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

Siqi Yang

Siqi Yang contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Computer VisionComputation and Languagehep-exhep-phArtificial IntelligenceMachine LearningMultimedia

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
6works
0followers
7topics
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

6 published item(s)

preprint2026arXiv

InstructAV2AV: Instruction-Guided Audio-Video Joint Editing

Recent diffusion-based methods have achieved impressive progress in video content manipulation. However, they typically ignore the accompanying audio, leaving the audio disjointed from the edited results. In this paper, we propose InstructAV2AV, the first end-to-end framework for instruction-guided audio-video joint editing. We first develop a scalable data synthesis pipeline and construct InsAVE-80K, the first large-scale audio-video editing dataset with high-quality source-to-target pairs. With this data foundation, we adapt an audio-video generation backbone to leverage its robust priors. We concatenate the audio-video input with noisy latent codes to anchor the source context, propose the source-instruction gated attention to improve instruction following and content preservation, and introduce a two-stage training strategy to effectively transfer these pre-trained priors. Extensive experiments demonstrate that InstructAV2AV outperforms state-of-the-art methods across 11 metrics spanning three aspects on two evaluation sets, highlighting its potential for controllable content creation. Project page: https://hjzheng.net/projects/InstructAV2AV/.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

V-FAT: Benchmarking Visual Fidelity Against Text-bias

Recent advancements in Multimodal Large Language Models (MLLMs) have demonstrated impressive performance on standard visual reasoning benchmarks. However, there is growing concern that these models rely excessively on linguistic shortcuts rather than genuine visual grounding, a phenomenon we term Text Bias. In this paper, we investigate the fundamental tension between visual perception and linguistic priors. We decouple the sources of this bias into two dimensions: Internal Corpus Bias, stemming from statistical correlations in pretraining, and External Instruction Bias, arising from the alignment-induced tendency toward sycophancy. To quantify this effect, we introduce V-FAT (Visual Fidelity Against Text-bias), a diagnostic benchmark comprising 4,026 VQA instances across six semantic domains. V-FAT employs a Three-Level Evaluation Framework that systematically increases the conflict between visual evidence and textual information: (L1) internal bias from atypical images, (L2) external bias from misleading instructions, and (L3) synergistic bias where both coincide. We introduce the Visual Robustness Score (VRS), a metric designed to penalize "lucky" linguistic guesses and reward true visual fidelity. Our evaluation of 12 frontier MLLMs reveals that while models excel in existing benchmarks, they experience significant visual collapse under high linguistic dominance.

0 citations0 reviewsNo code linkedOpen access
preprint2025arXiv

Reading or Reasoning? Format Decoupled Reinforcement Learning for Document OCR

Reading text from images or scanned documents via OCR models has been a longstanding focus of researchers. Intuitively, text reading is perceived as a straightforward perceptual task, and existing work primarily focuses on constructing enriched data engineering to enhance SFT capabilities. In this work, we observe that even advanced OCR models exhibit significantly higher entropy in formatted text (\emph{e.g.}, formula, table, etc.) compared to plain text, often by an order of magnitude. These statistical patterns reveal that advanced OCR models struggle with high output uncertainty when dealing with format sensitive document, suggesting that reasoning over diverse reading pathways may improve OCR performance. To address this, we propose format decoupled reinforcement learning (FD-RL), which leverages high-entropy patterns for targeted optimization. Our approach employs entropy-based data filtration strategy to identify format-intensive instances, and adopt format decoupled rewards tailored to different format types, enabling format-level validation rather than token-level memorization. FD-RL achieves an average score of 90.41 on OmniDocBench, setting a new record for end-to-end models on this highly popular benchmark. More importantly, we conduct comprehensive ablation studies over data, training, filtering, and rewarding strategies, thoroughly validating their effectiveness.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Factorization of the forward-backward charge asymmetry and measurements of the weak mixing angle and proton structure at hadron colliders

The forward-backward charge asymmetry (AFB) at hadron colliders is sensitive to both the electroweak (EW) symmetry breaking represented by the effective weak mixing angle, and the proton structure information in the initial state modeled by the parton distribution functions (PDFs). Due to their strong correlation, the precisions of the determination on the weak mixing angle and PDFs using the measured AFB spectrum are limited. In this paper, we define a set of structure parameters which factorize the unique proton information of the relative difference between quarks and antiquarks in the AFB observation. Other than the conventional way of extracting the weak mixing angle fro the convolution of PDF and EW calculations, we propose a new method to simultaneously determine the value of the weak mixing angle and the proton structure terms by fitting to the observed AFB distribution, and point out the necessity of specifying additional observations to further reduce the uncertainties on the proton structure terms respectively, so that the model-independent high precision measurements can be achieved at the future LHC experiments.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Reduction of the electroweak correlation in the PDF updating by using the forward-backward asymmetry of Drell-Yan process

We propose a new observable for the measurement of the forward-backward asymmetry $(A_{FB})$ in Drell-Yan lepton production. At hadron colliders, the $A_{FB}$ distribution is sensitive to both the electroweak (EW) fundamental parameter $\sin^2 θ_{W}$, the weak mixing angle, and the parton distribution functions (PDFs). Hence, the determination of $\sin^2 θ_{W}$ and the updating of PDFs by directly using the same $A_{FB}$ spectrum are strongly correlated. This correlation would introduce large bias or uncertainty into both precise measurements of EW and PDF sectors. In this article, we show that the sensitivity of $A_{FB}$ on $\sin^2 θ_{W}$ is dominated by its average value around the $Z$ pole region, while the shape (or gradient) of the $A_{FB}$ spectrum is insensitive to $\sin^2 θ_{W}$ and contains important information on the PDF modeling. Accordingly, a new observable related to the gradient of the spectrum is introduced, and demonstrated to be able to significantly reduce the potential bias on the determination of $\sin^2 θ_{W}$ when updating the PDFs using the same $A_{FB}$ data.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Unsupervised Domain Adaptive Object Detection using Forward-Backward Cyclic Adaptation

We present a novel approach to perform the unsupervised domain adaptation for object detection through forward-backward cyclic (FBC) training. Recent adversarial training based domain adaptation methods have shown their effectiveness on minimizing domain discrepancy via marginal feature distributions alignment. However, aligning the marginal feature distributions does not guarantee the alignment of class conditional distributions. This limitation is more evident when adapting object detectors as the domain discrepancy is larger compared to the image classification task, e.g. various number of objects exist in one image and the majority of content in an image is the background. This motivates us to learn domain invariance for category level semantics via gradient alignment. Intuitively, if the gradients of two domains point in similar directions, then the learning of one domain can improve that of another domain. To achieve gradient alignment, we propose Forward-Backward Cyclic Adaptation, which iteratively computes adaptation from source to target via backward hopping and from target to source via forward passing. In addition, we align low-level features for adapting holistic color/texture via adversarial training. However, the detector performs well on both domains is not ideal for target domain. As such, in each cycle, domain diversity is enforced by maximum entropy regularization on the source domain to penalize confident source-specific learning and minimum entropy regularization on target domain to intrigue target-specific learning. Theoretical analysis of the training process is provided, and extensive experiments on challenging cross-domain object detection datasets have shown the superiority of our approach over the state-of-the-art.

0 citations0 reviewsNo code linkedOpen access