Search arXiv⌕ Search

arXiv subjects

Yao Zhao

Publications and source records attributed to Yao Zhao.

At least 19 recordsLinked to original sources
arXiv2026

Grounded Credit Assignment in Reinforcement Learning for Multi-Turn Search Agents

Outcome-reward reinforcement learning such as GRPO gives every token of a search-agent rollout the same advantage, although the rollout exercises two capabilities: finding evidence and using it. We show that accuracy improves without grounding: Search-R1 raises accuracy on all seven QA benchmarks, yet on Bamboogle the evidence-supported share of correct answers falls from 84.0% to 42.9%. We attribute this to GRPO's credit assignment: rollouts that retrieve answer-bearing evidence are favoured in aggregate only in proportion to their accuracy gain over the rest, a grounding gap that answering from memory or misusing evidence shrinks or reverses. To address this, we propose Grounded Credit Assignment (GCA), which uses verifiable sub-outcomes (evidence retrieval and answer correctness) to correct search and answer credit separately: a wrong answer is blamed more than the search that found the evidence, and a correct answer that bypasses retrievable evidence is credited less. Never negative on evidence-finding search, the correction does not shrink with the grounding gap; capped at 10% of each token's advantage, it never overrides the outcome signal. Without extra models, rollouts, or training stages, GCA improves average EM over Search-R1 by 5.7 (Qwen2.5-3B) and 6.3 points (Qwen3-30B-A3B) and raises this share from 76.9% to 93.4%.

cs.LG↗
arXiv2026

Halve-Then-Greedy: Adaptive Sampling for Posterior Success with Near-Optimal Instance Dependence

We study fixed-budget adaptive experimentation aimed at maximizing the probability of obtaining sufficient posterior evidence that a recommended design's mean performance exceeds a target. Unlike best-arm identification, this objective does not require finding the best design. For Gaussian bandits with a Gaussian prior, we propose \emph{Halve-Then-Greedy} (HTG), which combines a large initial balanced allocation, successive halving, and posterior-greedy sampling. We prove instance-dependent, finite-budget exponential bounds on posterior-failure probability that capture both separation from unfavorable arms and the presence of multiple favorable arms. On one-positive-arm families, we establish an asymptotic sample-complexity lower bound that HTG matches up to a logarithmic factor. We further analyze uniform sampling and sequential halving, revealing regimes where they require more samples than HTG by linear and logarithmic factors, respectively. Simulations illustrate HTG's sample-efficiency advantages and their dependence on instance structure.

cs.LG↗
arXiv2026

Towards Optimality in Active Preference Learning

Aligning large language models (LLMs) depends on high-quality datasets of human preference labels, which are costly to collect. Although active learning has been studied to improve sample efficiency relative to passive collection, many existing approaches adopt classical experimental design criteria such as G- or D-optimality. These objectives are not tailored to the structure of preference learning, leaving open the design of problem-specific algorithms. In this work, we identify a simple intuition specific to preference learning that calls into question the suitability of these existing design objectives. Motivated by this insight, we propose two active learning algorithms. The first provides the first instance-dependent label complexity guarantee for this setting, and the second is a simple, practical greedy method. We show the optimality of our instance-dependent term up to a constant depending on the logit bound, compared with a lower-bound characteristic. We evaluate our algorithms on real-world preference datasets and synthetic data observing improved sample efficiency compared to existing methods.

cs.LG↗
arXiv2026

Robust Ensemble Guidance for Scientific Inverse Problems

Ensemble guidance combines pretrained diffusion priors with black-box forward models to solve inverse problems without differentiating through the physical simulator. However, observation coordinates with large predictive spread or extreme residuals can dominate the ensemble correction, degrading reconstruction accuracy. We show that two simple modifications, weighting and clipping, substantially improve this correction. Our method, Robust Ensemble Guidance (REG), uses ensemble predictive spread to balance observation scales and adaptively clips standardized residuals to limit the influence of extreme discrepancies. Both operations reuse existing particles and forward predictions, requiring no additional denoiser or forward-model evaluations. Under a local linear Gaussian model, we derive conditions for reduced one-step estimation risk, bound the influence of individual observation coordinates, and characterize when these benefits persist with finite ensembles. Experiments on Navier-Stokes inversion, black-hole imaging, and acoustic full-waveform inversion demonstrate improved reconstruction over the underlying ensemble solver. In particular, REG increases black-hole reconstruction PSNR by 6.2-8.2 dB across three observation regimes and reduces Navier-Stokes reconstruction error by 26.4\% in a matched-budget comparison. These findings highlight the importance of observation heterogeneity and residual influence in designing reliable generative solvers for scientific inverse problems.

cs.LG↗
arXiv2026

Decoupling Spherical Reasoning from Dense Prediction for 360 Depth Estimation

The equirectangular projection (ERP) is widely used for panoramic depth estimation, but its spatially varying distortion makes geometry-consistent feature modeling challenging. We revisit panoramic depth estimation by decoupling contextual modeling in native spherical space from dense ERP prediction. To this end, we propose a Fibonacci Spherical Graph (FSG) as an intermediate reasoning space to lift ERP features onto quasi-uniform Fibonacci nodes on the sphere and capture local and long-range dependencies through complementary spherical neighborhoods. The resulting spherical discretization distributes graph nodes approximately uniformly over the spherical surface, reducing the over-representation of highly stretched regions during relational modeling. Operating on a compact set of Fibonacci nodes also avoids the computational burden of constructing and processing a graph at full ERP resolution. To bridge spherical reasoning and dense prediction, we propose a Spherical Context Conditioning (SCC) module that adaptively modulates dense ERP features with the enhanced spherical representation, allowing spherical context to guide pixel-aligned depth prediction. Extensive experiments on three benchmarks demonstrate that the proposed method consistently achieves superior depth accuracy over existing approaches.

cs.CV↗
arXiv2026

Learning Better Reasoning for Generative Recommendation with Semantic IDs

Generative recommendation reformulates item retrieval as sequence generation, allowing a unified model to directly generate the next item from a user's interaction history. Semantic IDs further make this paradigm effective and scalable by representing each item as discrete codes, enabling knowledge sharing among semantically related items. Recent studies introduce explicit reasoning before Semantic-ID generation, helping models summarize user interests and infer possible preference transitions. However, reasoning is not inherently beneficial: Inaccurate or uninformative reasoning may mislead subsequent item generation and ultimately degrade recommendation performance. This raises a central challenge: how can a recommender select and learn effective reasoning traces and progressively evolve toward better reasoning from its own generations? In this work, we propose Evo-Rec, a three-stage framework for learning better reasoning and further enhancing it through reinforcement learning. First, we align Semantic IDs with their textual and behavioral contexts, enabling the model to understand and generate item identifiers. Second, we sample multiple candidate reasoning traces and retain those that improve the prediction of the ground-truth item, providing a stronger reasoning initialization through supervised fine-tuning. Third, we further optimize the reasoning policy through reinforcement learning with catalog-constrained item generation and ranking-aware recommendation feedback. Experiments on three Amazon Review benchmarks show that Evo-Rec consistently outperforms discriminative, generative, and reasoning-enhanced recommenders across all evaluation metrics. These results demonstrate the effectiveness of our framework in learning better reasoning for SID-based generative recommendation.

cs.IR↗
arXiv2026

From Interests to Semantic IDs: Retrieval-Grounded Credit Assignment for Generative Recommendation

Semantic IDs (SIDs) encode each catalog item as a short token sequence, enabling generative recommenders to predict the next item autoregressively. Reasoning-enhanced variants, an increasingly common extension, first generate a textual trace and then decode a next-item SID by beam search. Such recommenders are commonly trained with group-relative policy optimization under an exact-match SID reward, which is sparse in large catalogs. Two failure modes follow. When all rollouts in a group miss the target, the group yields zero advantage and no learning signal. Rollouts sharing the same SID reward receive identical advantages, however much their traces differ. In both cases the reward reflects only the decoded SID, never the reasoning that produced it. This creates a credit-assignment gap. We address this gap with retrieval-grounded query attribution. Each trace is structured into a history summary, a set of interest hypotheses, and a final SID. A frozen retriever executes every hypothesis as a catalog query, so that each hypothesis becomes independently verifiable rather than judged only through the final SID. A rollout is rewarded when any of its queries retrieves the target within the \mbox{top-$K$}, and per-query hit indicators localize that reward to individual hypotheses. Credit is thus assigned at the span level: only hypotheses that individually hit receive positive retrieval advantage, while the retrieval channel never updates the final SID span. Rollouts that share a SID reward can therefore receive different updates. Across experiments on three Amazon Reviews datasets, this yields consistent improvements in SID recommendation. On Video Games, an oracle analysis further reveals the potential of interest-conditioned SID decoding: selecting the target-relevant query among generated interests improves both recall and ranking.

cs.IR↗
arXiv2026

Differences in Neurovascular Coupling in Patients with Major Depressive Disorder: Evidence from Simultaneous Resting-State EEG-fNIRS

Neurovascular coupling (NVC), the relationship between neural activity and cerebral hemodynamic responses, remains poorly understood in major depressive disorder (MDD). To investigate alterations in NVC associated with depressive symptom severity, we simultaneously recorded resting-state electroencephalography (rsEEG) and functional near-infrared spectroscopy (fNIRS) in 206 participants, including 134 patients with MDD and 72 healthy controls stratified by age to disentangle disease-related alterations from age-associated effects. NVC in the prefrontal cortex (PFC) was characterized by the consistency and temporal lag between spontaneous electrophysiological peaks and corresponding hemodynamic responses. We found that age significantly enhances the NVC consistency (p < 0.05), whereas this relationship was altered in patients with MDD during both the oxygen-consumption phase and the subsequent hemodynamic compensation phase. Furthermore, among patients with MDD, the strength of this consistency decreased more significantly with greater illness severity (partial r = -0.336, p = 0.0601). These findings suggest that changes in neurovascular coupling effects are strongly associated with the severity of depression. By leveraging wearable neuroimaging techniques, this study provides multimodal evidence for altered neurovascular coupling in depression and highlights its potential as a biomarker for disease monitoring and recovery trajectories.

q-bio.NC↗
arXiv2026

EchoPath: Execution-Level Replayable Memory for GUI Agents

Computer-use agents increasingly operate browsers, software, and desktop applications via CLI or API portals, but graphical user interface (GUI) still plays an important role in common industrial production scenarios. GUI agents commonly employ fresh observe-plan-ground-act loops, which is inefficient for enterprise tasks that repeatedly update records, process forms, configure tools, and export reports. We introduce EchoPath, a model-agnostic harness that converts artifact-validated GUI trajectories into standardized, parameter-controlled callable memories, analogous to Model Context Protocol (MCP)-style tool calls rather than unstructured experience records. Each memory stores task-intent keys, application and state preconditions, flexible input parameters, GUI evidence, validation provenance, and lifecycle state, so the host agent invokes a targeted procedure only when it can be deterministically replayed in the current runtime. The core mechanism enabling replay is an image-based target-reaiming algorithm that treats stored coordinates as visual evidence, matches the remembered GUI target against the current screen, and emits corrected operation coordinates before execution. During replay, EchoPath rebinds only declared modifiable inputs and rejects ambiguous or incompatible steps to bounded grounding repair or fresh planning. In experiments with real computer-use tasks, EchoPath reduced median token cost by more than 90% and median execution time by about 60%. These results support a bounded form of enterprise GUI memory: validated execution experience can become a controllable callable asset for recurrent work rather than only context for another reasoning pass.

cs.AI↗
arXiv2026

Causal inference with bipartite designs: A generalized propensity score approach

Bipartite experiments, in which one set of units receives a treatment while outcomes are measured on another set, have recently garnered attention for their ability to capture interference across two distinct populations, such as buyers and sellers in online marketplaces. However, analyzing these experiments can be challenging, given that exposure is neither purely binary nor independent across units. In this paper, we propose a unified framework for causal inference in bipartite designs that leverages generalized propensity scores (GPS) to estimate exposure-response functions. Under standard unconfoundedness assumptions, we show that our GPS-based estimators are unbiased and derive theoretical bounds on their variance. We further introduce practical modeling and weighting strategies, along with double deconfounding methods, that integrate the GPS into both the outcome model and the assignment mechanism. Through extensive simulations, we demonstrate that these approaches achieve substantial bias reduction compared to naive methods. We also illustrate their effectiveness in a real-world application using an Amazon Pet Supplies dataset, where controlling for network structure proves critical to drawing valid causal conclusions. Our results underscore the importance of bipartite designs in contexts with significant interference and highlight how GPS-based methods can bolster the reliability of causal effect estimates in such settings.

stat.ME↗
arXiv2026

Let ViT Speak: Generative Language-Image Pre-training

In this paper, we present \textbf{Gen}erative \textbf{L}anguage-\textbf{I}mage \textbf{P}re-training (GenLIP), a minimalist generative pretraining framework for Vision Transformers (ViTs) designed for multimodal large language models (MLLMs). To better align vision encoders with the autoregressive nature of LLMs, GenLIP trains a ViT to predict language tokens directly from visual tokens using a standard language modeling objective, without contrastive batch construction or an additional text decoder. This design offers three key advantages: (1) \textbf{Simplicity}: a single transformer jointly models visual and textual tokens; (2) \textbf{Scalability}: it scales effectively with both data and model size; and (3) \textbf{Performance}: it achieves competitive or superior results across diverse multimodal benchmarks. Trained on 8B samples from Recap-DataComp-1B, GenLIP matches or surpasses strong baselines despite using substantially less pretraining data. After continued pretraining on multi-resolution images at native aspect ratios, GenLIP further improves on detail-sensitive tasks such as OCR and chart understanding, making it a strong foundation for vision encoders in MLLMs.

cs.CV↗
arXiv2026

SphereSOD: Geometry-Structure Coupled Learning for 360 Salient Object Detection

360° salient object detection (SOD) aims to accurately segment salient regions across a full field of view. However, equirectangular projection (ERP) introduces severe spatial distortion when mapping the spherical domain onto a planar representation. Existing methods mainly focus on compensating projection distortion while overlooking the interaction between panoramic geometry and salient object structure during feature perception and prediction refinement. To this end, we propose SphereSOD, an ERP-native framework that couples panoramic geometry with evolving salient structures. Specifically, spherical geometry governs feature sampling and spatial weighting, while coarse-grained saliency and contour prediction influence context aggregation during the progressive decoding process. SphereSOD first initializes deformable sampling based on spherical projection geometry and then employs bounded, content-adaptive offsets, yielding features that are better aligned with the underlying panoramic geometry. Subsequently, the decoder performs structure-guided context aggregation and progressive refinement to recover complete salient regions and accurate boundaries. Extensive experiments on three public 360° SOD benchmarks demonstrate state-of-the-art performance and a favorable accuracy-efficiency trade-off, supporting structurepreserving inference directly in ERP space as a promising alternative to projection-heavy panoramic pipelines.

cs.CV↗
arXiv2026

Control of laser-plasma instabilities by non-collinear polychromatic light

Normal broadband lasers with collinear polychromatic components have immense potential for mitigating laser plasma instabilities (LPIs). However, the projection complexity of collinear polychromatic light (CPL) is a significant challenge owing to the demand for a large bandwidth and beamlet number. Here, we propose a theoretical LPI model and optical design for non-collinear polychromatic light (NCPL), which has a small angle $\sim4^\circ$ and large frequency difference $\sim$1\% between the double-color beamlets. LPI models of the NCPL demonstrate a decoupling threshold for the shared daughter waves under a multibeam configuration. Compared with the CPL, the wavevector couplings are further reduced by the introduced angle. Therefore, both the growth rate and saturation level of LPIs are greatly reduced by using the NCPL. The two- and three-dimensional simulation results indicate that the NCPL reduces the absolute and convective decoupling thresholds of the CPL and is sufficient to effectively mitigate the reflectivity, hot-electron generation, and intensity of cross-beam energy transfer. An optical design for the efficient generation of ultraviolet NCPL has been presented based on the unsaturated optical parametric amplification and non-collinear sum-frequency generation.

physics.plasm-ph↗
arXiv2026

Puffin-World: Scaling a Unified Multimodal Model with Native 3D World States

We propose Puffin-World, a unified multimodal architecture that integrates physical understanding, spatial simulation, and 3D world generation and reconstruction without relying on external offline modules. To reliably construct and interact with 3D worlds, our framework jointly models three native world states: physics (gravity field and latitude), geometry (depth), and appearance (image), together with a unified Omni-Camera representation that supports diverse tasks and flexible motions. Beyond modeling these states, we introduce a strategy for propagating physical dynamics across future frames. By grounding absolute camera properties in the real world, Puffin-World enables physically consistent and visually stable world generation. We further couple appearance and geometry within a single generative process, jointly synthesizing each future view and reconstructing its underlying geometry. This unified paradigm enables interleaved closed-loop applications requiring synergy across multiple tasks, including mimic and self-calibrated world exploration. To scale Puffin-World to complex scenarios, we construct Puffin-16M, comprising 15 million vision-language-camera triplets and 1 million trajectories featuring various and challenging motions. To foster further research in this area, we released the code, models, and datasets.

cs.CV↗
arXiv2026

Post-training Quantization for Hybrid Iterative Generative Models

Iterative Generative Models (IGMs) span autoregressive and diffusion paradigms, and hybrid variants that couple them can achieve remarkable image-generation fidelity. However, their iterative inference incurs substantial computational overhead, making Post-training Quantization (PTQ) appealing for acceleration, while directly applying vanilla PTQ to hybrid IGMs can trigger model collapse. By analyzing these failures, we identify two critical challenges: Excessive Outliers (EOs) in the activations create an irreconcilable trade-off between preserving normal precision and covering EOs, resulting in severe degradation in generation quality; Amplified Anomalies (AAs) arising unpredictably from minor quantization errors, create a mismatch between calibration and inference, thus iteratively triggering model collapse. To address these challenges, we introduce HyGenQ, a PTQ framework for hybrid IGMs. HyGenQ comprises Hierarchical Cluster Decoupling (HCD) and Scaling Recalibration (SR). HCD identifies and decouples outlier channels via a multi-stage clustering process, effectively isolating EOs while maintaining normal value precision, thereby alleviating performance degradation. SR scales AAs beyond Gaussian Bound, thereby avoiding model collapse caused by aggressive truncation. Extensive experiments demonstrate that HyGenQ successfully quantizes representative hybrid IGMs to 8-bit precision (W8A8), significantly outperforming existing baselines and validating its robustness across different model families.

cs.LG↗
arXiv2026

StateFlow: Building, Evolving, and Accessing 3D World States for Previsualization

Previsualization is an intermediate layer between ideas and production in film, games, architecture, and urban design. It lets creators iteratively refine scenes, actions, cameras, and spatial-temporal dynamics. Yet existing generative methods rely on simple prompts to jointly control all of these factors through one-shot image or video synthesis, offering weak controllability and limited support for iterative editing. Fundamentally, a world comprises multiple elements with geometry, appearance, and other attributes, together with cameras. Different frames are produced through local modifications or recombinations of this shared state, which is otherwise largely reused. Therefore, we argue that the missing component is an explicit and persistent working state. To address this, we present StateFlow, a state-centric framework for generative previsualization. Rather than generating videos in one shot, StateFlow uses an editable 3D world to organize scene structure, evolution, and cameras, while off-the-shelf video models enhance visual quality when higher fidelity is desired. This world is maintained as a persistent structured 3D state of scene elements and camera configurations, serving as the core working representation for previsualization. Built on this insight, StateFlow has three stages to construct, evolve, and access the world state. State construction lifts generated 2D content into a coherent 3D world through prior-guided, conflict-aware dual-view initialization, while State evolution translates user intent into structured state transitions while preserving world memory, avoiding full-scene regeneration for each edit. State access uses render-feedback reflection to refine camera plans into visually feasible trajectories, avoiding reliance on VLM semantics alone. Experiments show that StateFlow produces high-quality 3D worlds for video creation and game-like prototyping.

cs.CV↗
arXiv2026

Intelligence Foundation Model: A New Perspective to Approach Artificial General Intelligence

We propose a new perspective for approaching artificial general intelligence (AGI) through an intelligence foundation model (IFM). Unlike existing foundation models (FMs), which specialize in pattern learning within specific domains such as language, vision, or time series, IFM aims to acquire the underlying mechanisms of intelligence by learning directly from diverse intelligent behaviors. Vision, language, and other cognitive abilities are manifestations of intelligent behavior; learning from this broad range of behaviors enables the system to internalize the general principles of intelligence. Based on the fact that intelligent behaviors emerge from the collective dynamics of biological neural systems, IFM consists of two core components: a novel network architecture, termed the state neural network, which captures neuron-like dynamic processes, and a new learning objective, neuron output prediction, which trains the system to predict neuronal outputs from collective dynamics. The state neural network emulates the temporal dynamics of biological neurons, allowing the system to store, integrate, and process information over time, while the neuron output prediction objective provides a unified computational principle for learning these structural dynamics from intelligent behaviors. Together, these innovations establish a biologically grounded and computationally scalable foundation for building systems capable of generalization, reasoning, and adaptive learning across domains, representing a step toward truly AGI.

cs.AI↗
arXiv2026

Geometric phase-space nonseparability triggers giant optical shifts

Nonseparability among multiple degrees of freedom has enabled fundamental advances in structured light and related applications. Here we unveil a previously overlooked form of nonseparability in phase space, which we term geometric phase-space nonseparability. The latter arises solely from the wavefront curvature of a conventional wave packet, such as a fundamental Gaussian beam. This phase-space structure manifests as a position-dependent transverse-momentum distribution across the beam profile leading to the giant spatial and angular beam shifts upon reflection at a planar interface that we predict analytically and observe experimentally. Remarkably, the curvature-induced phase-space correlation remains robust against spatial-coherence degradation, allowing the giant shifts to persist even in the nearly incoherent regime. Our results establish wavefront curvature as a general mechanism for engineering beam shifts across optical, acoustic, and matter-wave systems.

physics.optics↗