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.
Language Rectified Flow improves diffusion language generation by simplifying complex steps.
Discrete Flow Maps bypass sequential prediction limits for parallel text generation.
While normalizing flows have led to significant advances in modeling high-dimensional continuous distributions, their applicability to discrete distributions remains unknown. In this paper, we show that flows can in fact be extended to discrete events---and under a simple change-of-variables formula not requiring log-d…
Improved language generation with faster sampling speed.
FlowLLM uses LLMs and flow matching to efficiently generate novel materials.
MarketGPT models financial time series with realistic order flow data.
Unified framework reduces NFEs for inverse problems.
Branching Flows generates sequences of varying lengths using binary trees.
A new sampler for FLMs improves token-level decoding controls.
We interpret the physical -field renormalization group flow in the language of Courant algebroids, clarifying the sense in which this flow is the natural "Ricci flow" for generalized geometry. Next we show that the -field renormalization group flow preserves T-duality in a natural sense. As corollaries we obtain …
Paper introduces new methods for modeling categorical data.
Normalizing flows are a powerful class of generative models for continuous random variables, showing both strong model flexibility and the potential for non-autoregressive generation. These benefits are also desired when modeling discrete random variables such as text, but directly applying normalizing flows to discret…
We recast the Calabi flow in DeGiorgi's language of minimizing movements. We establish the long time existence of minimizing movements for K-energy with arbitrary initial condition. Furthermore we establish some a priori regularity of these solutions, and that sufficiently regular minimizing movements are smooth soluti…
We propose a new statistical model suitable for machine learning of systems with long distance correlations such as natural languages. The model is based on directed acyclic graph decorated by multi-linear tensor maps in the vertices and vector spaces in the edges, called tensor network. Such tensor networks have been …
We introduce and demonstrate a new approach to inference in expressive probabilistic programming languages based on particle Markov chain Monte Carlo. Our approach is simple to implement and easy to parallelize. It applies to Turing-complete probabilistic programming languages and supports accurate inference in models …
In this paper, we show the uniqueness of Schrödinger flow from a general complete Riemannian manifold to a complete Kähler manifold with bounded geometry. While following the ideas of McGahagan[16], we present a more intrinsic proof by using the distance functions and gauge language.
Unified framework for continuous-state discrete flow matching models.
Flow-based models use ODEs to generate complex data distributions.
Stan is a probabilistic programming language that has been increasingly used for real-world scalable projects. However, to make practical inference possible, the language sacrifices some of its usability by adopting a block syntax, which lacks compositionality and flexible user-defined functions. Moreover, the semantic…
Program comprehension is a fundamental task in software development and maintenance processes. Software developers often need to understand a large amount of existing code before they can develop new features or fix bugs in existing programs. Being able to process programming language code automatically and provide sum…
A method for eliciting expert beliefs using preferential questions and normalizing flows.
In dimension , there is a complete theory of weak solutions of Ricci flow - the singular Ricci flows introduced by Kleiner and Lott - which are unique across singularities, as was proved by Bamler and Kleiner. We show that uniqueness should not be expected to hold for Ricci flow weak solutions in dimensions $n\geq…
Study reveals structure of Bitcoin's crypto flow network.
The goal of program synthesis is to automatically generate programs in a particular language from corresponding specifications, e.g. input-output behavior. Many current approaches achieve impressive results after training on randomly generated I/O examples in limited domain-specific languages (DSLs), as with string tra…
Traffic flow forecasting is hot spot research of intelligent traffic system construction. The existing traffic flow prediction methods have problems such as poor stability, high data requirements, or poor adaptability. In this paper, we define the traffic data time singularity ratio in the dropout module and propose a …
Attention mechanism combines bottom-up and top-down signals in neural networks.
Paper proposes a faster, higher-quality RL method for text-to-image models.
Generative models tackle incompressible fluid flows by enforcing divergence-free constraints.
We prove that the Kontsevich tetrahedral flow , the right-hand side of which is a linear combination of two differential monomials of degree four in a bi-vector on an affine real Poisson manifold , does infinitesimally preserve the space of Poisson…
With the recent success of embeddings in natural language processing, research has been conducted into applying similar methods to code analysis. Most works attempt to process the code directly or use a syntactic tree representation, treating it like sentences written in a natural language. However, none of the existin…
Hybrid model improves traffic flow prediction accuracy.
Physics-informed model reduces RBC simulation costs.
Unified treatment of eigenvalue processes using Riemannian geometry.
Generative model predicts financial market order flow with high accuracy.
Generative models are reinterpreted in statistical terms, enabling better understanding and inference.
Study quantifies gender bias in language models across 7 languages.
Predict stock price movements using financial data and news articles with LLMs.
New ASR system handles multiple languages without needing language-specific encoding.
As deep neural networks become more adept at traditional tasks, many of the most exciting new challenges concern multimodality---observations that combine diverse types, such as image and text. In this paper, we introduce a family of multimodal deep generative models derived from variational bounds on the evidence (dat…
Recurrent Networks are one of the most powerful and promising artificial neural network algorithms to processing the sequential data such as natural languages, sound, time series data. Unlike traditional feed-forward network, Recurrent Network has a inherent feed back loop that allows to store the temporal context info…
New technique reduces language biases in large language models.
In this work we present a modification in the conventional flow of information through a LSTM network, which we consider well suited for RNNs in general. The modification leads to a iterative scheme where the computations performed by the LSTM cell are repeated over a constant input and cell state values, while updatin…
PHINN: A generative model for rare-event time series using persistent homology
FinBloom enhances LLMs for real-time financial queries.
The paper applies math and physics to language models, introducing entropy and geometric concepts.
This study improves stock price prediction using multimodal data.
Paper introduces TrufLL for language model training without labeled data.
Multilingual end-to-end (E2E) models have shown great promise in expansion of automatic speech recognition (ASR) coverage of the world's languages. They have shown improvement over monolingual systems, and have simplified training and serving by eliminating language-specific acoustic, pronunciation, and language models…