arXiv subjects

Yu Liu

Publications and source records attributed to Yu Liu.

At least 19 recordsLinked to original sources

arXiv2026

Difficulty-Estimated Policy Optimization

Recent advancements in Large Reasoning Models (LRMs), exemplified by DeepSeek-R1, have underscored the potential of scaling inference-time compute through Group Relative Policy Optimization (GRPO). However, GRPO frequently suffers from gradient signal attenuation when encountering problems that are either too trivial or overly complex. In these scenarios, the disappearance of inter-group advantages makes the gradient signal susceptible to noise, thereby jeopardizing convergence stability. While variants like DAPO attempt to rectify gradient vanishing, they do not alleviate the substantial computational overhead incurred by exhaustive rollouts on low-utility samples. In this paper, we propose Difficulty-Estimated Policy Optimization (DEPO), a novel framework designed to optimize the efficiency and robustness of reasoning alignment. DEPO integrates an online Difficulty Estimator that dynamically assesses and filters training data before the rollout phase. This mechanism ensures that computational resources are prioritized for samples with high learning potential. Empirical results show that DEPO achieves approximately 1.86x faster rollouts than DAPO in the profiled configuration, while improving benchmark performance over GRPO at comparable training cost. Our approach significantly lowers the computational barrier for training high-performance reasoning models, offering a more sustainable path for reasoning scaling.

cs.AI↗

arXiv2026

LISA: Linear-Indexed Sparse Attention for Efficient Long-Context Reasoning

Long chain-of-thought (CoT) reasoning models such as DeepSeek-R1 keep lengthening their inference chains under test-time scaling, yet standard self-attention incurs quadratic cost in context length, making long-CoT deployment costly in production. We therefore introduce LISA (Linear-Indexed Sparse Attention), a plug-and-play replacement for self-attention that requires no full pretraining. LISA combines a linear-attention branch for long-range memory with a Lightning Indexer that dynamically selects the most relevant tokens for a sparse self-attention branch. The two branches are fused through a learned gate, so the model retains broad context while focusing computation on informative tokens. We train LISA in two stages. In Stage 1, a linear-attention branch is integrated with sliding-window attention and distilled from a frozen teacher's full-attention distribution. In Stage 2, the Lightning Indexer replaces the static window and is trained with a per-head KL-divergence objective to mimic teacher attention patterns. The resulting model reduces eager-mode self-attention computation from quadratic in sequence length to linear in the number of generated tokens and the selected context budget. Experiments on DeepSeek-distilled Qwen models show that LISA cuts measured eager-mode mean sample latency by 23.9% at 16K context without degrading long-context reasoning quality. Across nine mathematical and code reasoning benchmarks, the gated LISA-1.5B model improves the equal-weight benchmark average by 4.2 percentage points relative to the full-attention baseline.

cs.AI↗

arXiv2026

RouteDRAG: Learning to Route and Control Agentic Execution for Multimodal Document RAG

Web-based applications increasingly rely on understanding long multimodal documents, where answering questions often requires combining information across pages, tables, and figures. Agentic retrieval-augmented generation addresses these tasks through iterative retrieval and reasoning. However, Agentic execution is not always necessary, introducing additional computation and potentially incorrect answers. Our empirical analysis reveals that Agentic execution provides query-dependent benefits, while additional operations may improve, preserve, or degrade answer accuracy. These findings motivate controlling both Agent invocation and subsequent execution. To address these challenges, we propose RouteDRAG, a learning-based framework comprising four components. Unified Document Index organizes multimodal content into a graph-backed index and derives query representations from initial retrieval. Adaptive Path Routing preserves exact answering and predicts the relative benefit of Agent over shorter execution paths. Dynamic Agent Execution maintains an evolving workspace and refines Planner proposals to control tool use and termination. Sequential Policy Learning first trains operation control from observed outcomes and then trains path routing using the resulting Agent's performance. Across two benchmarks, RouteDRAG achieves an average accuracy improvement of 3.47 percentage points over shorter-path-only execution while assigning only 79.3% of queries to Agent. Notably, it also surpasses uniform Agent execution on both benchmarks while requiring fewer answering calls. Our code is now available at https://github.com/Ereshkigal1/RouteDRAG.

cs.AI↗

arXiv2026

Constraint on magnetized black bounce spacetime from HFQPOs data and the selection of resonance models via information criterion

Black bounce spacetime, proposed by Simpson and Visser (SV), can describe the Schwarzschild solution, regular black holes, and traversable wormholes in a unified manner, depending on the value of the regularization parameter $a$. This paper primarily explores the dynamics of charged particle in the magnetized SV spacetime, and constrains the parameters of the SV spacetime along with its surrounding magnetic fields. The constraints are given by using $χ^2$ analysis combined with high-frequency quasi-periodic oscillation (HFQPO) data observed from three microquasars: GRS 1915+105, XTE 1550-564, and GRO 1655-40. The results indicate that the magnetic field significantly influences the position of the innermost stable circular orbit of charged particle and frequency distribution of epicyclic motion, which excites more resonance model variants, enhancing observational effects. Additionally, we employ the Akaike Information Criterion (AIC) to evaluate resonance model and its various variants. The support for different models from observational data shows significant variation: $E R_8$ as the best model is supported strongly, $ER_3$ model has moderate evidence of support, $ER_6$ and $ER_7$ models are considerably less support, while other resonance models have essentially no support. The result from AIC is broadly consistent with the results obtained from the Bayesian Information Criterion (BIC) and the Bayes factor. For models more supported by the observational data, the allowed ranges of the regularization parameter: $0\leq a/M<0.736$ ($68\%$ confidence level) suggests that HFQPOs data could support the magnetized black bounce spacetime as a regular black hole, and the smaller value of the regularization parameter indicates a possibility of the presence of quantum effects.

astro-ph.HE↗

arXiv2026

Zero-HLM: Generalizable Humanoid Loco-Manipulation without Real Robot Data

Learning generalizable humanoid loco-manipulation remains bottlenecked by costly real-world demonstrations. Simulation provides a scalable alternative, but requires visually faithful, physics-ready scenes and executable whole-body motions. We present Zero-HLM, a real-to-sim-to-real framework that transforms smartphone captures and simulated source demonstrations into scalable training data for humanoid loco-manipulation. Gaussian-guided agentic real-to-sim reconstruction aligns visual and physical scenes; motion-environment co-adaptation uses rollout feedback to correct whole-body references and interaction geometry. We generate 4,800 physically validated demonstrations for zero-shot transfer across three Unitree G1 tasks. Simulation-trained GR00T N1.7 achieves 66.7% mean full-task success, versus 46.7% with real-data training. On generalized configurations, expanded simulation training raises pooled success from 23% for base-trained real-data policies to 69%, without additional real demonstrations. Project page: https://young-bit.github.io/zero-hlm/

cs.RO↗

arXiv2026

RIGen: Beyond Action Following with Reciprocal Interaction in Generative Gameplay

Generative gameplay requires more than immediate responses to player actions.Actions should have persistent consequences, while the generated world should act according to its own state and rules, shaping the player's situation and subsequent choices. Video generation captures local dynamics, but limited history, partially observed game state, and the need for contextual behavior decisions make such interactions difficult to sustain through audiovisual prediction alone. We introduce \rig (Reciprocally Interactive Generation), a framework for reciprocal interaction in generative gameplay. \rig combines persistent state and configurable rules, contextual interaction reasoning, and action-conditioned audiovisual generation. Together, these components make interaction consequences and actor behavior jointly determine generated content while maintaining consistency between game state and visible outcomes. We construct a Xonotic gameplay dataset to train an audiovisual generator for learning dynamics and interactions. Further post-training enables \rigarena, an interactive generative first-person shooter (FPS) prototype. Experiments demonstrate more reliable combat outcomes and reciprocal interaction between the player and the generated world. Project page: https://murphylmf.github.io/RIGen/.

cs.CV↗

arXiv2026

Recursive Reasoning or Statistical Extrapolation? In-Context Learning in Multi-Agent Interdependent Decision-Making

In-context learning (ICL) enables large language model (LLM) agents to improve decisions using interaction history, yet it remains unclear whether such improvement reflects refined internal reasoning or mere extrapolation of statistical patterns. To disentangle these mechanisms, we study LLM agents in multi-agent incomplete-information games that require recursive belief reasoning. By constructing a public goods game and manipulating the statistical structure of historical feedback, we evaluate decision quality against a history-independent rational expectations equilibrium (REE) benchmark. Our experiments reveal that when historical statistical patterns are disrupted, the benefits of longer context largely vanish, degrading decision quality to the no-context baseline in a way sharply amplified by stronger strategic interdependence. These results suggest that, in such strategic environments, ICL behavior is more consistent with statistical extrapolation than with strategic reasoning. Our work extends the mechanistic study of ICL to strategic multi-agent settings, introduces REE as a diagnostic tool for distinguishing reasoning from extrapolation, and provides a reusable framework for probing the boundaries of LLM reasoning in recursive belief tasks.

cs.AI↗

arXiv2026

WorldCast: World-Consistent Character Video Generation with Large-Angle Camera Control

Precise camera control is essential for cinematic character video generation, yet existing methods struggle to follow large-angle trajectories with combined rotation, translation, and radius variation from arbitrary initial azimuths. Such viewpoint changes also reveal character regions unseen in the initial frame, causing geometric distortions, appearance degradation, and identity drift. To systematically study these coupled challenges, we introduce CLACT, a large-scale Character-centric dataset and benchmark covering diverse Large-Angle Camera Trajectories, with accurate camera geometry and multi-view character observations for training and evaluation. We further propose WorldCast, a framework for world-consistent character video generation under diverse large-angle camera motions. Specifically, we back-project the initial frame into a 3D point cloud using estimated depth, align the point-cloud scale with the target camera trajectory, and render it along the trajectory to obtain explicit geometric guidance and coverage masks. To complement the incomplete character information, we introduce Geometry-Aligned Memory Attention (GAMA). Using relative camera poses estimated jointly from the initial frame and multi-view character anchors, we express anchor and target-view geometry in a common coordinate frame. GAMA uses this geometric alignment to guide the retrieval and injection of view-relevant character features during generation. Evaluations across diverse large-angle camera trajectories demonstrate accurate trajectory following while preserving character geometry, appearance, and identity.

cs.CV↗

arXiv2026

Percolation Transition of K-Destinations Reveals Universal Backbone of Urban Mobility Networks

Cities are held together by a small subset of recurrent origin--destination ties, but extracting this backbone from noisy mobility data remains difficult, and most models still prioritize distance over connectivity. We build directed mobility networks from 48 months of mobile-phone data across eight U.S. cities (2018--2021) and apply a rank-based percolation filter that retains, for each origin, its top-K destinations. Defining K* as the smallest cutoff that yields a strongly connected network, we find that city-specific thresholds cluster tightly across space and time, with a representative value near $K\approx130$. Degree-preserving and gravity-style baselines do not recover this scale, underscoring the importance of selectively preserved long-range functional ties. We then define PPD(K), an origin-level mobility concentration measure, and show that socioeconomic associations peak around the percolation-derived integration regime. Together, these findings offer a principled method to extract a minimal mobility backbone and connect its structure to persistent socioeconomic variation.

physics.soc-ph↗

arXiv2026

Cross-Visibility-Gap Identity Persistence for 4D Scene Graphs

4D scene graphs require persistent object identity when an object disappears inside an opaque carrier and later reappears as a fresh detection. When appearance and target-centric motion are unavailable, identity recovery must rely on relational history. We formulate cross-visibility-gap identity persistence as historical-node recovery through a remembered target--carrier binding. We introduce CVG-SGM, a perception-frontend-compatible transfer--maintain--read memory. Geometry-weighted cross-node interaction transfers target-specific evidence into carrier states, a shared per-node GRU maintains this information during transport, and a two-time readout combines stored carrier--identity compatibility with current detection--carrier geometry to recover the historical identity. In the controlled one-identity-per-carrier setting, CVG-SGM achieves 1.000 re-link accuracy, compared with 0.570 for both no-memory and parameter-count-matched no-write controls. It also generalizes to scene-disjoint, unseen longer visibility gaps. On real RGB-D/SAM2 feature-level episodes, it correctly re-links 36/36 held-out detections, compared with 28/36 for the strongest fixed carrier--identity control. Counterfactual carrier-memory swaps redirect predictions according to the swapped bindings, supporting the role of stored relational information in identity recovery. The current approach remains limited by shared-carrier identity ambiguity and does not provide end-to-end raw-video re-identification.

cs.CV↗

arXiv2026

Controlled-Z Gates via Floquet Coupling to Two Auxiliary States in Transmon Circuits

High-fidelity entangling gates are essential for quantum information processing. In transmon circuits, higher excited states provide additional pathways for engineering interactions within the computational subspace. Here we demonstrate controlled-Z (CZ) gates that exploit two noncomputational states, $\ket{02}$ and $\ket{20}$, through selective Floquet coupling. The nontrivial parametric modulation can activate both auxiliary transitions from $\ket{11}$ while suppressing the leading exchange between $\ket{01}$ and $\ket{10}$. Based on this approach, we develop two gate protocols: a nonadiabatic round trip through a bright auxiliary superposition and adiabatic evolution along a zero-energy dark state, both achieving average gate fidelities above $99.99\%$ in numerical simulations. Control of the modulation phase further extends the bright-state protocol to conventional and unconventional geometric CZ gates. Effective-model simulations show that unconventional control improves tolerance to residual $ZZ$ coupling and static qubit-frequency offsets relative to conventional control over the parameter ranges examined. Simulations including neighboring spectator qubits show that the original nonadiabatic bright-state and adiabatic dark-state CZ protocols retain high gate fidelities. This framework offers new routes toward the experimental realization of CZ gates and their geometric implementations.

quant-ph↗

arXiv2026

CAESAR: Mixed Reality Search-and-Rescue Platform for Cognitive Attack Evaluation

Mixed Reality (MR) enables immersive training by tightly coupling virtual content with the physical environment and supporting embodied interaction. In safety-critical collaborative tasks such as search-and-rescue, however, users rely heavily on virtual elements for perception, coordination, and decision-making, making them vulnerable to cognitive attacks that subtly distort visual feedback, virtual content, or system synchronization. While prior MR security research has examined individual attack mechanisms, evaluations are typically conducted in isolated or single-user settings, and there is a lack of standardized collaborative testbeds for systematic adversarial experimentation. In this paper, we develop CAESAR, an end-to-end multi-user MR framework for systematic evaluation of cognitive attacks in collaborative search-and-rescue training. CAESAR features a modular architecture that integrates four classes of cognitive attacks within a synchronized collaborative scenario and incorporates a multimodal data collection pipeline that captures behavioral, performance, and subjective measures under adversarial conditions. We conduct a controlled user study with an in-depth analysis of the latency-based attack, demonstrating measurable effects on task performance and user behavior while characterizing users' subjective experience under adversarial conditions.

cs.HC↗

arXiv2026

CTFS : Collaborative Teacher Framework for Forward-Looking Sonar Image Semantic Segmentation with Extremely Limited Labels

As one of the most important underwater sensing technologies, forward-looking sonar exhibits unique imaging characteristics. Sonar images are often affected by severe speckle noise, low texture contrast, acoustic shadows, and geometric distortions. These factors make it difficult for traditional teacher-student frameworks to achieve satisfactory performance in sonar semantic segmentation tasks under extremely limited labeled data conditions. To address this issue, we propose a Collaborative Teacher Semantic Segmentation Framework for forward-looking sonar images. This framework introduces a multi-teacher collaborative mechanism composed of one general teacher and multiple sonar-specific teachers. By adopting a multi-teacher alternating guidance strategy, the student model can learn general semantic representations while simultaneously capturing the unique characteristics of sonar images, thereby achieving more comprehensive and robust feature modeling. Considering the challenges of sonar images, which can lead teachers to generate a large number of noisy pseudo-labels, we further design a cross-teacher reliability assessment mechanism. This mechanism dynamically quantifies the reliability of pseudo-labels by evaluating the consistency and stability of predictions across multiple views and multiple teachers, thereby mitigating the negative impact caused by noisy pseudo-labels. Notably, on the FLSMD dataset, when only 2% of the data is labeled, our method achieves a 5.08% improvement in mIoU compared to other state-of-the-art approaches.

cs.CV↗

arXiv2026

ScaleNav: Scalable Internet Video Pretraining for Language-Conditioned Local Navigation

We study language-conditioned local navigation, where a robot must navigate to a text-specified target visible in its initial observation and stop at the described location. End-to-end navigation policies require diverse training data to generalize across scenes, yet real-robot data is costly to collect and simulated data leaves a sim-to-real gap. Internet egocentric videos offer a scalable source of real-world navigation experience. Our key insight is to learn goal-directed visual dynamics from Internet egocentric videos, and use these dynamics to support robot action generation. To this end, we propose ScaleNav, a unified world-action framework with three-stage training. We first perform instruction-conditioned video pretraining on these videos to learn how observations evolve as the camera approaches instructed targets across diverse real-world scenes. Action alignment then uses action-labeled simulated data to train the model to predict executable action chunks. Joint fine-tuning subsequently optimizes all trainable components together for navigation. ScaleNav achieves success rates of 86.5% on a single-view adaptation of Short-Horizon OVON and 94.9% on our HabitatGS-LN benchmark, outperforming the evaluated baselines on both benchmarks. We further deploy and evaluate ScaleNav on a real quadruped robot, with results supporting the model's ability to generalize to real-world navigation. Code and videos are available at https://zju3dv.github.io/ScaleNav/

cs.RO↗

arXiv2026

Generative molecule evolution using 3D pharmacophore for efficient Structure-Based Drug Design

Recent breakthroughs in generative models have shown great promise for drug discovery, yet Structure-Based Drug Design (SBDD) faces two critical challenges: the scarcity of high-quality protein-ligand complex data and the practical synthesizability of generated molecules. While ligand-based approaches benefit from billion-scale small molecule datasets, they often lack the structural context of the target protein. To address these dual challenges of data scarcity and synthetic inaccessibility, we propose MEVO, a generative molecular evolutionary framework that operates within a synthesizable chemical space. MEVO introduces a reaction-aware generative model trained on reaction routes constructed by applying predefined reaction templates to Enamine Building Blocks. The model jointly generates molecules and corresponding synthesis sketches consistent with these templates. By integrating this generative capability into a pocket-aware evolutionary strategy, MEVO iteratively refines molecular structures within the synthesizable chemical space, guided by a hierarchical scoring function that combines fast approximation with rigorous Absolute Binding Free Energy Perturbation (ABFEP) calculations. This framework efficiently generates high-affinity, synthetically feasible binders for diverse protein targets. We validate MEVO's performance through rigorous ABFEP calculations, demonstrating its ability to design potent inhibitors for challenging targets such as KRAS$^{\textrm{G12D}}$, achieving affinities comparable to known active inhibitors. By integrating the data richness of ligand-based learning with the precision of structure-based optimization, MEVO offers a holistic solution that ensures both high binding affinity and synthetic accessibility, paving the way for more efficient and practical structure-based drug design.

cs.LG↗

arXiv2026

In-context Robot Learning Made Simple: A Democratized Recipe for Manipulation Tasks

We study robotic in-context learning (ICL), an emerging paradigm that enables robots to infer and execute tasks from visual demonstrations. Despite its growing promise, the problem itself remains under-defined: a visual demonstration simultaneously conveys action trajectories, object semantics, manipulation affordances, spatial relations, and task goals, making it unclear what information the robot is actually expected to follow. In this work, we first provide a clear problem definition of robot ICL that explicitly defines its learning target and resolves this fundamental prompt ambiguity. Building on this definition, we develop a minimalist and reproducible ICL framework (SimpleICL) with a visual prompt encoder and a low-cost data collection protocol. Without massive pre-training or specialized data infrastructure, our framework achieves strong performance in both simulation and real-world environments. Extensive experiments further reveal several key properties of robot ICL, including semantic discrimination and task-relevant disentanglement. We will fully open-source our data and training pipeline to facilitate systematic and reproducible research on robot ICL. The project page can be found at https://simpleicl.github.io/simpleicl.

cs.RO↗

arXiv2026

DLOSAC: Joint Spatial-Geometric Image Estimation via Delaunay-Guided Local Optimization RANSAC

Reliable image geometric estimation from noisy matching points is typically addressed using RANSAC, often combined with local optimization to refine inliers and model parameters. However, existing paradigms are predominantly residual-centric; even heuristic attempts to enforce spatial uniformity suffer from a sequential, two-stage nature that inherently decouples distribution optimization from geometric refinement, preventing an effective trade-off. In this paper, we propose DLOSAC, a local optimization framework that jointly incorporates both geometric consistency and spatial distribution information. We begin by constructing a Delaunay triangulation over candidate inliers to explicitly encode neighborhood topology. The geometric module introduces an edge-based consistency metric to evaluate geometric errors, while the spatial module derives distribution descriptors from triangle areas and edge attributes to quantify the clustering of inliers. Finally, we combine edge-based geometric constraints, spatial distribution regularization, and traditional point-level geometric constraints to improve the quality of geometric parameters. Extensive experiments on six publicly available real-world datasets, covering three types of geometric estimation tasks, achieve state-of-the-art performance.

cs.CV↗

arXiv2026

OmniStepBench: Deployment-Aware Agent Evaluation across Modalities and Task Complexity

As large language models evolve from question-answering systems into general-purpose agents, evaluation must move beyond static answer correctness to assess multimodal perception, multi-step execution, tool use, and artifact delivery. However, existing benchmarks are often tied to specific task types, execution environments, or scoring protocols, limiting their comparability, interpretability, and reliability for deployment decisions. We introduce \textbf{OmniStepBench}, a benchmark for deployment-aware agent evaluation across modalities and task complexity. Built on a shared OpenClaw execution environment, OmniStepBench organizes 233 tasks selected and adapted from 22 source benchmarks into a $2\times3$ workload matrix defined by input modality and execution form, and evaluates them under a unified contract-based protocol with evidence-based score auditing. We evaluate 30 commercial API models and 12 locally deployed open-weight models (42 models in total) over 9{,}218 applicable model--task runs, reporting accuracy and token consumption alongside reference costs for API models. Results show workload-dependent rankings, distinct accuracy--cost trade-offs between text and multimodal tasks, and competitive but weaker overall performance from local models. These findings support deployment decisions based on workload profiles, deployment mode, and accuracy--cost trade-offs rather than a single aggregate score.

cs.AI↗