Tom Goldstein

59

Papers

411

Total Citations

Papers (59)

Training Quantized Nets: A Deeper Understanding

NeurIPS 2017arXiv

Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach

Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives

Gemstones: A Model Suite for Multi-Faceted Scaling Laws

NeurIPS 2025arXiv

Efficient Fine-Tuning and Concept Suppression for Pruned Diffusion Models

ARGUS: Hallucination and Omission Evaluation in Video-LLMs

Making an Invisibility Cloak: Real World Adversarial Attacks on Object Detectors

Adaptive Primal-Dual Splitting Methods for Statistical Learning and Image Processing

ACE: Adapting to Changing Environments for Semantic Segmentation

PUP 3D-GS: Principled Uncertainty Pruning for 3D Gaussian Splatting

Diffusion Art or Digital Forgery? Investigating Data Replication in Diffusion Models

A New Rank Constraint on Multi-View Fundamental Matrices, and Its Application to Camera Location Recovery

Zero-Shot Vision Encoder Grafting via LLM Surrogates

Can Neural Nets Learn the Same Model Twice? Investigating Reproducibility and Double Descent From the Decision Boundary Perspective

Robust Optimization As Data Augmentation for Large-Scale Graphs

Can Watermarking Large Language Models Prevent Copyrighted Text Generation and Hide Training Data?

Object Recognition as Next Token Prediction

InstructZero: Efficient Instruction Optimization for Black-Box Large Language Models

WAVES: Benchmarking the Robustness of Image Watermarks

ODIN: Disentangled Reward Mitigates Hacking in RLHF

Spotting LLMs With Binoculars: Zero-Shot Detection of Machine-Generated Text

Estimating Sparse Signals With Smooth Support via Convex Programming and Block Sparsity

Adaptive Relaxed ADMM: Convergence Theory and Practical Implementation

Battle of the Backbones: A Large-Scale Comparison of Pretrained Models across Computer Vision Tasks

Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise

A Performance-Driven Benchmark for Feature Selection in Tabular Deep Learning

Understanding and Mitigating Copying in Diffusion Models

Hard Prompts Made Easy: Gradient-Based Discrete Optimization for Prompt Tuning and Discovery

Tree-Rings Watermarks: Invisible Fingerprints for Diffusion Images

On the Exploitability of Instruction Tuning

What Can We Learn from Unlearnable Datasets?

Dealbreaker: A Nonlinear Latent Variable Model for Educational Data

Training Neural Networks Without Gradients: A Scalable ADMM Approach

Convex Phase Retrieval without Lifting via PhaseMax

Adaptive Consensus ADMM for Distributed Optimization

Linear Spectral Estimators and an Application to Phase Retrieval

Transferable Clean-Label Poisoning Attacks on Deep Neural Nets

Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural Networks

Visualizing the Loss Landscape of Neural Nets

Adversarial training for free!

Detection as Regression: Certified Object Detection with Median Smoothing

Certifying Strategyproof Auction Networks

Certifying Confidence via Randomized Smoothing

MetaPoison: Practical General-purpose Clean-label Data Poisoning

Adversarially Robust Few-Shot Learning: A Meta-Learning Approach

Center Smoothing: Certified Robustness for Networks with Structured Outputs

Can You Learn an Algorithm? Generalizing from Easy to Hard Problems with Recurrent Networks

VQ-GNN: A Universal Framework to Scale up Graph Neural Networks using Vector Quantization

GradInit: Learning to Initialize Neural Networks for Stable and Efficient Training

Long-Short Transformer: Efficient Transformers for Language and Vision

Gradient-Free Adversarial Training Against Image Corruption for Learning-based Steering

Encoding Robustness to Image Style via Adversarial Feature Perturbations

Adversarial Examples Make Strong Poisons

Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Models

Where do Models go Wrong? Parameter-Space Saliency Maps for Explainability

Sleeper Agent: Scalable Hidden Trigger Backdoors for Neural Networks Trained from Scratch

End-to-end Algorithm Synthesis with Recurrent Networks: Extrapolation without Overthinking

Autoregressive Perturbations for Data Poisoning

Robustness Disparities in Face Detection