Model removes objects from general scenes using weak supervision.
problem Automatic object removal from general scene images with weak supervision.
method Two-stage editor architecture with mask generator and image in-painter; novel GAN prior for mask generator.
result Effectively removes a wide variety of objects from general scenes using weak supervision.
ED-NeRF efficiently edits 3D scenes using latent space NeRF and improved loss functions.
problem Slow training speeds and inadequate editing loss functions in existing NeRF editing techniques.
method Embedding real-world scenes into latent space of LDM, using a unique refinement layer and an improved loss function.
result ED-NeRF achieves faster editing speed and improved output quality compared to state-of-the-art models.
BPI models 2D patterns on multiple planes and 3D scene from a single image.
problem Understanding and editing images with multiple 2D planes and 3D scene from a single image.
method Box Program Induction (BPI) with neural networks and search-based algorithm.
result Holistic, structured scene representation enables 3D-aware image editing.
Agents use object-oriented reasoning to solve problems more quickly.
problem Finding solutions to problems more efficiently.
method Hierarchical controller directs low-level agent to simulate alternate states of the world.
result Achieves similar reward levels as non-hierarchical agents but with better data efficiency.
Adversarial edit attacks improve machine learning model security for tree data.
problem Improving security of machine learning models for tree-structured data.
method Extends adversarial attacks to tree-structured data using tree edit distance and black-box queries.
result Many tree classifiers can be effectively attacked, demonstrating the vulnerability of these models.
RS-Del provides robustness for sequence classifiers against edit distance attacks.
problem Certifying robustness of discrete sequence classifiers against edit distance attacks.
method Randomized deletion (RS-Del) for discrete sequence classifiers, focusing on edit distance-bounded adversaries.
result Achieved a certified accuracy of 91% at an edit distance radius of 128 bytes on malware detection.
Facial attribute editing aims to manipulate single or multiple attributes of a face image, i.e., to generate a new face with desired attributes while preserving other details. Recently, generative adversarial net (GAN) and encoder-decoder architecture are usually incorporated to handle this task with promising results.…
This work proposes a method to compose visual relations more faithfully.
problem Composing relations between objects in images is challenging due to their entanglement.
method Represent each relation as an unnormalized density (energy-based model) to compose relations factorizedly.
result The proposed method generates and edits scenes with multiple sets of relations more faithfully.
Differentiable language model attacks improve adversarial examples for categorical sequence classifiers.
problem Challenges in adversarial attacks for categorical sequence models due to non-differentiability.
method Fine-tuning a language model as a generator of adversarial examples, using a differentiable loss function that combines a surrogate classifier score and approximate edit distance.
result Semantically better and more resistant adversarial examples.
Generative model learns to navigate image manifolds for unsupervised video generation.
problem Generating natural-looking scenes of moving objects without labeled data.
method Two-step training scheme with GANs, using a recurrent model to generate coherent sequences.
result Recurrent model learns to navigate the image manifold induced by the frames generator, producing more natural scenes.
A stroke-based method reconstructs characters from noisy images.
problem Challenges in reconstructing characters from noisy images.
method Uses a weighted quadratic Bezier curve (WQBC) to represent strokes.
result Achieves excellent reconstruction in real scenes.
Proposes a deep structured generative model for complex scenes.
problem Difficulty in generating images with complicated structures.
method Boosts GANs with structure information encoded by a stochastic and-or graph (sAOG).
result Successfully captures and generates images with complex structures.
This paper tackles disentanglement in image editing and reconstruction.
problem Learning disentangled image representations and balancing disentanglement strength and reconstruction quality.
method Distance covariance based decorrelation regularization for disentanglement, soft target representation for reconstruction, and collapsing AE decoder and GAN generator.
result The proposed model improves the disentanglement strength and perceptual quality of generated images.
DNCF framework recovers real scenes from imperfect images robustly.
problem Recovering real scenes from imperfect images.
method Nonparametric deep network that learns physical image formation equations.
result DNCF framework robustly defends against adversarial attacks.
The paper teaches a reinforcement learning agent to generate diverse programs based on symbolic instructions.
problem Learning to generate diverse programs for diverse scenes given a symbolic instruction.
method Instruction-conditioned reinforced adversarial learning.
result The agent's stochastic policy more accurately captures the diversity in the goal distribution.
Models learn to represent edits from natural language and code.
problem Learning distributed representations of edits.
method Combining a neural editor with an edit encoder.
result Models can capture the structure and semantics of edits.
A novel estimator for linear coefficients in semiparametric models without assuming model structure.
problem Estimating nuisances in semiparametric models without knowing the underlying structure.
method Proposes a novel estimator and a method called TAME for debiasing and improving on double machine learning.
result Establishes a new estimator with improved error rate compared to double machine learning.
Generative model improves scene graph prediction for rare compositions.
problem Challenges in inferring unseen or rare scene graph compositions.
method Synthesizing rare scene graphs by perturbing real ones, using conditional GANs.
result Marginal but consistent improvements in zero- and few-shot metrics.
Researchers develop methods to prevent GANs from generating certain types of images.
problem Pre-trained GANs sometimes produce undesirable samples.
method Post-editing GANs to prevent certain types of outputs, using three algorithms.
result Our algorithms effectively prevent GANs from generating certain types of images while maintaining high quality.
SeqAttnGAN generates interactive images based on multi-turn text descriptions.
problem Interactive image editing with multi-turn textual commands.
method SeqAttnGAN uses a neural state tracker and GAN framework for sequential image generation and refinement.
result SeqAttnGAN outperforms state-of-the-art models on interactive image editing tasks.
Proposes a method to quantify uncertainty in deterministic image classifiers.
problem Uncertainty in deterministic image classifiers.
method Introduces Wellington Posterior for inductive transfer from scenes.
result Validates Wellington Posterior using various methods.
We present a new, efficient method for automatically detecting severe conflicts `edit wars' in Wikipedia and evaluate this method on six different language WPs. We discuss how the number of edits, reverts, the length of discussions, the burstiness of edits and reverts deviate in such pages from those following the gene…
A new smooth edit distance for easier optimization in machine learning.
problem Hard optimization of edit distance for variable-length sequences.
method Soft edit distance (SED) as a differentiable approximation.
result SED can be optimized with gradient methods and used for clustering.
As the number of contributors to online peer-production systems grows, it becomes increasingly important to predict whether the edits that users make will eventually be beneficial to the project. Existing solutions either rely on a user reputation system or consist of a highly specialized predictor that is tailored to …
Light-based attacks can misclassify images without altering physical objects.
problem Physical attacks on deep learning classifiers are limited by the ability to modify inputs directly.
method Constructs an experimental setup with a light projection source, object, and camera. Uses differential evolution to select light patterns.
result Projected light can degrade classification accuracy from 98% to 22% for 2D objects and from 89% to 43% for 3D objects.
Neural networks model code edits from source code changes.
problem Modeling dynamic changes in source code.
method Developed neural networks to learn and predict code edits.
result Attentional and pointer network components provide best performance.
The paper explores how information theory aids in statistical learning models.
problem Characterizing fundamental performance limits in statistical learning models.
method Introduces divergence measures and evidence lower bound (ELBO) in model training.
result Provides a systematic derivation for generative diffusion models.
Ground-A-Video edits videos without training, preserving intended changes.
problem Complex multi-attribute video editing with omitted or wrong changes.
method Grounding-guided video-to-video translation with Cross-Frame Gated Attention.
result Zero-shot multi-attribute video editing with improved accuracy and frame consistency.
Adversarial training helps classifiers resist universal perturbations.
problem Vulnerability of classifiers to universal perturbations.
method Adversarial training with shared adversarial examples.
result Adversarial training reduces sensitivity to universal perturbations.
A framework that edits existing examples to generate structured outputs.
problem Generating complex outputs like code is hard.
method Retrieve a training example, edit it, and use a computationally efficient retrieval model.
result Retrieve-and-edit framework significantly boosts performance on code and card tasks.
Paper fine-tunes LLMs using user edits, unifying preference, supervision, and reward feedback.
problem Adapting LLMs to user preferences and feedback types.
method Derives bounds for learning algorithms from user edits, proposes an ensembling procedure.
result Ensembling procedure outperforms individual feedback methods and robustly adapts to different user-edit distributions.
The increasingly photorealistic sample quality of generative image models suggests their feasibility in applications beyond image generation. We present the Neural Photo Editor, an interface that leverages the power of generative neural networks to make large, semantically coherent changes to existing images. To tackle…
We propose a new generative model of sentences that first samples a prototype sentence from the training corpus and then edits it into a new sentence. Compared to traditional models that generate from scratch either left-to-right or by first sampling a latent sentence vector, our prototype-then-edit model improves perp…
MEGAN models chemical reactions as graph edits, improving synthesis planning.
problem Generating and predicting chemical reactions under constraints.
method End-to-end encoder-decoder neural model inspired by arrow pushing formalism.
result State-of-the-art accuracy in standard benchmarks for retrosynthesis prediction.
New method uses coupled SDEs to edit images with high fidelity and consistency.
problem Challenges in editing image content with text-to-image models.
method Using coupled stochastic differential equations to guide generative model sampling.
result Achieves high prompt fidelity and near-pixel-level consistency.
Paper proposes efficient image inversion and editing using rectified stochastic differential equations.
problem Inversion and editing of real images using generative models.
method Proposes RF inversion using dynamic optimal control and a linear quadratic regulator, extending to stochastic sampler for Flux.
result Allows state-of-the-art performance in zero-shot inversion and editing, outperforming prior works.
New method detects watermarks in LLM-generated text with human edits.
problem Dilution of watermark signals by human edits on LLM-generated text.
method Truncated goodness-of-fit test (Tr-GoF) for robust detection.
result Tr-GoF achieves optimality in robust detection of Gumbel-max watermark.
A technique to quickly fix mistakes in neural networks.
problem Fixing model errors in neural networks quickly and without affecting other samples.
method Editable Training, a model-agnostic training technique.
result Effectiveness demonstrated on large-scale image classification and machine translation tasks.
Contrastive Code Representation Learning improves code summarization and type inference.
problem Code representations are sensitive to edits, hindering downstream semantic understanding tasks.
method ContraCode: a contrastive pre-training task that learns code functionality.
result Contrastive pre-training improves code summarization and type inference accuracy.
Machine learning experiments show IID assumption is flawed for bathymetry editing.
problem Flawed IID assumption in machine learning for bathymetry editing.
method Real-world computer-assisted labeling task, IID assumption analysis.
result Common random split leads to poor performance in machine learning.
Metric learning improves tree edit distance for better classification.
problem Improving tree edit distance for better classification accuracy.
method Adaptive symbol embeddings to learn tree edit distance indirectly.
result Embedding edit distance learning (BEDL) improves upon state-of-the-art on multiple tree data sets.
A neural scene representation framework enforcing 3D transformations.
problem Learning 3D scene representations from images without 3D supervision.
method Introducing a loss enforcing equivariance of the scene representation with 3D transformations.
result Real-time neural rendering with comparable results to models requiring minutes for inference.
GMED edits stored examples to improve continual learning.
problem Catastrophic forgetting in task-free continual learning.
method Gradient-based memory editing of stored examples.
result GMED-edited examples help prevent forgetting.
Two JSMA variants improve speed and accuracy for image classification attacks.
problem Fooling deep neural networks with adversarial images.
method Maximizing pixel values to mislead classification models.
result Two JSMA variants perform well on digit and scene datasets.
New seq2seq model can copy entire spans, outperforming simpler models in editing tasks.
problem Editing documents or source code using seq2seq models with explicit token copying.
method Extended seq2seq model capable of copying entire input spans to output in one step, new training and inference methods.
result New model consistently outperforms simpler baselines in editing tasks of natural language and source code.
Professional-grade software applications are powerful but complicated−expert users can achieve impressive results, but novices often struggle to complete even basic tasks. Photo editing is a prime example: after loading a photo, the user is confronted with an array of cryptic sliders like "clarity", "temp", and "high…
GENESIS generates and samples 3D scenes by capturing object interactions.
problem Lack of models that explicitly capture object interactions in scene generation.
method Object-centric latent variables, spatial GMM, amortized inference, autoregressive prior.
result First object-centric generative model of 3D visual scenes.
3D adversarial logos can fool object detectors in real-world settings.
problem Creating robust adversarial attacks in 3D rendering views.
method Constructing 3D adversarial logos via texture mapping and differentiable rendering.
result 3D adversarial logos are more versatile and robust than traditional adversarial patches.