it achieves the optimal deal relative to all baselines. These results indicate that casting belief estimation as constraints provides a simple, general mechanism for reliable strategic dialogue.

0 citations0 reviewsNo code linkedOpen access
preprint2023arXiv

Post-Selection Inference for the Cox Model with Interval-Censored Data

We develop a post-selection inference method for the Cox proportional hazards model with interval-censored data, which provides asymptotically valid p-values and confidence intervals conditional on the model selected by lasso. The method is based on a pivotal quantity that is shown to converge to a uniform distribution under local alternatives. The proof can be adapted to many other regression models, which is illustrated by the extension to generalized linear models and the Cox model with right-censored data. Our method involves estimation of the efficient information matrix, for which several approaches are proposed with proof of their consistency. Thorough simulation studies show that our method has satisfactory performance in samples of modest sizes. The utility of the method is illustrated via an application to an Alzheimer's disease study.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Properties and device performance of BN thin films grown on GaN by pulsed laser deposition

Wide and ultrawide-bandgap semiconductors lie at the heart of next-generation high-power, high-frequency electronics. Here, we report the growth of ultrawide-bandgap boron nitride (BN) thin films on wide-bandgap gallium nitride (GaN) by pulsed laser deposition. Comprehensive spectroscopic (core level and valence band XPS, FTIR, Raman) and microscopic (AFM and STEM) characterizations confirm the growth of BN thin films on GaN. Optically, we observed that BN/GaN heterostructure is second-harmonic generation active. Moreover, we fabricated the BN/GaN heterostructure-based Schottky diode that demonstrates rectifying characteristics, lower turn-on voltage, and an improved breakdown capability (234 V) as compared to GaN (168 V), owing to the higher breakdown electrical field of BN. Our approach is an early step towards bridging the gap between wide and ultrawide-bandgap materials for potential optoelectronics as well as next-generation high-power electronics.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Stability of Oxygenated Groups on Pristine and Defective Diamond Surfaces

The surface functionalization of diamond has been extensively studied through a variety of techniques, such as oxidation. Several oxygen groups have been correspondingly detected on the oxidized diamond, such as COC (ester), CO (ketonic), and COH (hydroxyl). However, the composition and relative concentration of these groups on diamond surfaces can be affected by the type of oxygenation treatment and the diamond surface quality. To investigate the stability of the oxygenated groups at specific diamond surfaces, we evaluated through fully atomistic reactive molecular mechanics (FARMM) simulations, using the ReaxFF force field, the formation energies of CO, COC, and COH groups on pristine and defective diamond surfaces (110), (111), and (311). According to our findings, the COH group has the lowest formation energy on a perfect (110) surface, while the COC is favored on a defective surface. As for the (111) surface, the COC group is the most stable for both pristine and defective surfaces. Similarly, COC group is also the most stable one on the defective/perfect (311) surface. In this way, our results suggest that if in a diamond film the (110) surface is the major exposed facet, the most adsorbed oxygen group could be either COH or COC, in which the COC would depend on the level of surface defects.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

A Reactive Molecular Dynamics Study of Hydrogenation on Diamond Surfaces

Hydrogenated diamond has been regarded as a promising material in electronic device applications, especially in field-effect transistors (FETs). However, the quality of diamond hydrogenation has not yet been established, nor has the specific orientation that would provide the optimum hydrogen coverage. In addition, most theoretical work in the literature use models with 100% hydrogenated diamond surfaces to study electronic properties, which is far from the experimentally observed hydrogen coverage. In this work, we have carried out a detailed study using fully atomistic reactive molecular dynamics (MD) simulations on low indices diamond surfaces i.e. (001), (013), (110), (113) and (111) to evaluate the quality and hydrogenation thresholds on different diamond surfaces and their possible effects on electronic properties. Our simulation results indicate that the 100% surface hydrogenation in these surfaces is hard to achieve because of the steric repulsion between the terminated hydrogen atoms. Among all the considered surfaces, the (001), (110), and (113) surfaces incorporate a larger number of hydrogen atoms and passivate the surface dangling bonds. Our results on hydrogen stability also suggest that these surfaces with optimum hydrogen coverage are robust under extreme conditions and could provide homogeneous p-type surface conductivity in the diamond surfaces, a key requirement for high-field, high-frequency device applications.

0 citations0 reviewsNo code linkedOpen access