arXiv subjects

Junyu Zhang

Publications and source records attributed to Junyu Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2026

Adaptive Newton-CG methods with global and local analysis for unconstrained optimization with Hölder continuous Hessian

In this paper, we study Newton-conjugate gradient (Newton-CG) methods for minimizing a nonconvex function $f$ whose Hessian is $(H_f,ν)$-Hölder continuous with modulus $H_f>0$ and exponent $ν\in(0,1]$. Recently proposed Newton-CG methods for this problem adopt (i) non-adaptive regularization and (ii) a nested line-search procedure, where (i) often leads to inefficient early progress and the loss of local superlinear convergence, and (ii) may incur high computational cost due to multiple solves of the Newton system per iteration. To address these limitations, we propose two novel Newton-CG algorithms, depending on the availability of $ν$, that adaptively regularize the Newton system by leveraging the auto-conditioning technique to eliminate the nested line search. The proposed algorithms achieve the best-known iteration complexity ${\mathcal O}\big(H_f^{1/(1+ν)}ε^{-(2+ν)/(1+ν)}\big)$ for finding an $ε$-stationary point and, simultaneously, enjoy local superlinear convergence near nondegenerate local minimizers. Numerical experiments further demonstrate the practical advantages of our algorithms over existing approaches.

math.OC↗

arXiv2026

Clumps in High-Redshift Galaxies: Mass Scaling and Radial Trends from JADES

Massive star-forming clumps are a prominent feature of high-redshift galaxies and are thought to trace gravitational fragmentation, feedback, and bulge growth in gas-rich disks. We present a statistical analysis of clumps in $\sim$3600 galaxies spanning $2 \lesssim z \lesssim 8$ from deep JWST/NIRCam imaging in the JADES GOODS--South field. Clumps are identified as residual features after subtracting smooth Sérsic profiles, enabling a uniform, rest-frame optical census of sub-galactic structure. We characterize their physical properties, size--mass relations, and spatial distributions to constrain models of sub-galactic structure formation and evolution. We find that clumps in our sample are typically low-mass ($10^{\sim7-8}M_\odot$), actively star-forming, and show diverse gas-phase metallicity, dust attenuation, and stellar population properties. Their sizes and average pairwise separations increase with cosmic time (toward lower redshift), consistent with inside-out disk growth. The clump mass function follows a power law with slope $α= -1.50_{-0.17}^{+0.19}$, consistent with fragmentation in turbulent disks. We find a deficit of relatively young clumps near galaxy centers and a radial transition in the size--mass relation: outer clumps exhibit steeper, near-virial slopes ($R_{\rm e}\propto M_*^{\sim 0.3}$), while inner clumps follow flatter trends ($R_{\rm e}\propto M_*^{\sim 0.2}$), consistent with structural evolution via migration or disruption. These results provide new constraints on the formation, survival, and dynamical evolution of clumps, highlighting their role in shaping galaxy morphology during the peak of cosmic star formation.

astro-ph.GA↗

arXiv2026

Shuffling the Stochastic Mirror Descent via Dual Lipschitz Continuity and Kernel Conditioning

The global Lipschitz smoothness condition underlies most convergence and complexity analyses via two key consequences: the descent lemma and the gradient Lipschitz continuity. How to study the performance of optimization algorithms in the absence of Lipschitz smoothness remains an active area. The relative smoothness framework from Bauschke-Bolte-Teboulle (2017) and Lu-Freund-Nesterov (2018) provides an extended descent lemma, ensuring convergence of Bregman-based proximal gradient methods and their vanilla stochastic counterparts. However, many widely used techniques (e.g., momentum schemes, random reshuffling, and variance reduction) additionally require the Lipschitz-type bound for gradient deviations, leaving their analysis under relative smoothness an open area. To resolve this issue, we introduce the dual kernel conditioning (DKC) regularity condition to regulate the local relative curvature of the kernel functions. Combined with the relative smoothness, DKC provides a dual Lipschitz continuity for gradients: even though the gradient mapping is not Lipschitz in the primal space, it preserves Lipschitz continuity in the dual space induced by a mirror map. We verify that DKC is widely satisfied by popular kernels and is closed under affine composition and conic combination. With these novel tools, we establish the first complexity bounds as well as the iterate convergence of random reshuffling mirror descent for constrained nonconvex relative smooth problems.

math.OC↗

arXiv2026

BACH-V: Bridging Abstract and Concrete Human-Values in Large Language Models

Do large language models (LLMs) genuinely understand abstract concepts, or merely manipulate them as statistical patterns? We introduce an abstraction-grounding framework that decomposes conceptual understanding into three capacities: interpretation of abstract concepts (Abstract-Abstract, A-A), grounding of abstractions in concrete events (Abstract-Concrete, A-C), and application of abstract principles to regulate concrete decisions (Concrete-Concrete, C-C). Using human values as a testbed - given their semantic richness and centrality to alignment - we employ probing (detecting value traces in internal activations) and steering (modifying representations to shift behavior). Across six open-source LLMs and ten value dimensions, probing shows that diagnostic probes trained solely on abstract value descriptions reliably detect the same values in concrete event narratives and decision reasoning, demonstrating cross-level transfer. Steering reveals an asymmetry: intervening on value representations causally shifts concrete judgments and decisions (A-C, C-C), yet leaves abstract interpretations unchanged (A-A), suggesting that encoded abstract values function as stable anchors rather than malleable activations. These findings indicate LLMs maintain structured value representations that bridge abstraction and action, providing a mechanistic and operational foundation for building value-driven autonomous AI systems with more transparent, generalizable alignment and control.

cs.CL↗

arXiv2026

Line-search and Adaptive Step Sizes for Nonconvex-strongly-concave Minimax Optimization

In this paper, we propose a novel reformulation of the smooth nonconvex-strongly-concave (NC-SC) minimax problems that casts the problem as a joint minimization. We show that our reformulation preserves not only first-order stationarity, but also global and local optimality, second-order stationarity, and the Kurdyka-Łojasiewicz (KL) property, of the original NC-SC problem, which is substantially stronger than its nonsmooth counterpart in the literature. With these enhanced structures, we design a versatile parameter-free and nonmonotone line-search framework that does not require evaluating the inner maximization. Under mild conditions, global convergence rates can be obtained, and, with KL property, full sequence convergence with asymptotic rates is also established. In particular, we show our framework is compatible with the gradient descent-ascent (GDA) algorithm. By equipping GDA with Barzilai-Borwein (BB) step sizes and nonmonotone line-search, our method exhibits superior numerical performance against the compared benchmarks.

math.OC↗

arXiv2026

Abundant Population of Broad H$α$ Emitters in the GOODS-N Field Revealed by CONGRESS, FRESCO, and JADES

We present a spectroscopic search for broad H$α$ emitters at z$\approx$3.7-6.5 in the GOODS-N field, utilizing JWST/NIRCam slitless spectroscopy from FRESCO and CONGRESS, complemented by JADES imaging. We identify 19 broad H$α$ emitters with FWHM$>$1000 km/s at z$\approx$4-5.5, including 9 new sources. The broad H$α$ luminosity function (LF) derived from our sample is consistent with those of other JWST-selected broad-line AGN reported in the literature. The black hole masses and AGN bolometric luminosities, inferred from the broad H$α$ components, indicate that most sources are accreting at ~10% of the Eddington limit. We derive their host stellar masses via SED fitting and find higher $M_{BH}/M_{*}$ ratios relative to the local $M_{BH}-M_{*}$ relations, consistent with previous studies. We find that 42% of the sample do not satisfy the widely-used color selection criteria for Little Red Dots (LRDs), with the majority of these sources lacking the characteristic steep rest-optical red slope, indicating that the LRD selection is highly incomplete when selecting AGN galaxies. A comparison of the average SEDs between our sample and LRDs selected in the same field reveals that the steep red slopes observed in some LRDs are likely due to line-boosting effects as previously suggested. Furthermore, we find that 68% of color-selected LRDs with H$α$ detections in the NIRCam/Grism spectra do not exhibit broad-line features. While the limited sensitivity of the grism spectra may hinder the detection of broad-line components in faint sources, our findings still highlight the enigmatic nature of the LRD population.

astro-ph.GA↗

arXiv2025

When Reasoning Meets Its Laws

Despite the superior performance of Large Reasoning Models (LRMs), their reasoning behaviors are often counterintuitive, leading to suboptimal reasoning capabilities. To theoretically formalize the desired reasoning behaviors, this paper presents the Laws of Reasoning (LoRe), a unified framework that characterizes intrinsic reasoning patterns in LRMs. We first propose compute law with the hypothesis that the reasoning compute should scale linearly with question complexity. Beyond compute, we extend LoRe with a supplementary accuracy law. Since the question complexity is difficult to quantify in practice, we examine these hypotheses by two properties of the laws, monotonicity and compositionality. We therefore introduce LoRe-Bench, a benchmark that systematically measures these two tractable properties for large reasoning models. Evaluation shows that most reasoning models exhibit reasonable monotonicity but lack compositionality. In response, we develop an effective finetuning approach that enforces compute-law compositionality. Extensive empirical studies demonstrate that better compliance with compute laws yields consistently improved reasoning performance on multiple benchmarks, and uncovers synergistic effects across properties and laws. Project page: https://lore-project.github.io/

cs.AI↗

arXiv2025

The Luminosity Function and Clustering of H$α$ Emitting Galaxies at $z\approx4-6$ from a Complete NIRCam Grism Redshift Survey

We study the luminosity function (LF) and clustering properties of 888 H$α$ emitters (HAEs) at $3.75 < z < 6$ in the GOODS-N field. The sample, built from JWST CONGRESS and FRESCO NIRCam grism surveys using a novel redshift assignment algorithm, spans $\sim$62 arcmin$^2$ and reaches $L_{\rm Hα} \sim 10^{41.2} {\rm erg s^{-1}}$. We identify two prominent filamentary protoclusters at $z \approx 4.41$ and $z \approx 5.19$, hosting 98 and 144 HAEs, respectively. The observed H$α$ LFs show similar shallow faint-end slopes for both protocluster and field galaxies at $z=3.75-5$, and for the protocluster at $z=5-6$ ($α\approx 1.2$ to $-1.3$). In contrast, the field LF at $z=5-6$ is much steeper ($α=-1.87_{-0.23}^{+0.30}$), suggesting that protocluster galaxies at $z > 5$ are more evolved, resembling those at $z=3.75-5$. The observed star formation rate density from H$α$, integrated down to 0.45 ${\rm M_\odot yr^{-1}}$, is $0.050^{+0.002}_{-0.003}$ and $0.046^{+0.006}_{-0.004} M_\odot {\rm yr}^{-1} {\rm Mpc}^{-3}$ at $z=3.75-5$ and $z=5-6$, with protoclusters contributing $\sim$25% and 55%, respectively. This implies that a large fraction of star formation at $z > 4$ occurs in protoclusters. We conduct the first star-formation-rate-limited 3D clustering analysis at $z > 4$. We find the filamentary protocluster geometry flattens the power-law shape of the HAE auto-correlation functions, with slopes much shallower than typically assumed. The auto-correlation function of field HAEs have correlation lengths of $r_0 = 4.61^{+1.00}_{-0.68} h^{-1}{\rm Mpc}$ at $z \approx 4-5$ and $r_0 = 6.23^{+1.68}_{-1.13} h^{-1}{\rm Mpc}$ at $z=5-6$. Comparing the observed correlation functions with the UniverseMachine simulation, we infer the dark matter (sub-)halo masses of HAEs to be $\log (M_h/M_\odot)=11.0-11.2$ at $z\approx 4-6$, with a scatter of 0.4 dex.

astro-ph.GA↗

arXiv2025

The Large-scale Environments of Low-luminosity AGNs at $3.9 < z < 6$ and Implications for Their Host Dark Matter Halos from a Complete NIRCam Grism Redshift Survey

We study the large-scale environments and clustering properties of 28 low-luminosity AGNs at $z=3.9-6$ in the GOODS-N field. Our sample, identified from the JWST NIRCam Imaging and WFSS data in CONGRESS and FRESCO surveys with either broad H$α$ emission lines or V-shape continua, are compared to 782 H$α$ emitters (HAEs) selected from the same data. These AGNs are located in diverse large-scale environments and do not preferentially reside in denser environments compared to HAEs. Their overdensity field, $δ$, averaged over (15 $h^{-1}$cMpc)$^3$, ranges from $-0.56$ to 10.56, and shows no clear correlation with broad-line luminosity, black hole (BH) masses, or the AGN fraction. It suggests that $> 10$ cMpc structures do not significantly influence BH growth. We measure the two-point cross-correlation function of AGNs with HAEs, finding a comparable amplitude to that of the HAE auto-correlation. This indicates similar bias parameters and host dark matter halo masses for AGNs and HAEs. The correlation length of field AGNs is 4.26 $h^{-1}$cMpc, and 7.66 $h^{-1}$cMpc at $3.9 < z < 5$ and $5 < z < 6$, respectively. We infer a median host dark matter halo mass of $\log (M_h/M_\odot)\approx 11.0-11.2$ and host stellar masses of $\log (M_\star/M_\odot) \approx 8.4-8.6$ by comparing with the UniverseMachine simulation. Our clustering analysis suggests that low-luminosity AGNs at high redshift reside in normal star-forming galaxies with overmassive BHs. They represent an intrinsically distinct population from luminous quasars and could be a common phase in galaxy evolution.

astro-ph.GA↗

arXiv2025

Quasar Radiative Feedback May Suppress Galaxy Growth on Intergalactic Scales at $z = 6.3$

We present observational evidence that intense ionizing radiation from a luminous quasar suppresses nebular emission in nearby galaxies on intergalactic scales at $z=6.3$. Using JWST/NIRCam grism spectroscopy from the SAPPHIRES and EIGER programs, we identify a moderate but statistically significant decline in [O\,\textsc{iii}]\,$\lambda5008$ luminosity relative to the UV continuum ($L_{5008}/L_{1500}$) among galaxies within $\sim$ 7 comoving Mpc (cMpc) of the quasar J0100$+$2802, the most UV-luminous quasar known at this epoch ($M_{1450}=-29.26$). While $L_{1500}$ remains roughly constant with transverse distance, $L_{5008}$ increases significantly, suggesting suppression of very recent star formation toward the quasar. The effect persists after controlling for completeness, local density, and UV luminosity, and correlates with the projected photoionization-rate profile $Γ_{\mathrm{qso}}$. A weaker but directionally consistent suppression in $L_{5008}/L_{1500}$ is also observed along the line of sight. The transverse suppression radius ($\sim$ 7 cMpc) implies a recent radiative episode with a cumulative duration $\sim$ 3.1 Myr, shorter than required for thermal photoheating to dominate and thus more naturally explained by rapid H$_2$ photodissociation and related radiative processes. Environmental effects alone appear insufficient to explain the signal. Our results provide direct, geometry-based constraints on large-scale quasar radiative feedback and recent quasar lifetimes.

astro-ph.GA↗

arXiv2025

The Augmented Lagrangian Methods: Overview and Recent Advances

Large-scale constrained optimization is pivotal in modern scientific, engineering, and industrial computation, often involving complex systems with numerous variables and constraints. This paper provides a unified and comprehensive perspective on constructing augmented Lagrangian functions (based on Hestenes-Powell-Rockafellar augmented Lagrangian) for various optimization problems, including nonlinear programming and convex and nonconvex composite programming. We present the augmented Lagrangian method (ALM), covering its theoretical foundations in both convex and nonconvex cases, and discuss several successful examples and applications. Recent advancements have extended ALM's capabilities to handle nonconvex constraints and ensure global convergence to first and second-order stationary points. For nonsmooth convex problems, ALM utilizes proximal operations, preserving desirable properties such as locally linear convergence rates. Furthermore, recent progress has refined the complexity analysis for ALM and tackled challenging integer programming instances. This review aims to offer a thorough understanding of ALM's benefits and limitations, exploring different ALM variants designed to enhance convergence and computational performance. We also illustrate effective algorithms for ALM subproblems across different types of optimization problems and highlight practical implementations in several fields.

math.OC↗

arXiv2025

Stochastic Bregman Proximal Gradient Method Revisited: Kernel Conditioning and Painless Variance Reduction

We investigate stochastic Bregman proximal gradient (SBPG) methods for minimizing a finite-sum nonconvex function $Ψ(x):=\frac{1}{n}\sum_{i=1}^nf_i(x)+ϕ(x)$, where $ϕ$ is convex and nonsmooth, while $f_i$, instead of gradient global Lipschitz continuity, satisfies a smooth-adaptability condition w.r.t. some kernel $h$. Standard acceleration techniques for stochastic algorithms (momentum, shuffling, variance reduction) depend on bounding stochastic errors by gradient differences that are further controlled via Lipschitz property. Lacking this, existing SBPG results are limited to vanilla stochastic approximation that cannot yield the optimal $O(\sqrt{n})$ complexity dependence on $n$. Moreover, existing works report complexities under various nonstandard stationarity measures that largely deviate from the standard minimal limiting Fréchet subdifferential $\mathrm{dist}(0,\partialΨ(\cdot))$. Our analysis reveals that these popular stationarity measures are often much smaller than $\mathrm{dist}(0,\partialΨ(\cdot))$, leading to overstated solution quality and non-stationary output. To resolve these issues, we design a new gradient mapping $\mathcal{D}_{ϕ,h}^λ(\cdot)$ by BPG residuals in dual space and a new kernel-conditioning (KC) regularity, under which the mismatch between $\|\mathcal{D}_{ϕ,h}^λ(\cdot)\|$ and $\mathrm{dist}(0,\partialΨ(\cdot))$ is provably $O(1)$ and instance-free. Moreover, KC-regularity guarantees Lipschitz-like bounds for gradient differences, providing general analysis tools for momentum, shuffling, and variance reduction under smooth-adaptability. We illustrate this point on variance reduced SBPG methods and establish an $O(\sqrt{n})$ complexity for $\|\mathcal{D}_{ϕ,h}^λ(\cdot)\|$, providing instance-free (worst-case) complexity under $\mathrm{dist}(0,\partialΨ(\cdot))$.

math.OC↗

arXiv2025

The First Photometric Evidence of a Transient/Variable Source at z>5 with JWST

The James Webb Space Telescope (JWST) discovered 79 transients out to $z$$\sim$4.8 through the JADES Transient Survey (JTS), but the JTS did not find any $z$$>$5 transients. Here, we present the first photometric evidence of a $z$$>$5 transient/variable source with JWST. The source, AT 2023adya, resides in a $z_{\mathrm{spec}}$$=$5.274 galaxy in GOODS-N, which dimmed from $m_{\rm F356W}$$=$26.05$\pm$0.02 mag to 26.24$\pm$0.02 mag in the rest-frame optical over approximately two rest-frame months, producing a clear residual signal in the difference image ($m_{\rm F356W}$$=$28.01$\pm$0.17 mag; SN$_\mathrm{var}$$=$6.09) at the galaxy center. Shorter-wavelength bands (F090W/F115W) show no rest-frame ultraviolet brightness change. Based on its rest-frame V-band absolute magnitude of M$_\mathrm{V}$$=$$-$18.48 mag, AT 2023adya could be any core-collapse supernova (SN) subtype or an SN Ia. However, due to low SN Ia rates at high redshift, the SN Ia scenario is unlikely. Alternatively, AT 2023adya may be a variable active galactic nucleus (AGN). However, the JWST NIRCam/Grism spectrum shows no broad H$α$ emission line (FWHM$=$130$\pm$26 km s$^{-1}$), disfavoring the variable AGN scenario. It is also unlikely that AT 2023adya is a tidal disruption event (TDE) because the TDE models matching the observed brightness changes have low event rates. Although it is not possible to determine AT 2023adya's nature based on the two-epoch single-band photometry alone, this discovery indicates that JWST can push the frontier of transient/variable science past $z$$=$5 and towards the epoch of reionization.

astro-ph.HE↗

arXiv2025

EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents

Leveraging Multi-modal Large Language Models (MLLMs) to create embodied agents offers a promising avenue for tackling real-world tasks. While language-centric embodied agents have garnered substantial attention, MLLM-based embodied agents remain underexplored due to the lack of comprehensive evaluation frameworks. To bridge this gap, we introduce EmbodiedBench, an extensive benchmark designed to evaluate vision-driven embodied agents. EmbodiedBench features: (1) a diverse set of 1,128 testing tasks across four environments, ranging from high-level semantic tasks (e.g., household) to low-level tasks involving atomic actions (e.g., navigation and manipulation); and (2) six meticulously curated subsets evaluating essential agent capabilities like commonsense reasoning, complex instruction understanding, spatial awareness, visual perception, and long-term planning. Through extensive experiments, we evaluated 24 leading proprietary and open-source MLLMs within EmbodiedBench. Our findings reveal that: MLLMs excel at high-level tasks but struggle with low-level manipulation, with the best model, GPT-4o, scoring only 28.9\% on average. EmbodiedBench provides a multifaceted standardized evaluation platform that not only highlights existing challenges but also offers valuable insights to advance MLLM-based embodied agents. Our code and dataset are available at https://embodiedbench.github.io.

cs.AI↗

arXiv2025

AlphaOne: Reasoning Models Thinking Slow and Fast at Test Time

This paper presents AlphaOne ($α$1), a universal framework for modulating reasoning progress in large reasoning models (LRMs) at test time. $α$1 first introduces $α$ moment, which represents the scaled thinking phase with a universal parameter $α$. Within this scaled pre-$α$ moment phase, it dynamically schedules slow thinking transitions by modeling the insertion of reasoning transition tokens as a Bernoulli stochastic process. After the $α$ moment, $α$1 deterministically terminates slow thinking with the end-of-thinking token, thereby fostering fast reasoning and efficient answer generation. This approach unifies and generalizes existing monotonic scaling methods by enabling flexible and dense slow-to-fast reasoning modulation. Extensive empirical studies on various challenging benchmarks across mathematical, coding, and scientific domains demonstrate $α$1's superior reasoning capability and efficiency. Project page: https://alphaone-project.github.io/

cs.CL↗

arXiv2025

Algorithms for zero-sum stochastic games with the risk-sensitive average criterion

This paper is an attempt to compute the value and saddle points of zero-sum risk-sensitive average stochastic games. For the average games with finite states and actions, we first introduce the so-called irreducibility coefficient and then establish its equivalence to the irreducibility condition. Using this equivalence,we develop an iteration algorithm to compute $\varepsilon$-approximations of the value (for any given $\varepsilon>0$) and show its convergence. Based on $\varepsilon$-approximations of the value and the irreducibility coefficient, we further propose another iteration algorithm, which is proved to obtain $\varepsilon$-saddle points in finite steps. Finally, a numerical example of energy management in smart grids is provided to illustrate our results.

math.OC↗

arXiv2025

SMILES: Potentially Higher Ionizing Photon Production Efficiency in Overdense Regions

The topology of reionization and the environments where galaxies efficiently produce ionizing photons are key open questions. For the first time, we investigate the trend between ionizing photon production efficiency, $ξ_{\rm ion}$, and galaxy overdensity, $\log(1+δ)$. We analyze the ionizing properties of 79 galaxies between $1.0 < z < 5.2$ using JWST NIRSpec medium-resolution spectra from the Systematic Mid-infrared Instrument (MIRI) Legacy Extragalactic Survey (SMILES) program. Among these, 67 galaxies have H$α$ coverage, spanning $1.0 < z < 3.1$. The galaxy overdensity, $\log(1+δ)$, is measured using the JADES photometric catalog, which covers the SMILES footprint. For the subset with H$α$ coverage, we find that $\logξ_{\rm ion}$ is positively correlated with $\log(1+δ)$, with a slope of $0.94_{-0.46}^{+0.46}$. Additionally, the mean $ξ_{\rm ion}$ for galaxies in overdense regions ($\log(1+δ) > 0.1$) is 2.43 times that of galaxies in lower density regions ($\log(1+δ) < 0.1$). This strong trend is found to be independent of redshift evolution. Furthermore, our results confirm the robust correlations between $ξ_{\rm ion}$ and the rest-frame equivalent widths of the [O III] or H$α$ emission lines. Our results suggest that galaxies in high-density regions are efficient producers of ionizing photons.

astro-ph.GA↗

arXiv2025

Generative Physical AI in Vision: A Survey

Generative Artificial Intelligence (AI) has rapidly advanced the field of computer vision by enabling machines to create and interpret visual data with unprecedented sophistication. This transformation builds upon a foundation of generative models to produce realistic images, videos, and 3D/4D content. Conventional generative models primarily focus on visual fidelity while often neglecting the physical plausibility of the generated content. This gap limits their effectiveness in applications that require adherence to real-world physical laws, such as robotics, autonomous systems, and scientific simulations. As generative models evolve to increasingly integrate physical realism and dynamic simulation, their potential to function as "world simulators" expands. Therefore, the field of physics-aware generation in computer vision is rapidly growing, calling for a comprehensive survey to provide a structured analysis of current efforts. To serve this purpose, the survey presents a systematic review, categorizing methods based on how they incorporate physical knowledge, either through explicit simulation or implicit learning. It also analyzes key paradigms, discusses evaluation protocols, and identifies future research directions. By offering a comprehensive overview, this survey aims to help future developments in physically grounded generation for computer vision. The reviewed papers are summarized at https://tinyurl.com/Physics-Aware-Generation.

cs.CV↗