Related papers: Block-Wise Differentiable Sinkhorn Attention: Tail…
Entropic optimal transport (EOT) via Sinkhorn iterations is widely used in modern machine learning, yet GPU solvers remain inefficient at scale. Tensorized implementations suffer quadratic HBM traffic from dense $n\times m$ interactions,…
Optimal transport (OT) is a widely used tool in machine learning, but computing high-accuracy solutions for large instances remains costly. Entropic regularization and the Sinkhorn algorithm improve scalability; however, when the…
The recent emergence of new algorithms for permuting models into functionally equivalent regions of the solution space has shed some light on the complexity of error surfaces, and some promising properties like mode connectivity. However,…
Selective State Space Models (SSMs) achieve linear-time inference, yet their gradient-based sensitivity analysis remains bottlenecked by O(L) memory scaling during backpropagation. This memory constraint precludes genomic-scale modeling (L…
To help address the growing demand for ever-longer sequence lengths in transformer models, Liu et al. recently proposed Ring Attention, an exact attention algorithm capable of overcoming per-device memory bottle- necks by distributing…
Entropic regularization provides a simple way to approximate linear programs whose constraints split into two or more tractable blocks. The resulting objectives are amenable to cyclic Kullback-Leibler (KL) Bregman projections, with…
Sinkhorn's algorithm is a method of choice to solve large-scale optimal transport (OT) problems. In this context, it involves an inverse temperature parameter $\beta$ that determines the speed-accuracy trade-off. To improve this trade-off,…
Efficient inference on GPUs using large language models remains challenging due to memory bandwidth limitations, particularly during data transfers between High Bandwidth Memory (HBM) and SRAM in attention computations. Approximate…
The self-attention mechanism is central to the success of Transformer architectures. However, standard row-stochastic attention has been shown to suffer from significant signal degradation across layers. In particular, it can induce rank…
Deployment of Transformer models on edge devices is becoming increasingly challenging due to the exponentially growing inference cost that scales quadratically with the number of tokens in the input sequence. Token pruning is an emerging…
Transformers usually expose one inference cost per trained model, while deployed systems often need multiple cost-quality operating points. We study Budgeted Attention Allocation, a monotone head-gating mechanism conditioned on a requested…
Standard transformer architectures apply a single attention mechanism uniformly across all tokens and sequence positions, irrespective of local context or computational budget. We propose Meta-Attention, a framework that dynamically routes…
Transformer attention computes a single softmax-weighted average over values -- a one-pass estimate that cannot correct its own errors. We introduce \emph{gradient-boosted attention}, which applies the principle of gradient boosting…
We tackle high-dimensional, path-dependent valuation and control and introduce a deep BSDE/2BSDE solver that couples truncated log-signatures with a neural rough differential equation (RDE) backbone. The architecture aligns stochastic…
Entropic regularized optimal transport (OT) via the Sinkhorn algorithm has become a fundamental tool in machine learning, yet existing implementations either suffer from numerical instability for small regularization parameters or incur…
The transition from fitting empirical data to achieving true human utility is fundamentally constrained by a granularity mismatch, where fine-grained autoregressive generation is often supervised by coarse or uniform signals. This position…
Differentiable logic gate networks (DLGNs) learn compact, interpretable Boolean circuits via gradient-based training, but all existing variants are restricted to the 16 two-input binary gates. Extending DLGNs to Ternary Kleene $K_3$ logic…
We propose Sparse Sinkhorn Attention, a new efficient and sparse method for learning to attend. Our method is based on differentiable sorting of internal representations. Concretely, we introduce a meta sorting network that learns to…
Graph transformers achieve strong results on molecular and long-range reasoning tasks, yet remain hampered by over-smoothing (the progressive collapse of node representations with depth) and attention entropy degeneration. We observe that…
Solutions to the Schr\"{o}dinger bridge problem and its generalizations yield feedback control policies for optimal density steering over a controlled diffusion. To numerically compute the same, the dynamic Sinkhorn recursion has become a…