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

Tianyi Xu

Tianyi Xu contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Artificial IntelligenceMachine Learningastro-ph.EPComputation and LanguageComputer Visioncond-mat.mes-halleess.IVmath.OCNetworking and Internet Architecture

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
7works
0followers
9topics
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

7 published item(s)

preprint2026arXiv

EVOCHAMBER: Test-Time Co-evolution of Multi-Agent System at Individual, Team, and Population Scales

We argue that multi-agent test-time evolution is not single-agent evolution replicated N times. A single-agent learner can only evolve its own context and memory. A multi-agent system additionally evolves who collaborates, how they collaborate, and how knowledge flows across the population. These components have no single-agent counterpart and can produce phenomena such as emergent specialization. Yet prior test-time methods either confine experiences to individual agents, forfeiting cross-agent learning, or broadcast symmetrically to all agents, erasing the specialization that makes collaboration valuable. We present EVOCHAMBER, a training-free framework that instantiates test-time evolution at three levels over a coevolving agent pool. At its core is CODREAM (Collaborative Dreaming), a post-task protocol triggered on team failure or disagreement, in which agents collaboratively reflect, distill insights, and route them asymmetrically from strong to weak agents on the failed niche, preserving specialization while filling knowledge gaps. Team-level operators assemble niche-conditioned teams and select collaboration structures online. Population-level lifecycle operators fork, merge, prune, and seed agents under performance pressure. On three heterogeneous task streams with Qwen3-8B, EVOCHAMBER reaches 63.9% on competition math, 75.7% on code, and 87.1% on multi-domain reasoning, outperforming the best baseline by 32% relative on math and confirming asymmetric cross-agent transfer as the primary driver in ablation. Starting from several identically initialized agents, four to five stable niche specialists spontaneously emerge, a structural signature of multi-agent evolution that no single-agent learner can express. See our code at: https://github.com/Mercury7353/EvoChamber

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

From Soliloquy to Agora: Memory-Enhanced LLM Agents with Decentralized Debate for Optimization Modeling

Optimization modeling underpins real-world decision-making in logistics, manufacturing, energy, and public services, but reliably solving such problems from natural-language requirements remains challenging for current large language models (LLMs). In this paper, we propose \emph{Agora-Opt}, a modular agentic framework for optimization modeling that combines decentralized debate with a read-write memory bank. Agora-Opt allows multiple agent teams to independently produce end-to-end solutions and reconcile them through an outcome-grounded debate protocol, while memory stores solver-verified artifacts and past disagreement resolutions to support training-free improvement over time. This design is flexible across both backbones and methods: it reduces base-model lock-in, transfers across different LLM families, and can be layered onto existing pipelines with minimal coupling. Across public benchmarks, Agora-Opt achieves the strongest overall performance among all compared methods, outperforming strong zero-shot LLMs, training-centric approaches, and prior agentic baselines. Further analyses show robust gains across backbone choices and component variants, and demonstrate that decentralized debate offers a structural advantage over centralized selection by enabling agents to refine candidate solutions through interaction and even recover correct formulations when all initial candidates are flawed. These results suggest that reliable optimization modeling benefits from combining collaborative cross-checking with reusable experience, and position Agora-Opt as a practical and extensible foundation for trustworthy optimization modeling assistance. Our code and data are available at https://github.com/CHIANGEL/Agora-Opt.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

SITA: Learning Speaker-Invariant and Tone-Aware Speech Representations for Low-Resource Tonal Languages

Tonal low-resource languages are widely spoken yet remain underserved by modern speech technology. A key challenge is learning representations that are robust to nuisance variation such as gender while remaining tone-aware for different lexical meanings. To address this, we propose SITA, a lightweight adaptation recipe that enforces Speaker-Invariance and Tone-Awareness for pretrained wav2vec-style encoders. SITA uses staged multi-objective training: (i) a cross-gender contrastive objective encourages lexical consistency across speakers, while a tone-repulsive loss prevents tone collapse by explicitly separating same-word different-tone realizations; and (ii) an auxiliary Connectionist Temporal Classification (CTC)-based ASR objective with distillation stabilizes recognition-relevant structure. We evaluate primarily on Hmong, a highly tonal and severely under-resourced language where off-the-shelf multilingual encoders fail to represent tone effectively. On a curated Hmong word corpus, SITA improves cross-gender lexical retrieval accuracy, while maintaining usable ASR accuracy relative to an ASR-adapted XLS-R teacher. We further observe similar gains when transferring the same recipe to Mandarin, suggesting SITA is a general, plug-in approach for adapting multilingual speech encoders to tonal languages.

0 citations0 reviewsNo code linkedOpen access
preprint2023arXiv

Online Learning for Adaptive Probing and Scheduling in Dense WLANs

Existing solutions to network scheduling typically assume that the instantaneous link rates are completely known before a scheduling decision is made or consider a bandit setting where the accurate link quality is discovered only after it has been used for data transmission. In practice, the decision maker can obtain (relatively accurate) channel information, e.g., through beamforming in mmWave networks, right before data transmission. However, frequent beamforming incurs a formidable overhead in densely deployed mmWave WLANs. In this paper, we consider the important problem of throughput optimization with joint link probing and scheduling. The problem is challenging even when the link rate distributions are pre-known (the offline setting) due to the necessity of balancing the information gains from probing and the cost of reducing the data transmission opportunity. We develop an approximation algorithm with guaranteed performance when the probing decision is non-adaptive, and a dynamic programming based solution for the more challenging adaptive setting. We further extend our solutions to the online setting with unknown link rate distributions and develop a contextual-bandit based algorithm and derive its regret bound. Numerical results using data traces collected from real-world mmWave deployments demonstrate the efficiency of our solutions.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Quantum cascade of new correlated phases in trigonally warped bilayer graphene

Divergent density of states offers the unique opportunity to explore a wide variety of correlated electron physics. In the thinnest limit, this has been predicted and verified in the ultra-flat bands of magic-angle twisted bilayer graphene, the band touching points of few-layer rhombohedral graphite, and the lightly doped rhombohedral trilayer graphene. The simpler and seemingly better understood Bernal bilayer graphene is also susceptible to orbital magnetism-driven phases at charge neutrality, such as layer antiferromagnet and quantum anomalous Hall octet. Here we report the discovery of a cascade of novel correlated phases in the vicinity of electric-field-controlled Lifshitz transitions and van Hove singularities in trigonally warped bilayer graphene. We provide compelling evidence for the observation of Stoner ferromagnets - half and quarter metals. More prominently, we identify signatures consistent with a topologically nontrivial Wigner-Hall crystal at zero magnetic field and its transition to a trivial Wigner crystal, as well as two correlated metals whose behavior deviates from standard Fermi liquids. Our results in this reproducible, tunable, simple system opens a new chapter for studying strongly correlated electrons.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Attention-based Residual Speech Portrait Model for Speech to Face Generation

Given a speaker's speech, it is interesting to see if it is possible to generate this speaker's face. One main challenge in this task is to alleviate the natural mismatch between face and speech. To this end, in this paper, we propose a novel Attention-based Residual Speech Portrait Model (AR-SPM) by introducing the ideal of the residual into a hybrid encoder-decoder architecture, where face prior features are merged with the output of speech encoder to form the final face feature. In particular, we innovatively establish a tri-item loss function, which is a weighted linear combination of the L2-norm, L1-norm and negative cosine loss, to train our model by comparing the final face feature and true face feature. Evaluation on AVSpeech dataset shows that our proposed model accelerates the convergence of training, outperforms the state-of-the-art in terms of quality of the generated face, and achieves superior recognition accuracy of gender and age compared with the ground truth.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Unveiling the secrets of the mid-infrared Moon

The Moon's optical characteristics in visible and long-wavelength infrared (LWIR) have long been observed with our eyes or with instruments. What the mid-infrared (MIR) Moon looks like is still a mystery. For the first time we present detailed appearance of the MIR Moon observed by a high-resolution geostationary satellite and reveal the essence behind its appearance. The appearance of the MIR Moon is opposite to its normal visible appearance. In addition the MIR Moon shows limb darkening. Both the absolute and the relative brightness distribution of the MIR lunar disk changes with the solar incidence angle. The signatures of the MIR Moon are controlled by both the reflection and emission of the lunar surface. We also show first-ever brightness temperature maps of the lunar disk without needing a mosaic, which better show the temperature variation across the lunar disk. They reveal that the relationship between brightness temperature and solar incidence angle i is cos1/bi, and the power parameter is smaller than the Lambertian temperature model of cos1/4i observed for lunar orbit-based measurements. The slower decrease of the brightness temperature when moving away from the sub-solar point than the Lambertian model is due to topographic effects. The brightness temperature is dominated by albedo and the solar incidence angle and influenced by the topography. Our results indicate that the Moon in the MIR exhibits many interesting phenomena which were previously unknown, and contains abundant information about lunar reflection and thermal emission for future study.

0 citations0 reviewsNo code linkedOpen access