Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

4.2%8.3%12.5%16.7% · Oct 199519922001200920172026
48 results for phase collapse

This paper explains a mechanism called phase collapse that improves image classification accuracy.

problem Understanding the role of non-linearities and convolutional filters in image classification.
method Demonstrates phase collapse as a mechanism that eliminates spatial variability and linearly separates classes.
result Phase collapse improves classification accuracy, while thresholding operators degrade performance.

Deep nets exhibit 'Neural Collapse' during training's final phase, simplifying decision-making.

problem Understanding and optimizing deep learning training phases.
method Direct measurements on three deepnet architectures across seven datasets.
result Deep nets exhibit 'Neural Collapse' during training's final phase, simplifying decision-making.

Novel anti-grokking phase discovered in neural networks, revealed by HTSR layer quality metric.

problem Understanding and detecting overfitting in neural networks.
method 3-layer MLP, weight decay, HTSR layer quality metric αα, correlation traps, Kolmogorov–Smirnov tests.
result Anti-grokking phase detected late in training, revealed by α<2α< 2 and correlation traps.

The paper analyzes the dynamics of tokens in transformer models at moderate interaction levels.

problem Understanding the evolution of tokens in transformer models at moderate interaction levels.
method Modeling transformer models as a system of particles interacting in a mean-field way and studying the corresponding dynamics.
result Characterization and convergence of the limiting dynamics in different phases of the system.

Trust is a collective, self-fulfilling phenomenon that suggests analogies with phase transitions. We introduce a stylized model for the build-up and collapse of trust in networks, which generically displays a first order transition. The basic assumption of our model is that whereas trust begets trust, panic also begets…

2014-09-22abs ↗pdf ↗

Paper explains neural collapse in neural networks using a new model.

problem Understanding neural collapse in neural networks during training.
method Introducing the unconstrained layer-peeled model (ULPM) to prove gradient flow convergence to critical points of a minimum-norm separation problem.
result Proves that all critical points are strict saddle points except the global minimizers exhibiting neural collapse.

Reward collapse occurs when ranking-based reward models yield uniform rewards for different prompts.

problem Reward collapse in aligning large language models with human preferences.
method Introduced a prompt-aware optimization scheme to derive closed-form expressions for reward distributions.
result Our prompt-aware utility functions significantly alleviate reward collapse during training.

Generative diffusion models gradually memorize training data, losing independent dimensions.

problem Understanding how generative diffusion models memorize training data, especially on low-dimensional manifolds.
method Measuring latent dimensionality via the learned score field, proposing a geometric memorization theory.
result Generative diffusion models experience a smooth collapse of their capacity to vary across independent directions as data become scarce, leading to near point-wise replication of salient features.

Study shows how anisotropic data affects learning dynamics in phase retrieval.

problem Understanding learning dynamics in phase retrieval with anisotropic Gaussian inputs.
method Developed a tractable reduction to reveal a three-phase trajectory and derived scaling laws.
result Found that anisotropy leads to a three-phase trajectory: fast escape, slow convergence, and spectral-tail learning.

An elementary family of local Hamiltonians H,¸,=1,2,3,ldotsH_{\c ,\ell}, \ell = 1,2,3, ldots, is described for a 22-dimensional quantum mechanical system of spin =1/2={1/2} particles. On the torus, the ground state space G,G_{\circ,\ell} is (log)(\log) extensively degenerate but should collapse under łłperturbation" to an anyonic syste…

2001-10-09abs ↗pdf ↗

Improved RL training for DMs reduces mode collapse and preserves diversity.

problem Mode collapse and training instability in RL fine-tuned diffusion models.
method Dynamic hierarchical RL training with sliding-window parameter regularisation.
result Models trained with HRF achieve better preservation of diversity in downstream tasks.

Modeling how network connectivity affects economic collapse and robustness.

problem Impact of network topology on systemic risk and collapse of complex economic systems.
method Proposed a model to study the effects of network structure on economic systems by varying connectivity.
result Emergent systemic risks arise with increased interconnections, leading to phase transitions and tipping points.

This work investigates how neural collapse improves transfer learning for large-scale models.

problem Improving transfer learning for large-scale models with limited labeled data.
method Investigates neural collapse and develops a fine-tuning method using skip-connections.
result Feature collapse on downstream data correlates with higher transfer accuracy.

A new approach uses circuit topology to study complex polymer interactions.

problem Understanding structural phase transitions in entangled polymer systems.
method Braided circuit topology framework for multiple-chain systems.
result Circuit topological motif fractions are effective order parameters for structural transitions.

This paper extends neural collapse to imbalanced data under cross-entropy loss.

problem Analyzing neural collapse in deep networks with imbalanced data.
method Using the unconstrained feature model and cross-entropy loss, the paper studies neural collapse in imbalanced datasets.
result Feature vectors within the same class collapse to a single mean vector, but angles between them depend on sample size.

This paper extends neural collapse to class-imbalanced datasets using an unconstrained ReLU feature model.

problem Understanding neural collapse in class-imbalanced datasets with cross-entropy loss.
method Generalized neural collapse to class-imbalanced settings using an unconstrained ReLU feature model.
result Class-means converge to orthogonal vectors with different lengths, and classifier weights align to these vectors.

We analyze neural collapse in neural networks, showing that features collapse to vertices of a Simplex ETF.

problem Understanding and optimizing the features learned in the last layer of neural networks during training.
method Simplified unconstrained feature model, studying the global optimization landscape of cross-entropy loss with weight decay.
result The global minimizers of the loss are Simplex ETFs, and other critical points are strict saddles with negative curvature.

Persistent entropy detects phase transitions in complex systems.

problem Detecting phase transitions in complex systems.
method Established a general theorem for persistent entropy to reliably detect phase transitions, introduced operational framework for finite-time computations.
result Persistent entropy exhibits an asymptotically non-vanishing gap across phases, robust numerical signatures across experiments.

High-dimensional models become unstable when sample size falls below a critical level, leading to a phase transition.

problem Instability in high-dimensional learning models when sample size is insufficient.
method Proved the necessity of a Fisher eigenvalue threshold for stability, introduced Fisher floor for verification.
result A sharp phase transition between reliable concentration and inevitable failure in high-dimensional learning.

Study on line bundle flow on Kähler surfaces converging to a singular solution.

problem Analyzing the mean curvature flow on Kähler surfaces.
method Investigates the flow under hypercritical phase and semipositivity conditions.
result The flow converges to a singular solution away from curves of negative self-intersection.

For surfaces, we brush a reasonably sharp picture of the influence of the fundamental group upon the complexity of foliated-dynamics. A metaphor emerges with phase-changes through the solid-liquid-gaseous states. Groups of ranks 0r10\le r\le 1 are frozen with intransitivity reigning ubiquitously. When 2r32\le r \le 3, th…

2011-11-24abs ↗pdf ↗

New estimates quantify blow-up rates of spacelike singularities in gravitational collapse.

problem Quantifying the blow-up rates of spacelike singularities in gravitational collapse.
method Deriving new quantitative estimates using spherical symmetry and double-null coordinates.
result Polynomial blow-up rates O(1/rN)O(1/r^N) for various quantities, with improved estimates for rurr\partial_u r and rvrr\partial_v r.

Deep neural networks near edge of chaos show universal scaling laws.

problem Understanding the behavior of deep neural networks near critical points.
method Analogy to absorbing phase transitions in statistical mechanics, deterministic propagation dynamics, mean-field and directed percolation universality classes.
result Deep neural networks exhibit universal scaling laws near the edge of chaos.

This work explains neural collapse in shallow neural networks and its impact on generalization.

problem Understanding neural collapse in shallow neural networks and its effect on generalization.
method Analysis of two and three-layer ReLU neural networks, focusing on data dimension, sample size, and signal-to-noise ratio.
result Neural collapse occurs in shallow ReLU networks under certain conditions related to data properties and network architecture.

Theoretical study shows AI models can recover from contaminated training data.

problem Data contamination in AI training can degrade model performance.
method Theoretical analysis and experiments on various data types.
result Models converge to true distribution under mild conditions, with rate dependent on real data fraction.

The aim of this work is to explore the possible types of phenomena that simple macroeconomic Agent-Based models (ABM) can reproduce. We propose a methodology, inspired by statistical physics, that characterizes a model through its 'phase diagram' in the space of parameters. Our first motivation is to understand the lar…

2013-07-11abs ↗pdf ↗

Ancient solutions found on flag manifolds from invariant Einstein metrics.

problem Understanding the behavior of Ricci flow on flag manifolds.
method Global study of the dynamical system induced by the Ricci flow, using invariant Einstein metrics and Poincaré compactification.
result Non-collapsed ancient solutions emerge from invariant Einstein metrics, with a Type I singularity in finite time.

Proposes using continuum percolation to analyze data manifolds and improve generative models.

problem Disentangling geometric support from probability distributions in high-dimensional data.
method Establishes a correspondence between topological phase transitions of random geometric graphs and data manifolds, using Percolation Shift metric.
result Demonstrates that Percolation Shift metric captures structural pathologies like mode collapse and guides training to prevent manifold shrinkage and improve fidelity.

Bayesian deep learning faces posterior collapse due to likelihood vs. prior competition.

problem Posterior collapse in Bayesian deep learning models.
method Identified competition between likelihood and prior regularization in a linear latent variable model.
result Posterior collapse is related to neural and dimensional collapse, suggesting a broader learning issue.

Study on Neural Collapse limits in deep learning.

problem Understanding the limits of Neural Collapse in deep learning.
method Investigated Neural Collapse in the context of generalization and feature learning, refining conjectures and conducting experiments.
result Neural Collapse primarily occurs on the train set and not on the test set, suggesting it is an optimization phenomenon with unclear connections to generalization.

Novel Ricci flow normalization for homogeneous spaces, focusing on flag manifolds.

problem Understanding the limiting behavior and symmetry properties of Ricci flow on homogeneous spaces.
method Introducing a novel normalization for the homogeneous Ricci flow and characterizing Gromov-Hausdorff limits.
result Full classification of Gromov-Hausdorff limits and detailed phase portraits for three-isotropy-summands flag manifolds.

Mathematical analysis shows annealing prevents mode collapse in Gaussian mixtures.

problem Mode collapse in variational inference for multimodal distributions.
method Analyzed annealing strategies for Gaussian mixtures, derived formulas, and tested on neural networks.
result Appropriately chosen annealing schemes can robustly prevent mode collapse.

Despite excellent progress in recent years, mode collapse remains a major unsolved problem in generative adversarial networks (GANs).In this paper, we present spectral regularization for GANs (SR-GANs), a new and robust method for combating the mode collapse problem in GANs. Theoretical analysis shows that the optimal …

2019-08-29abs ↗pdf ↗

New method controls posterior collapse in VAEs without network architecture constraints.

problem Posterior collapse in VAEs reduces diversity of generated samples.
method Introduces Latent Reconstruction (LR) loss to control posterior collapse.
result Controls posterior collapse on various datasets without architectural constraints.

Collapsibility is a combinatorial strengthening of contractibility. We relate this property to metric geometry by proving the collapsibility of any complex that is CAT(0) with a metric for which all vertex stars are convex. This strengthens and generalizes a result by Crowley. Further consequences of our work are: (1) …

2011-07-28abs ↗pdf ↗