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.
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,657 papers · 148 categories
Trend · papers per month
48 results for “memory editing”
We introduce the problem of learning distributed representations of edits. By combining a "neural editor" with an "edit encoder", our models learn to represent the salient information of an edit and can be used to apply edits to new inputs. We experiment on natural language and source code edit data. Our evaluation yie…
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.
Edit wars in Wikipediastat.ML
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…
FlowChef steers RFMs to efficiently guide image generation tasks.
problem Efficiently guiding image generation tasks with RFMs.
method Developed a theoretical and empirical understanding of RFMs' vector field dynamics, proposing FlowChef for gradient-free navigation.
result FlowChef significantly outperforms baselines in performance, memory, and time requirements.
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 …
LLMs struggle to optimize hyperparameters efficiently, but hybrid methods can improve performance.
problem Optimizing hyperparameters of small language models using LLMs.
method Comparison of classical HPO algorithms and LLM-based methods, introducing Centaur hybrid approach.
result Hybrid Centaur approach achieves best results, outperforming classical and pure LLM methods.
Edit distance, also known as Levenshtein distance, is an essential way to compare two strings that proved to be particularly useful in the analysis of genetic sequences and natural language processing. However, edit distance is a discrete function that is known to be hard to optimize. This fact hampers the use of this …
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.
Principled Fine-tuning of LLMs from User-Edits: A Medley of Preference, Supervision, and Rewardcs.LG
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.
Programming languages are emerging as a challenging and interesting domain for machine learning. A core task, which has received significant attention in recent years, is building generative models of source code. However, to our knowledge, previous generative models have always been framed in terms of generating stati…
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.
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.
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.
For the task of generating complex outputs such as source code, editing existing outputs can be easier than generating complex outputs from scratch. With this motivation, we propose an approach that first retrieves a training example based on the input (e.g., natural language description) and then edits it to the desir…
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.
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.…
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 complicatedexpert 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…
New metric learning approach for tree data reduces computation cost.
problem Efficiently computing distances between ordered labeled trees.
method Introduced pq-grams and a differentiable weighted pq-gram distance, combined with LMNN for optimization.
result Significantly reduces computation time for tree classification problems.
String kernels are attractive data analysis tools for analyzing string data. Among them, alignment kernels are known for their high prediction accuracies in string classifications when tested in combination with SVM in various applications. However, alignment kernels have a crucial drawback in that they scale poorly du…
ARED introduces a new dataset for Argentina's real estate market.
problem Lack of mixed modality datasets for Argentina's real estate market.
method Developed a comprehensive real estate price prediction dataset series for Argentina.
result Dataset captures time-dependent phenomena on a market level.
Metric learning has the aim to improve classification accuracy by learning a distance measure which brings data points from the same class closer together and pushes data points from different classes further apart. Recent research has demonstrated that metric learning approaches can also be applied to trees, such as m…
The Wikimedia Foundation has recently observed that newly joining editors on Wikipedia are increasingly failing to integrate into the Wikipedia editors' community, i.e. the community is becoming increasingly harder to penetrate. To sustain healthy growth of the community, the Wikimedia Foundation aims to quantitatively…
Many machine learning models can be attacked with adversarial examples, i.e. inputs close to correctly classified examples that are classified incorrectly. However, most research on adversarial attacks to date is limited to vectorial data, in particular image data. In this contribution, we extend the field by introduci…
The paper introduces COAR to estimate component attributions and enable model editing.
problem Understanding how ML models transform inputs into predictions.
method Component modeling and COAR algorithm for estimating component attributions.
result COAR enables model editing across various tasks.
The study examines Euclid's Book I, focusing on area applications and construction methods.
problem Exploring Euclid's geometric constructions and proofs, particularly those involving area calculations.
method Summarizing medieval editions and ancient commentaries, comparing constructions and proofs.
result Medieval editions often avoid Euclid's use of superposition in area proofs, offering alternative constructions.
LoMC: Localized Multidirectional Correction for Refusal Suppression in Routed Foundation Modelsstat.ML
Localized Multidirectional Correction improves non-refusal target-response behavior in foundation models.
problem Controlled post-training refusal suppression in routed MoE and hybrid-MoE foundation models.
method Introduce Localized Multidirectional Correction (LoMC), a support-gated intervention framework.
result Substantially improves non-refusal target-response behavior while maintaining general capability under a compact intervention footprint.
Causal KL improves on existing metrics for evaluating causal models.
problem Insufficient discrimination between causal models using edit-distance and KL divergence.
method Introducing Causal KL, an augmented KL divergence that considers causal relationships.
result Causal KL variants effectively distinguish between observationally equivalent models.
While neural sequence generation models achieve initial success for many NLP applications, the canonical decoding procedure with left-to-right generation order (i.e., autoregressive) in one-pass can not reflect the true nature of human revising a sentence to obtain a refined result. In this work, we propose XL-Editor, …
Tree Edit Distance Learning via Adaptive Symbol Embeddings: Supplementary Materials and Resultscs.LG
Metric learning has the aim to improve classification accuracy by learning a distance measure which brings data points from the same class closer together and pushes data points from different classes further apart. Recent research has demonstrated that metric learning approaches can also be applied to trees, such as m…
This is a survey written for a special edition of the journal of differential geometry.
We consider the setting of Reeb graphs of piecewise linear functions and study distances between them that are stable, meaning that functions which are similar in the supremum norm ought to have similar Reeb graphs. We define an edit distance for Reeb graphs and prove that it is stable and universal, meaning that it pr…
Many prediction problems can be phrased as inferences over local neighborhoods of graphs. The graph represents the interaction between entities, and the neighborhood of each entity contains information that allows the inferences or predictions. We present an approach for applying machine learning directly to such graph…
Proposes CLSM for better subsequence generation in music sequences.
problem Editing subsequences in music sequences without losing context.
method Context-informed prior and decoder for generative model, context position-informed encoder for inference.
result Contextual latent space is smoother in interpolation and generates higher quality samples.
New diffusion models improve counterfactual image generation with semantic control.
problem Challenges in preserving identity, maintaining quality, and ensuring causal model faithfulness in counterfactual image generation.
method Integrates semantic representations into diffusion models through Pearlian causality, introducing spatial, semantic, and dynamic abduction.
result Demonstrates high-level semantic identity preservation and principled trade-offs between faithful causal control and identity preservation.
LLMs optimize quantum circuits by iteratively improving proposals with feedback and memory traces.
problem Optimizing quantum circuits using large language models (LLMs) under black-box evaluation.
method Closed-loop, test-time optimization with LLMs, score-difference feedback, and restart-from-the-best sampling.
result The approach improves circuit synthesis performance and success rate, especially for larger qubit settings.
Paper introduces STSL, a second-order Tweedie sampler for efficient posterior sampling in inverse problems.
problem Computational challenges in sampling from posterior distributions using latent diffusion models.
method Introduces STSL, a novel second-order Tweedie sampler with tractable reverse process.
result STSL achieves 4X and 8X reduction in neural function evaluations compared to state-of-the-art solvers.
PCBMs turn any neural network into interpretable models without dense annotations.
problem Restrictive nature of CBMs and lack of dense concept annotations in training data.
method Introduce PCBMs that can turn any neural network into interpretable models without dense annotations.
result PCBMs can turn any neural network into interpretable models without dense annotations, improving interpretability and performance.
We introduce agents that use object-oriented reasoning to consider alternate states of the world in order to more quickly find solutions to problems. Specifically, a hierarchical controller directs a low-level agent to behave as if objects in the scene were added, deleted, or modified. The actions taken by the controll…
MoleculeSTM learns from molecule structures and texts for better drug design.
problem Lack of integration between chemical structures and textual knowledge in AI drug discovery.
method Jointly learns chemical structures and texts via contrastive learning, using a large dataset.
result MoleculeSTM achieves state-of-the-art performance in zero-shot tasks like structure-text retrieval and molecule editing.
Most existing text-to-image synthesis tasks are static single-turn generation, based on pre-defined textual descriptions of images. To explore more practical and interactive real-life applications, we introduce a new task - Interactive Image Editing, where users can guide an agent to edit images via multi-turn textual …
This is a chapter that is to appear in the "Handbook of Knot Theory", edited by William W. Menasco and Morwen B. Thistlethwaite.
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.