Searcharxiv⌕ Search

arXiv subjects

Miao Liu

Publications and source records attributed to Miao Liu.

At least 19 recordsLinked to original sources

ALIVE: Making Streaming Video Models Interactive

Streaming video LLMs can continuously observe the world, yet often handle user requests in isolation. We introduce ALIVE, an interactive streaming video model supporting proactive monitoring, streaming task coordination, and feedback-seeking interaction. We construct ALIVE-SFT-288K and ALIVE-RL-3K for two-stage training. Contrastive supervised fine-tuning teaches when and how to respond, while reinforcement learning with Behavior-Aligned Credit Assignment (BACA) combines trajectory-level and local advantages to refine decisions, responses, and silence across multi-turn interactions. We also introduce FSI-Bench to evaluate feedback seeking and subsequent task completion. ALIVE achieves the highest overall scores among the compared models on four public benchmarks and FSI-Bench, demonstrating strong streaming understanding and interactive capabilities.

cs.CV↗

LogT: Logically Think with Images for Visual Search

Thinking with images empowers vision-language models (VLMs) to use an external Zoom-in tool for fine-grained visual search. Recent works employ supervised fine-tuning (SFT) and reinforcement learning (RL) with tool-specific designs to enable such capabilities. However, achieving logical reasoning alongside accurate and efficient tool use remains a significant challenge. To address this, we shift the focus away from increasing RL complexity and instead identify SFT data quality as a key bottleneck. We then propose LogT (short for Logically Think with Images), a fully automated SFT data engine that explicitly enforces logic-chain-guided reasoning. LogT generates an SFT dataset featuring challenging, visually grounded questions and structured tool-use traces with human-like logic. We validate our approach by fine-tuning two base models, Qwen2.5-VL-7B and InternVL3.5-8B, on the dataset and evaluating them across seven benchmark test sets. We show that LogT-SFT based on Qwen2.5-VL-7B outperforms the state-of-the-art trained with both SFT and RL, despite using 3.1$\times$ less data and $>120\times$ fewer GPU hours. Furthermore, applying a simple RL recipe without tool-specific design on LogT-SFT yields a model that surpasses prior methods with complex RL setups by a substantial 9.59-point margin. In addition to strong performance gains, our model exhibits higher tool-use accuracy and efficiency, as well as more structured and logically consistent reasoning traces, directly validating the effectiveness of our approach.

cs.LG↗

HEIR: Harness Egocentric Intent for Human-Robot Interactions

For robots to assist, serve, and collaborate with humans, they must not only interpret human requests conveyed through language and multimodal intentional cues, but also execute long-horizon tasks that interleave navigation and manipulation. In this paper, we introduce HEIR, a resource for studying how to Harness Egocentric Intent for Human-Robot Interaction. HEIR pairs human speech, measured gaze, and egocentric video with robot observations and teleoperated navigation-manipulation trajectories, while each session preserves successive episodes, with each request fulfilled through multiple execution subtasks. The captured data, together with human-authored annotations, support the study of intent grounding, execution planning, and task completion within a unified interaction setting. To demonstrate the value of HEIR, we conduct extensive offline experiments to evaluate how models benefit from additional human-centric context across these tasks, together with real-robot experiments that assess execution task completion. HEIR thus provides a valuable resource for studying user-centric robot interaction in naturalistic household environments. Our demo and dataset are available online.

cs.RO↗

Collective phase modes in twisted $d$-wave superconducting bilayers

Twisted cuprate bilayers have been predicted to host high-temperature chiral $d+id'$ superconductivity, originating from higher-order Josephson coupling processes. In such two-dimensional superconducting systems, long-wavelength fluctuations in the phase of the superconducting order parameter constitute gapless collective modes and therefore remain significant even at zero temperature. Here, we perform a theoretical analysis of the low-energy phase fluctuations in twisted $d$-wave superconducting bilayers within a self-consistent harmonic approximation, systematically retaining Josephson coupling to all orders. We demonstrate that higher-order Josephson coupling processes lead to nontrivial modifications of the phase dynamics. The momentum-resolved summand of the relative-phase stiffness is nonzero even in the normal state because interlayer tunneling explicitly breaks the intralayer U(1) symmetry, but its momentum integral vanishes for the continuum dispersion. The relative-phase stiffness is smaller in the $d+id'$ phase than in the $d$-wave phase, while the overall-phase stiffness has the opposite behavior. Furthermore, phase fluctuations strongly soften the Josephson plasma frequency near a twist angle of $45^\circ$ and also substantially reduce the Josephson critical current.

cond-mat.supr-con↗

Quantum phase fluctuations quench chiral superconductivity in twisted overdoped cuprate bilayers

Twisted cuprate bilayers have been proposed to host chiral superconductivity at temperatures comparable to their large pairing scale, yet experimental signatures of broken time-reversal symmetry remain unclear. We show that quantum phase fluctuations strongly suppress chiral order in twisted overdoped cuprate bilayers despite robust intralayer pairing. At a $45^\circ$ twist, we analytically demonstrate that the chiral transition temperature obeys $T_χ\propto t^4$ when the single-electron interlayer tunneling amplitude $t$ is weak. This collapse originates from competition between the small Josephson energy scale supporting chirality and the temperature-dependent collective zero-point energy. Chirality consequently survives only in a narrow angular window at ultralow temperatures, with relative-phase plasma softening above the chiral transition. A distinct transport signature arising from the quantum phase fluctuations is a pronounced low-temperature enhancement of the critical current near a magic angle. These findings separate pairing and chiral ordering scales and offer a possible explanation for the difficulty of observing robust chiral superconductivity in twisted cuprates.

cond-mat.supr-con↗

Learning Motion-Expressive 4D Gaussian Representations from Monocular Videos with Flow Splatting

Reconstructing dynamic 3D scenes from monocular videos remains challenging because limited observations provide insufficient motion constraints. Conventional 4D Gaussian representations model spatiotemporal appearance efficiently, but the restricted temporal evolution of each primitive makes complex nonlinear dynamics difficult to represent compactly. We propose a motion-expressive 4D Gaussian representation with time-varying means and covariances. By analytically differentiating Gaussian trajectories, we obtain Gaussian velocities and introduce Flow Splatting to render optical flow while accounting for camera motion, enabling dense motion supervision. This formulation jointly models scene appearance and motion within a unified set of 4D Gaussian primitives. We further introduce an initialization strategy based on the sampling theorem and a velocity consistency regularizer to stabilize optimization. Experiments on DAVIS and NVIDIA Dynamic Scenes demonstrate a favorable trade-off between reconstruction quality and efficiency under challenging monocular settings.

cs.CV↗

CausalAlign: Causal Alignment of Video Generation Models for Consistent Ego-Motion

Diffusion-based video generation models (VGMs) can produce visually plausible videos, yet their camera ego-motion often drifts over long horizons. Camera ego-motion determines how the viewpoint changes between frames and is essential for maintaining consistent geometry and trajectories. To address this problem, we propose CausalAlign, a causal alignment framework with two components. First, we diagnose how precise ego motion is represented, used, and retained within VGMs using linear probes and causal interventions, revealing a rise-and-fall pattern during denoising. Second, we strengthen ego-motion state in VGM by aligning the representations at the identified layers to a 3D foundation model through delta-token space. This space encodes the changes between consecutive frames, allowing the alignment to focus on ego-motion dynamics rather than static scene content. With Wan2.2 fine-tuned on DL3DV and RealEstate10K, CausalAlign can largely improve motion accuracy, geometric consistency, and temporal smoothness, while preserving visual quality. For example, it improves geometric consistency by more than 20% relative to the base model. Moreover, with the fine-tuned VGM, CausalAlign improves the success rate of world action models (WAMs) by 10 points on mobile manipulation tasks of RoboCasa365 benchmark. Overall, CausalAlign improves ego-motion consistency in diffusion-based VGMs and yields noticeable gains in embodied models.

cs.CV↗

ReTrace: Visual Backtracking for Efficient Multimodal Large Language Models

Multimodal Large Language Models (MLLMs) encode images into dense visual token sequences, creating substantial attention and KV-cache overhead. Existing visual token compression methods reduce this cost by pruning low-saliency tokens in shallow layers, but this practice couples temporary computational deactivation with irreversible information deletion. We show that this coupling breaks an intrinsic inference pattern in MLLMs. Through layer-wise attention analysis, we identify \textbf{Visual Backtracking}: visual pivots remain spatially stable, yet in middle layers they actively retrieve context from tokens that appeared unimportant earlier. Thus, early low-saliency tokens can be inactive for computation while still being useful for later reasoning. Motivated by this finding, we propose \textbf{ReTrace}, a backtracking-aware compression framework that introduces a dormant token state. ReTrace caches non-pivot tokens outside expensive layer-wise computation, restores them during the backtracking window, and prunes only after their information has been absorbed. Experiments on Qwen2.5-VL and InternVL3 demonstrate nearly 90\% reductions in FLOPs and KV cache, a 2.83$\times$ prefilling speedup, and strong performance retention across complex vision-language benchmarks.

cs.CV↗

ConfAL-WM: Confidence-Guided Active Learning for Action-Conditioned World Models

Action-conditioned world models have become an important foundation for embodied prediction, planning, and synthetic data generation, but their errors under new task and scene distributions are often concentrated in localized spatiotemporal regions such as robot arms, manipulated objects, contact areas, and occluded objects. This paper presents ConfAL-WM, a confidence-guided active learning framework for post-training embodied world models. Building upon EnerVerse-AC (EVAC), we attach a lightweight confidence probe to UNet decoder features and predict dense confidence maps in the latent space. These maps are aggregated into task-, frame-, and patch-level scores, enabling data-budget allocation and localized training enhancement. Our pipeline trains the probe and warms up EVAC on a small target-domain subset. EVAC-v1 then supplies task-level acquisition and optional frame/patch weighting signals; all selected-data models are initialized from the original pretrained EVAC checkpoint for retraining. Experiments on RoboTwin2.0 at the default 40% data budget show that confidence-guided selection improves post-training quality, while dense frame and patch weighting offers complementary reconstruction and semantic gains compared with scalar reward, progress, and judge-based scoring baselines. A quick visual overview of this work is available at https://ConfAL-WM.github.io.

cs.RO↗

pymatgen: 15 years of community growth, new functionality, and future prospects

We present a retrospective on the Python Materials Genomics (pymatgen) library, reflecting on 15 years of community growth and established best practices. The current version of pymatgen builds on its robust, open-source foundation, emphasising its collaborative nature. Over the past decade, pymatgen has thrived as one of the largest open-source materials science codebases. We detail how pymatgen supports advanced computational materials science workflows, how it adapts to changing demands, and the lessons learned from its growing community, providing guidance beyond pymatgen's open-source community. Pymatgen will continue to evolve in response to the current rapidly changing landscape of computational materials science and materials informatics. We discuss the challenges and potential solutions to managing the community-driven evolution of pymatgen and provide an outlook on its future development.

cond-mat.mtrl-sci↗

A Generalist Hanabi Agent

Traditional multi-agent reinforcement learning (MARL) systems can develop cooperative strategies through repeated interactions. However, these systems are unable to perform well on any other setting than the one they have been trained on, and struggle to successfully cooperate with unfamiliar collaborators. This is particularly visible in the Hanabi benchmark, a popular 2-to-5 player cooperative card-game which requires complex reasoning and precise assistance to other agents. Current MARL agents for Hanabi can only learn one specific game-setting (e.g., 2-player games), and play with the same algorithmic agents. This is in stark contrast to humans, who can quickly adjust their strategies to work with unfamiliar partners or situations. In this paper, we introduce Recurrent Replay Relevance Distributed DQN (R3D2), a generalist agent for Hanabi, designed to overcome these limitations. We reformulate the task using text, as language has been shown to improve transfer. We then propose a distributed MARL algorithm that copes with the resulting dynamic observation- and action-space. In doing so, our agent is the first that can play all game settings concurrently, and extend strategies learned from one setting to other ones. As a consequence, our agent also demonstrates the ability to collaborate with different algorithmic agents -- agents that are themselves unable to do so. The implementation code is available at: $\href{https://github.com/chandar-lab/R3D2-A-Generalist-Hanabi-Agent}{R3D2-A-Generalist-Hanabi-Agent}$

cs.MA↗

Unifying Distributional Training for One-Step Visual Generation

Distributional training provides collective supervision for one-step visual generation by matching real and generated features in frozen representation spaces. We introduce a unified theoretical framework that separates distribution modeling from matching discrepancy and connects global objectives to pointwise feature updates through Wasserstein gradient flow. Under this framework, FD-Loss and Gaussian-kernel Drifting are recovered through Gaussian optimal transport and kernel-density-based KL matching, respectively. The framework motivates MGFlow, which models feature distributions with Gaussian mixtures at an adjustable granularity between global moments and sample-based representations. MGFlow supports both optimal transport and score-based matching, and couples mass-constrained sample assignment with paired component updates to address mode collapse that mixture expressivity alone does not resolve. On ImageNet $256\times256$, MGFlow substantially surpasses the FD-Loss baseline, achieving state-of-the-art results with 1.45 $\mathrm{FDr}^6$ on pMF-H and 1.64 on JiT-H. For text-to-image generation, MGFlow post-trains FLUX.2 [klein] 4B into a one-step generator that outperforms the original four-step model on both GenEval and PickScore.

cs.LG↗

From Scores to Samples: Elastic Forcing for Autoregressive Video Generation

Few-step autoregressive video generation commonly relies on Distribution Matching Distillation (DMD), requiring a bidirectional diffusion teacher and an online fake-score model. We instead learn the rollout distribution directly from reference videos, eliminating both score models during post-training. Our framework minimizes maximum mean discrepancy (MMD) in frozen self-supervised video representation spaces, using a hybrid Nyström--Monte Carlo estimator to balance approximation bias and sampling variance. Memory-efficient replay and gradient subsampling make this objective practical. Using the same architecture and initialization as Self-Forcing, our 1.3B model improves the VBench Total score from 83.80 to 84.64 while retaining 17 FPS. Removing auxiliary score models also enables 14B post-training on eight H200 GPUs. Beyond distillation, learning from reference videos enables the acquisition of new visual styles, semantic concepts, and spatial priors without a target-specific diffusion teacher.

cs.CV↗

Batch-fabrication of all-dielectric vapor cells enabling optically addressed Rydberg atom electrometry

Millimeter-scale atomic vapor cells can be accurately and economically batch-fabricated by anodically bonding silicon and glass wafers, enabling the manufacturing of miniature atomic clocks and quantum sensors. However, silicon's high dielectric constant and conductive losses at millimeter wave frequencies limit its suitability for Rydberg-atom electrometry, which enables highly sensitive electric-field measurements by exploiting the extreme polarizability of Rydberg states in alkali atoms. To address this, we present an all-glass wafer-level microfabrication process that eliminates silicon, creating hermetically sealed vapor cells that are stable over long timelines with embedded cesium dispensers. Femtosecond laser machining precisely defines the cell geometry, and laser-activated alkali loading ensures reliable filling. We demonstrate long-term vacuum stability and robust Rydberg excitation through electromagnetically induced transparency measurements. We then use these cells to measure a 34 GHz millimeter wave field resonant with the $58D_{5/2}\rightarrow 60P_{3/2}$ transition using Autler-Townes splitting and observe the expected linear dependence with field strength. This work demonstrates that the all-glass bonding approach offers a highly durable low-loss cell alternative for miniaturized millimeter wave and microwave quantum sensing, with the potential to flexibly incorporate a range of other dielectric and semiconductor materials and integrate with photonic and electronic technologies.

physics.atom-ph↗

ORAV: Benchmarking Audio-Video Generation from Multimodal Contexts

Audio-video generation using heterogeneous multimodal references has emerged as a new challenge, requiring both compositional control over generation and grounded understanding of multimodal context. In this paper, we introduce ORAV Bench for Omni Reference Audio-Video Generation, comprising 380 task instances with 2-10 references, 9 semantic roles, and 30 role compositions. Instructions specify the relationships among references; the media supply the identities, dynamics, and audio characteristics to be realized. To evaluate these open-ended outputs, we develop a reference-aware pairwise protocol that prepares visual and auditory evidence, compares the intended contribution of each reference, and checks the overall verdict in both presentation orders. On held-out instances, it achieves 86.08% effective agreement with human judgments. Across 5 frontier systems, overall rankings conceal distinct strengths across reference compositions. A recurring failure is to reproduce unintended source content in place of the requested result, despite closely resembling a reference. Reproducible pointwise diagnostics of quality, reference affinity, and speech reveal distinct dimensions of model behavior. ORAV thus offers a benchmark for tracking progress toward controllable, compositional, and reference-faithful audio-video generation.

cs.CV↗

The Roadmap of Inorganic Computational Materials Databases: Capabilities, Credibility, Coverage, and the Open Frontier

Computational materials databases have become central infrastructure for data-driven discovery of inorganic materials, yet their growth remains strikingly uneven across property families. This perspective synthesizes a systematic survey of mainstream density functional theory (DFT) software, the computational cost and credibility of nineteen material-property families, and the coverage of existing computational databases, into a coherent picture of where the field stands and where it should go. We show that the ecosystem of first-principles codes is methodologically mature: for nearly every property of technological interest, at least one production-grade code can compute it.The binding constraint is no longer methodological capability but the economics of trust - which properties can be computed cheaply enough, and accurately enough, to be harvested at database scale. Mapping database coverage onto a Gartner-style readiness cycle reveals a sharp divide: ground-state structure, energetics, elasticity, and topology have reached routine production, while nine property families - including NMR/EPR parameters, core-level spectra, electron-phonon properties, thermal conductivity, and quantum transport - remain without any systematic computational database. We argue that these blank zones define the scientific opportunity of the next decade, and we propose a three-horizon roadmap: consolidating coverage and interoperability in the near term, industrializing mid-cost properties through surrogate-accelerated workflows in the medium term, and conquering the high-cost frontier through machine-learned interatomic potentials, autonomous computing infrastructure, and community governance in the long term.

cond-mat.mtrl-sci↗

A proof of Chvátal's conjecture via a sharp correlation inequality

We prove Chvátal's conjecture, posed in 1972: every hereditary family of subsets of a finite set has a largest intersecting subfamily that is a star. More generally, we prove a sharp correlation inequality for increasing Boolean functions $f,g:\{0,1\}^n\to\{0,1\}$. Writing $g^*(x)=1-g(1-x)$, we show that $$ \sum_{\varnothing\ne S\subseteq[n]}\hat{g}(S)^2\max_{i\in S}\mathrm{Inf}_i[f]\le\frac{2\mathrm{Cov}(f,g)\mathrm{Cov}(f,g^*)}{\mathrm{Cov}(f,g)+\mathrm{Cov}(f,g^*)}. $$ When $g$ is antipodal, that is, $g=g^*$, this yields $\mathrm{Cov}(f,g)\ge\frac{1}{4}\min_{i\in[n]}\mathrm{Inf}_i[f]$, the correlation formulation of Chvátal's conjecture due to Friedgut, Kahn, Kalai and Keller.

math.CO↗

Extremal problems for cancellative and locally thin hypergraphs

We study Turán-type extremal problems for cancellative and locally thin uniform hypergraphs. An $r$-uniform hypergraph is $t$-cancellative if $(\cup_{i=1}^t A_i)\cup B\ne (\cup_{i=1}^t A_i)\cup C$ whenever $A_1,\ldots,A_t,B,C$ are distinct edges. Let $C_t(n,r)$ denote the maximum number of edges in such a hypergraph on $n$ vertices. For all fixed integers $t,k\ge2$, we prove that $C_{2(t-1)}(n,tk)=(1+o(1))\frac{\binom{n}{k}}{\binom{tk-1}{k-1}}$ as $n\to\infty$. In the case $t=2$, this shows that Füredi's 2012 upper bound for $C_2(n,2k)$ is asymptotically sharp. The lower bound uses locally sparse induced packings, while the upper bound follows from double counting and a matching argument. More generally, for integers $s\ge t\ge1$, an $r$-uniform hypergraph is locally $(s,t)$-thin if among any $s$ distinct edges, at least $t$ contain a vertex that lies in none of the other $s-1$ edges. This notion includes cancellative hypergraphs as special cases. We establish general upper and lower bounds for the corresponding extremal numbers and determine their polynomial order of growth under suitable divisibility assumptions.

math.CO↗