A framework to explain decoder-only sequence classification models using intermediate predictions.
problem Explaining predictions of decoder-only sequence classification models.
method Progressive Inference framework with Single Pass-Progressive Inference and Multi Pass-Progressive Inference methods.
result Significantly better attributions compared to prior work on text classification tasks.
Study estimates heterogeneous principal causal effects with binary treatments and intermediate variables.
problem Estimating subgroup effects within strata defined by potential values of an intermediate variable.
method Proposes a framework for estimating and forming confidence intervals for heterogeneous principal causal effects under principal ignorability assumption. Develops several estimators with varying robustness properties.
result Established large-sample theory and analyzed bias contributions of each approach.
We study the geometry of multidimensional scalar 2 n d 2^{nd} 2 n d order PDEs (i.e. PDEs with n n n independent variables) with one unknown function, viewed as hypersurfaces E \mathcal{E} E in the Lagrangian Grassmann bundle M ( 1 ) M^{(1)} M ( 1 ) over a ( 2 n + 1 ) (2n+1) ( 2 n + 1 ) -dimensional contact manifold ( M , C ) (M,\mathcal{C}) ( M , C ) . We develop the theory of character…
Neural network models have a reputation for being black boxes. We propose to monitor the features at every layer of a model and measure how suitable they are for classification. We use linear classifiers, which we refer to as "probes", trained entirely independently of the model itself. This helps us better understand …
New algorithm tackles non-stationary delayed feedback in recommender systems.
problem Challenges in learning from delayed feedback in non-stationary environments.
method Developed a UCRL-based algorithm for non-stationary, delayed bandits with intermediate observations.
result Sublinear regret guarantees for the proposed algorithm in non-stationary delayed environments.
This paper generalizes the envelope of mid-lines to intermediate lines for a plane curve.
problem Understanding the envelope of intermediate lines for a plane curve.
method Using singularity theory techniques to analyze the local behavior of the envelope of intermediate lines.
result The envelope of intermediate lines ( E I L EIL E I L ) is formed by three disconnected sets: AEIL, the curve itself, and IPTL. Study on rigidity with non-negative intermediate curvature on low-dimensional manifolds.
problem Extending non-existence theorem of positive scalar curvature to product manifolds.
method Introduced intermediate curvature and studied rigidity conditions.
result Rigidity when intermediate curvature is non-negative in low dimensions.
Preserves positive intermediate curvature on manifolds.
problem Obstructs positive intermediate curvature on partial tori.
method Shows smooth interpolation of metrics with positive intermediate curvature.
result Proves non-existence of certain manifolds with positive intermediate curvature.
Improved PQM for pattern classification on quantum computers.
problem Pattern classification in quantum computing.
method Parametric Probabilistic Quantum Memory (PQM) with quantum circuit.
result Classical evaluation and quantum experiments validate PQM viability.
This paper analyzes microstructure dynamics in coupled markets using CFMMs.
problem Quantifying contributions of CFMMs to market dynamics in coupled markets.
method Examined constant function market makers (CFMMs) in coupled markets, focusing on basket inflation/deflation.
result CFMMs contribute significantly to basket inflation/deflation in coupled markets.
The study connects manifold topology to metrics with positive intermediate curvature.
problem Understanding the relationship between manifold topology and metrics with positive intermediate curvature.
method Formulated a conjecture and proved it for specific dimensions and conditions.
result Closed, aspherical 6-manifolds cannot admit metrics with positive 4-intermediate curvature.
Workshop reviews techniques to understand neural NLP models.
problem Understanding the inner workings of neural networks in natural language processing.
method Systematic manipulation of inputs, decoding intermediate representations, modifying architectures, and testing on simplified languages.
result Various techniques can improve explainability of neural network models.
New rigidity results for manifolds with maximal symmetry rank and positive intermediate Ricci curvature.
problem Understanding the structure of manifolds with maximal symmetry rank and positive intermediate Ricci curvature.
method Recovering stronger topological rigidity results using higher intermediate Ricci curvatures and nontrivial fundamental groups.
result Stronger topological rigidity results for manifolds with maximal symmetry rank and positive intermediate Ricci curvature.
Study on spaces of metrics with intermediate curvature bounds.
problem Understanding spaces of metrics with lower bounds on intermediate curvatures.
method Analyzing spaces of Riemannian metrics with specific curvature bounds on high-dimensional Spin-manifolds.
result Spaces of metrics with positive p-curvature and k-positive Ricci curvature have non-trivial homotopy groups.
Proves metrics with positive intermediate Ricci curvature on complex manifolds.
problem Establishing metrics with positive intermediate Ricci curvature on complex manifolds.
method Canonical variation and surgery techniques.
result Existence of metrics with positive intermediate Ricci curvature on various examples.
Extends Perelman's theorem to positive intermediate curvature conditions.
problem Positive intermediate curvature conditions and their implications.
method Generalization of Perelman's gluing theorem to positive intermediate curvature conditions.
result Observer moduli space can have non-trivial higher homotopy groups.
The study proves that certain manifolds with boundary cannot have metrics with positive intermediate curvatures.
problem Proving the nonexistence of metrics with positive intermediate curvatures on manifolds with boundary.
method Curvature obstruction theorems for manifolds with boundary.
result Topologically nontrivial compact manifolds with boundary cannot have metrics of positive m m m -intermediate curvature if the boundary is m m m -convex. LIT compresses deep networks by training intermediate representations, outperforming traditional methods.
problem Reducing the computational overhead of deep network inference.
method LIT trains a student model with the same width but shallower depth, using the intermediate representations from the teacher model.
result LIT achieves substantial network depth reductions without accuracy loss, outperforming traditional methods.
Countries rarely enter unrelated products, but those that do grow faster.
problem Understanding the benefits of entering unrelated products for economic growth.
method Identifying periods of unrelated product entry through analysis of 93 countries' exports from 1965 to 2014.
result Countries entering more unrelated products experience a small but significant increase in future economic growth.
Study finds metrics with positive intermediate Ricci curvature on specific low-dimensional manifolds.
problem Existence of invariant metrics with positive intermediate Ricci curvature on low-dimensional cohomogeneity one manifolds.
method Construction of invariant metrics with positive intermediate Ricci curvature on specific manifolds.
result Invariant metrics with positive 4th-intermediate Ricci curvature exist but not for 3rd-intermediate Ricci curvature on certain manifolds.
Symmetry rank bound for manifolds with positive intermediate Ricci curvature.
problem Bounding the symmetry rank of manifolds with positive intermediate Ricci curvature.
method Local argument and effective isometric action on manifolds.
result Symmetry rank bound r ≤ ⌊ n + k 2 f l o o r r \leq \lfloor \frac{n+k}{2}
floor r ≤ ⌊ 2 n + k f l oor for positive intermediate Ricci curvature. Sharp dimension constraints for positive intermediate curvature metrics are established.
problem Proving sharp dimension constraints for metrics with positive intermediate curvature.
method Constructing counterexamples and extending rigidity results.
result Sharp dimension constraints for positive intermediate curvature metrics are established.
This paper introduces a novel approach to measuring privacy risks in deep computer vision models based on intermediate outputs.
problem The exposure of intermediate results in hidden layers of deep computer vision models poses significant privacy concerns.
method The approach leverages Degrees of Freedom (DoF) to evaluate the amount of information retained in each layer and combines this with the rank of the Jacobian matrix to assess sensitivity to input variations.
result The proposed framework provides deeper insights into privacy risks associated with intermediate representations without requiring adversarial attack simulations.
The paper proves manifold splitting theorems with nonnegative intermediate curvature.
problem Proving rigidity results for manifolds with nonnegative intermediate curvatures.
method New recursion theorem for spectral intermediate curvatures and cylindrical splitting theorems.
result Smooth metrics with uniformly positive intermediate curvature constructed.
Polyhedral semantics for intermediate logics; Nerve Criterion ensures completeness.
problem Characterize polyhedrally-complete intermediate logics.
method Developed Nerve Criterion to characterize polyhedrally-complete logics combinatorially.
result Nerve Criterion provides a necessary and sufficient condition for polyhedrally-completeness.
Paper proposes an AutoML framework for efficient device-edge co-inference.
problem Finding optimal hyper-parameters for model sparsity and feature compression.
method Sequential decision problem solved using deep reinforcement learning (DRL).
result Achieves better communication-computation trade-off and significant speedup.
We give a homological construction of aperiodic tiles for certain open Riemannian surfaces admitting actions of Grigorchuk groups of intermediate growth.
Optimizes real estate prices with dynamic strategies.
problem Optimizing prices for limited real estate goods over time.
method Develops a mathematical model considering variable demand, time value, and growth of real estate value.
result Enhanced model for better revenue management in real estate.
A cost-effective framework for gradual domain adaptation using multifidelity.
problem Degrading prediction performance due to large domain distance.
method Combines multifidelity and active domain adaptation.
result Improves prediction performance with reduced sample cost.
Gradual domain adaptation improves model transfer between domains with intermediate training.
problem Challenges in unsupervised domain adaptation when distribution shifts are large.
method Gradual self-training using intermediate domains along the Wasserstein geodesic.
result GOAT framework generates intermediate domains for improved adaptation.
Study shows Gromov's Betti number bound fails for certain intermediate Ricci curvatures.
problem Gromov's Betti number bound for sectional curvature bounded below does not hold for intermediate Ricci curvatures.
method Established a surgery result for Riemannian metrics with R i c k > 0 Ric_k>0 R i c k > 0 and showed failure of Gromov's bound for specific ranges of k k k . result Gromov's Betti number bound fails for R i c k > 0 Ric_k>0 R i c k > 0 when ⌊ n / 2 f l o o r + 2 ≤ k ≤ n − 1 \lfloor n/2
floor+2 \le k \le n-1 ⌊ n /2 f l oor + 2 ≤ k ≤ n − 1 . This paper develops fast rates for ERM and SA algorithms under error bound conditions.
problem Developing fast and adaptive optimization algorithms for statistical learning.
method Empirical Risk Minimization (ERM) and Stochastic Approximation (SA) algorithms with fast convergence rates under error bound conditions.
result Fast and adaptive convergence rates for ERM and SA algorithms, spanning from O ( 1 / n ) O(1/\sqrt{n}) O ( 1/ n ) to O ( 1 / n ) O(1/n) O ( 1/ n ) , depending on error bound conditions. Study analyzes financial intermediation costs in decentralized lending protocols.
problem Understanding the cost of financial intermediation in decentralized lending protocols.
method Analysis of publicly available data on rates, supply, borrow activity, and accounts.
result Ex-post margins are 1% and lower for stablecoin markets.
Improved U-Nets with various intermediate blocks enhance singing voice separation.
problem Improving singing voice separation accuracy using U-Net architectures.
method Implemented and compared U-Nets with different intermediate spectrogram transformation blocks.
result A specific block type achieves state-of-the-art SDR by 0.9 dB.
Intermediate logic of all convex polyhedra is axiomatized.
problem Defining and axiomatizing intermediate logic for convex polyhedra.
method Using Jankov-Fine formulas, classical polyhedral geometry, and p-morphic images to establish completeness.
result A finite axiomatisation of PL for all convex polyhedra.
Log-Sobolev inequality proven for submanifolds in specific types of manifolds.
problem Proving Log-Sobolev inequality for submanifolds in asymptotic non-negative intermediate Ricci curvature manifolds.
method Extending previous results, proving inequality for submanifolds in specific types of manifolds.
result Sharp Log-Sobolev inequality proven for submanifolds in complete non-compact Riemannian manifolds with asymptotic non-negative intermediate Ricci curvature and Euclidean volume growth.
Optimizes transport on submanifolds for curvature inequalities.
problem Proving Michael-Simon-Sobolev inequalities in manifolds with intermediate Ricci curvature bounds.
method Generalizes optimal transport theory to submanifolds and applies to curvature inequalities.
result Proves a variant of the Michael-Simon-Sobolev inequality in manifolds with nonnegative intermediate Ricci curvatures.
Extends metric properties over surgeries to higher codimensions.
problem Extending metrics of positive scalar curvature over surgeries in higher codimensions.
method Generalizes Gromov-Lawson, Schoen-Yau, and Walsh's construction to ( p , n ) (p,n) ( p , n ) -intermediate scalar curvature. result Shows infinitely many path components for certain metrics on ( 4 n − 1 ) (4n-1) ( 4 n − 1 ) -manifolds. Proves positive Euler characteristic for certain manifolds with positive second intermediate Ricci curvature.
problem Hopf's conjecture on positive sectional curvature and its failure under relaxed conditions.
method Non-trivial extension of the Four Periodicity Theorem to higher degrees.
result Proves positive Euler characteristic for specific manifolds with positive second intermediate Ricci curvature.
Study improves understanding of Ricci curvature in manifolds.
problem Understanding Ricci curvature in manifolds with specific assumptions.
method Exploring m-intermediate Ricci curvature and proving comparison theorems.
result Stable weighted slicing in manifolds with non-negative m-intermediate Ricci curvature has almost non-negative Ricci curvature.
We compare systematically several classes of stochastic volatility models of stock market fluctuations. We show that the long-time return distribution is either Gaussian or develops a power-law tail, while the short-time return distribution has generically a stretched-exponential form, but can assume also an algebraic …
Study connects Chern class components to Kato surface germs.
problem Understanding intermediate Kato surfaces through Chern class components.
method Relates coefficients of Chern classes to birational germs.
result Established relations between Chern class components and Kato surface germs.
A new approach for pricing FX options that uses a single model for all markets.
problem Consistent pricing of FX options across different markets.
method Intermediate currency approach, calibrating to domestic market volatility smile.
result Model automatically reproduces correct foreign market volatility smiles.
Study finds almost contact structures in thermal QCD-like theories at intermediate coupling.
problem Understanding (Almost) Contact Structures in thermal QCD-like theories.
method Explicitly obtained (Almost) Contact Structures and SU(3) structures.
result Subspaces of C3S and AC3S are not mutually 'N-path connected' in the Infra-Red.
Study manifolds with positive intermediate Ricci curvature and large symmetry rank.
problem Closed, simply connected manifolds with positive 2nd-intermediate Ricci curvature and large symmetry rank.
method New tools for studying isometric actions on closed manifolds with positive kth-intermediate Ricci curvature, including isotropy rank lemma, symmetry rank bound, and connectedness principle.
result Even dimensional manifolds with large symmetry rank must have trivial odd degree integral cohomology and are either spheres or complex projective spaces.
New algorithms reduce communication costs in collaborative learning.
problem Reducing communication costs in collaborative learning.
method Distributed boosting and adaptation to classification noise.
result Communication-efficient algorithms for collaborative PAC learning robust to noise.
Establishes metrics with positive 2nd intermediate Ricci curvature on products of curved spaces.
problem Examines the limitations of positive curvature metrics on product spaces.
method Uses examples to show R i c 2 > 0 \mathrm{Ric}_2>0 Ric 2 > 0 does not imply positive curvature for products of spaces. result The R i c 2 > 0 \mathrm{Ric}_2>0 Ric 2 > 0 class of manifolds is distinct from positively curved manifolds. MLShrink integrates machine learning with wavelet shrinkage for denoising.
problem Denoising signals with uncertain magnitudes
method Combines wavelet shrinkage with machine learning
result Preserves simplicity for signal coefficients while allowing data-adaptive decisions for ambiguous coefficients