Joint KL yields horizon-free approximation but an information-theoretic lower bound of order Omega(H) for estimation error in autoregressive learning, with matching computationally efficient upper bounds.
Canonical reference
Title resolution pending
Canonical reference. 71% of citing Pith papers cite this work as background.
163
Pith papers citing it
Background
71% of classified citations
citation-role summary
background 24
method 5
baseline 2
citation-polarity summary
representative citing papers
WildChat releases a dataset of 1 million ChatGPT conversations with timestamps, demographics, and headers, claimed to be the most diverse and multilingual such resource available.
GPT-2 small solves indirect object identification via a circuit of 26 attention heads organized into seven functional classes discovered through causal interventions.
FlexiSLM is the first spoken language model supporting dynamic and controllable frame rates on speech input and output, outperforming fixed-rate 7B models at high quality and enabling faster inference at lower rates like 6.25 Hz.
MediEncoder jointly learns nonlinear low-dimensional covariate and mediator representations via a coupled encoder-decoder with cross-factor network, then applies them in an efficient influence function estimator for natural direct and indirect effects.
LOSCAR-SGD combines local updates, sparse model averaging, and communication-computation overlap with a delay-corrected merge rule, providing convergence rates for smooth non-convex objectives under worker heterogeneity.
Proposes pointwise Riemannian Dimension from feature eigenvalues to derive tighter, representation-aware generalization bounds for deep networks in the nonlinear regime.
Ringmaster LMO extends delay-thresholding from ASGD to LMO-based momentum updates, providing convergence guarantees under (L0, L1)-smoothness and time-complexity bounds that recover optimal rates in the Euclidean case.
Introduces the task of counterfactual time series forecasting with textual conditions plus a text-attribution mechanism that improves accuracy by distinguishing mutable from immutable factors.
M³C replaces the hard hyperparameter optimization with a sequence of simpler problems using a majorant for the log-determinant approximated via Monte Carlo, with proven high-probability convergence to a critical point under assumptions.
TokAlign++ learns token alignments between LLM vocabularies from monolingual representations to enable faster adaptation, better text compression, and effective token-level distillation across 15 languages with minimal steps.
PG-OT builds prompt-specific Pareto frontiers and applies distribution-aware optimal transport to improve multi-reward alignment while introducing JDR and JCR metrics to measure synergy and hacking.
ASAP amortizes Sinkhorn-based doubly-stochastic attention by learning a parametric map from 1D potentials to the Sinkhorn dual and reconstructing the plan via two-sided entropic c-transform, delivering 5.3x faster inference at matched accuracy.
Port-Hamiltonian neural networks extended to PDEs recover the Hamiltonian and dissipation of nonlinear string dynamics from data and outperform non-physics-informed baselines.
CAWI replaces standard random initialization of input-to-hidden weights in randomized neural networks with samples drawn from a data-fitted copula that preserves observed feature dependencies, yielding consistent accuracy gains on 83 classification benchmarks.
Inkheart SGD and M4 use bidirectional compression to achieve time complexities in distributed SGD that improve with worker count n and surpass prior lower bounds under a necessary structural assumption.
TRACE creates valid conformal prediction sets for complex generative models by scoring outputs via averaged denoising or velocity errors along stochastic transport paths instead of likelihoods.
ABGD parametrizes piecewise linear functions as difference of max-affine functions and converges linearly to an epsilon-accurate solution with O(d max(sigma/epsilon,1)^2) samples under sub-Gaussian noise, which is minimax optimal up to logs.
K-DSM uses per-feature kurtosis to set noise scales in DSM, enabling effective single-scale anomaly detection on tabular benchmarks in both semi-supervised and unsupervised settings.
The paper proves statistical consistency of contrastive loss to optimal ranking via an AUC criterion and derives generalization bounds O(1/m + 1/sqrt(n)) for supervised and O(1/sqrt(m) + 1/sqrt(n)) for self-supervised CRL that explain benefits of large negative sets.
LAPRAS uses predictions to answer likely queries with the offline Matrix Mechanism and paces residual budget for unpredicted queries via unbiased stopping-time estimation from the first few unexpected arrivals, achieving near-offline utility when overlap is high.
Adversarial training on simplified Vision Transformers achieves benign overfitting with near-zero robust loss and generalization error when signal-to-noise ratio and perturbation budget meet specific conditions.
Primal-dual policy gradient algorithms achieve global non-asymptotic convergence for safe RLHF cast as infinite-horizon discounted CMDPs without fitting reward models.
FaceParts performs unsupervised segmentation of facial features in Gaussian Splatting avatars and supports precise editing and cross-avatar part transfer using feature disentanglement, density clustering, and FLAME anchoring.
citing papers explorer
-
Autoregressive Learning in Joint KL: Sharp Oracle Bounds and Lower Bounds
Joint KL yields horizon-free approximation but an information-theoretic lower bound of order Omega(H) for estimation error in autoregressive learning, with matching computationally efficient upper bounds.
-
WildChat: 1M ChatGPT Interaction Logs in the Wild
WildChat releases a dataset of 1 million ChatGPT conversations with timestamps, demographics, and headers, claimed to be the most diverse and multilingual such resource available.
-
Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small
GPT-2 small solves indirect object identification via a circuit of 26 attention heads organized into seven functional classes discovered through causal interventions.
-
FlexiSLM: A Dynamic and Controllable Frame Rate Spoken Language Model
FlexiSLM is the first spoken language model supporting dynamic and controllable frame rates on speech input and output, outperforming fixed-rate 7B models at high quality and enabling faster inference at lower rates like 6.25 Hz.
-
MediEncoder: Nonlinear Representation Learning for High-Dimensional Causal Mediation Analysis
MediEncoder jointly learns nonlinear low-dimensional covariate and mediator representations via a coupled encoder-decoder with cross-factor network, then applies them in an efficient influence function estimator for natural direct and indirect effects.
-
LOSCAR-SGD: Local SGD with Communication-Computation Overlap and Delay-Corrected Sparse Model Averaging
LOSCAR-SGD combines local updates, sparse model averaging, and communication-computation overlap with a delay-corrected merge rule, providing convergence rates for smooth non-convex objectives under worker heterogeneity.
-
Pointwise Generalization in Deep Neural Networks
Proposes pointwise Riemannian Dimension from feature eigenvalues to derive tighter, representation-aware generalization bounds for deep networks in the nonlinear regime.
-
Ringmaster LMO: Asynchronous Linear Minimization Oracle Momentum Method
Ringmaster LMO extends delay-thresholding from ASGD to LMO-based momentum updates, providing convergence guarantees under (L0, L1)-smoothness and time-complexity bounds that recover optimal rates in the Euclidean case.
-
What if Tomorrow is the World Cup Final? Counterfactual Time Series Forecasting with Textual Conditions
Introduces the task of counterfactual time series forecasting with textual conditions plus a text-attribution mechanism that improves accuracy by distinguishing mutable from immutable factors.
-
A Majorization-Minimization with Monte Carlo Approach for Hyperparameter Estimation
M³C replaces the hard hyperparameter optimization with a sequence of simpler problems using a majorant for the log-determinant approximated via Monte Carlo, with proven high-probability convergence to a critical point under assumptions.
-
TokAlign++: Advancing Vocabulary Adaptation via Better Token Alignment
TokAlign++ learns token alignments between LLM vocabularies from monolingual representations to enable faster adaptation, better text compression, and effective token-level distillation across 15 languages with minimal steps.
-
Pareto-Guided Optimal Transport for Multi-Reward Alignment
PG-OT builds prompt-specific Pareto frontiers and applies distribution-aware optimal transport to improve multi-reward alignment while introducing JDR and JCR metrics to measure synergy and hacking.
-
ASAP: Amortized Doubly-Stochastic Attention via Sliced Dual Projection
ASAP amortizes Sinkhorn-based doubly-stochastic attention by learning a parametric map from 1D potentials to the Sinkhorn dual and reconstructing the plan via two-sided entropic c-transform, delivering 5.3x faster inference at matched accuracy.
-
Identifying the nonlinear string dynamics with port-Hamiltonian neural networks
Port-Hamiltonian neural networks extended to PDEs recover the Hamiltonian and dissipation of nonlinear string dynamics from data and outperform non-physics-informed baselines.
-
CAWI: Copula-Aligned Weight Initialization for Randomized Neural Networks
CAWI replaces standard random initialization of input-to-hidden weights in randomized neural networks with samples drawn from a data-fitted copula that preserves observed feature dependencies, yielding consistent accuracy gains on 83 classification benchmarks.
-
Scalable Distributed Stochastic Optimization via Bidirectional Compression: Beyond Pessimistic Limits
Inkheart SGD and M4 use bidirectional compression to achieve time complexities in distributed SGD that improve with worker count n and surpass prior lower bounds under a necessary structural assumption.
-
TRACE: Transport Alignment Conformal Prediction via Diffusion and Flow Matching Models
TRACE creates valid conformal prediction sets for complex generative models by scoring outputs via averaged denoising or velocity errors along stochastic transport paths instead of likelihoods.
-
Locally Near Optimal Piecewise Linear Regression in High Dimensions via Difference of Max-Affine Functions
ABGD parametrizes piecewise linear functions as difference of max-affine functions and converges linearly to an epsilon-accurate solution with O(d max(sigma/epsilon,1)^2) samples under sub-Gaussian noise, which is minimax optimal up to logs.
-
Kurtosis-Guided Denoising Score Matching for Tabular Anomaly Detection
K-DSM uses per-feature kurtosis to set noise scales in DSM, enabling effective single-scale anomaly detection on tabular benchmarks in both semi-supervised and unsupervised settings.
-
Statistical Consistency and Generalization of Contrastive Representation Learning
The paper proves statistical consistency of contrastive loss to optimal ranking via an AUC criterion and derives generalization bounds O(1/m + 1/sqrt(n)) for supervised and O(1/sqrt(m) + 1/sqrt(n)) for self-supervised CRL that explain benefits of large negative sets.
-
LAPRAS : Learning-Augmented PRivate Answering for linear query Streams
LAPRAS uses predictions to answer likely queries with the offline Matrix Mechanism and paces residual budget for unpredicted queries via unbiased stopping-time estimation from the first few unexpected arrivals, achieving near-offline utility when overlap is high.
-
Benign Overfitting in Adversarial Training for Vision Transformers
Adversarial training on simplified Vision Transformers achieves benign overfitting with near-zero robust loss and generalization error when signal-to-noise ratio and perturbation budget meet specific conditions.
-
Policy Gradient Primal-Dual Method for Safe Reinforcement Learning from Human Feedback
Primal-dual policy gradient algorithms achieve global non-asymptotic convergence for safe RLHF cast as infinite-horizon discounted CMDPs without fitting reward models.
-
FaceParts: Segmentation and Editing of Gaussian Splatting
FaceParts performs unsupervised segmentation of facial features in Gaussian Splatting avatars and supports precise editing and cross-avatar part transfer using feature disentanglement, density clustering, and FLAME anchoring.
-
LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory
LongMemEval benchmarks long-term memory in chat assistants, revealing 30% accuracy drops across sustained interactions and proposing indexing-retrieval-reading optimizations that boost performance.
-
Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts
Loss-Free Balancing keeps expert loads balanced in MoE models by dynamically adjusting routing-score biases based on recent usage, avoiding auxiliary-loss interference and yielding better performance.
-
Your Absorbing Discrete Diffusion Secretly Models the Conditional Distributions of Clean Data
Absorbing discrete diffusion models the conditional distributions of clean data; reparameterizing yields a time-independent RADD that unifies with AO-ARMs and reaches SOTA perplexity among diffusion models on zero-shot language benchmarks.
-
Improving Dictionary Learning with Gated Sparse Autoencoders
Gated SAEs decouple which features to use from how large their activations should be, applying the L1 penalty only to selection and thereby eliminating shrinkage while halving the number of firing features needed for good fidelity.
-
Learning Interactive Real-World Simulators
UniSim learns a universal real-world simulator from orchestrated diverse datasets, enabling zero-shot deployment of policies trained purely in simulation.
-
Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation
A new shared video-image tokenizer enables large language models to surpass diffusion models on standard visual generation benchmarks.
-
EvoPrompt: Connecting LLMs with Evolutionary Algorithms Yields Powerful Prompt Optimizers
EvoPrompt uses LLMs to run evolutionary operators on populations of prompts, outperforming human-engineered prompts by up to 25% on BIG-Bench Hard tasks across 31 datasets.
-
DreamFusion: Text-to-3D using 2D Diffusion
Optimizes a Neural Radiance Field via probability density distillation from a 2D diffusion model to produce text-conditioned 3D scenes viewable from any angle.
-
Open-ended Multi-agent Autocurricula via Visual Inspection of Policies with Multi-modal LLMs
A lightweight VLM inspecting SMAC policy videos produces open-ended multi-agent curricula that outperform text-only ablations and PLR scalar-score methods on held-out maps.
-
Stochastic Order Learning: An Approach to Rank Estimation Using Noisy Data
Stochastic Order Learning associates each instance with multiple plausible ranks and trains embeddings via complementary discriminative and stochastic-order losses that remain robust to ordinal label noise.
-
Sparse Delta Memory: Scaling the State of Linear RNNs through Sparsity
SDM sparsifies the Gated DeltaNet update rule to enable 1000x larger recurrent memory states at iso-FLOP, improving long-context recall and short-context reasoning over GDN and matching full attention at 8B scale.
-
Geometric Self-Distillation for Reasoning Generalization
A geometric self-distillation objective using Hellinger loss and Fisher-Rao proximal regularization prevents predictive drift in LLM post-training, improving out-of-distribution reasoning by 5.7-8.6 points.
-
EvoPolicyGym: Evaluating Autonomous Policy Evolution in Interactive Environments
EvoPolicyGym is a new benchmark suite of 16 compact RL environments that evaluates autonomous policy evolution, with GPT-5.5 achieving the top aggregate rank and top-two performance on all tasks.
-
Multi-Objective Exploration and Preference Optimization via Mutual Information
Maximizing joint conditional mutual information I(Y; C_Z, W, Z | X) decomposes multi-objective LLM alignment into preference-specific DPO terms plus an I(Y;W|X) exploration term that reduces reward-distribution overlap.
-
AA: A Multi-view Multimodal Dataset for Screen-based Gaze Estimation
AA is a multi-view multimodal dataset for screen-based gaze estimation featuring eight screen-mounted and two side-view cameras with precise gaze targets and subject-independent splits.
-
Wait, am I Being Fair? Characterizing Deductive Stereotyping and Mitigating It with Fair-GCG
The paper characterizes deductive stereotyping in LLMs and introduces Fair-GCG to discover injection phrases that improve fairness across benchmarks, reasoning, and real-world tasks.
-
SGD at the Edge of Stability: Stochastic Stabilization with Large Learning Rates
SGD on multiclass cross-entropy loss alternates between curvature-driven oscillations and stable regimes but self-stabilizes to enable best-iterate convergence with large learning rates for linear and two-layer models.
-
Dynamic Prediction of Alternating Recurrent Events via Neural Network
Presents a neural network model with IPW pseudo-observations for dynamic prediction of alternating recurrent events, with simulation and medical resident mood application results.
-
LiftQuant: Continuous Bit-Width LLM via Dimensional Lifting and Projection
LiftQuant enables continuous bit-width LLM quantization via dimensional lifting and projection from a 1-bit lattice, allowing 2.4-bit compression of 70B models that outperforms fixed 2-bit baselines on identical hardware.
-
Onsager-Machlup Posterior Transport for Deep Gaussian Processes
OM-Path frames DGP inference as learning a deterministic transport map from reference to posterior inducing variables using probability-flow ODE on Doob-bridged SDE with Onsager-Machlup path regularization, yielding statistically significant gains over DBVI on power and protein UCI datasets.
-
Why SGD is not Brownian Motion: A New Perspective on Stochastic Dynamics
SGD is reformulated via a master equation from discrete updates, producing a discrete Fokker-Planck equation that predicts non-stationary variance growth proportional to learning rate in flat Hessian directions.
-
DelTA: Discriminative Token Credit Assignment for Reinforcement Learning from Verifiable Rewards
DelTA estimates token coefficients to amplify discriminative directions in token-gradient vectors, reweighting the RLVR surrogate to produce more contrastive side-wise centroids and yielding 3.26 and 2.62 point gains on math benchmarks for 8B and 14B Qwen3 models.
-
Divide and Contrast: Learning Robust Temporal Features without Augmentation
Di-COT is an unsupervised contrastive method that stochastically partitions time-series windows into overlapping sub-blocks to learn representations without augmentation, reporting SOTA results on classification and transfer tasks across multiple benchmarks while cutting training time.
-
Behavior-Consistent Deep Reinforcement Learning
QED bounds cross-run KL divergence in Boltzmann policies by setting temperature proportional to Q-disagreement and reduces return variance by two orders of magnitude on 18 continuous-control tasks without performance loss.
-
HORST: Composing Optimizer Geometries for Sparse Transformer Training
HORST uses non-commutative operator composition and a hyperbolic mirror map to combine stability from adaptive optimizers with L1 sparsity bias, outperforming AdamW across sparsity levels on vision and language tasks.
-
Towards Understanding Self-Pretraining for Sequence Classification
Self-pretraining improves Transformer sequence classification by enabling learning of proximity-biased attention from positional encodings that label supervision alone cannot easily acquire from random starts.