REVIEW 5 cited by
Editable Neural Networks
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
abstract
These days deep neural networks are ubiquitously used in a wide range of tasks, from image classification and machine translation to face identification and self-driving cars. In many applications, a single model error can lead to devastating financial, reputational and even life-threatening consequences. Therefore, it is crucially important to correct model mistakes quickly as they appear. In this work, we investigate the problem of neural network editing $-$ how one can efficiently patch a mistake of the model on a particular sample, without influencing the model behavior on other samples. Namely, we propose Editable Training, a model-agnostic training technique that encourages fast editing of the trained model. We empirically demonstrate the effectiveness of this method on large-scale image classification and machine translation tasks.
Forward citations
Cited by 5 Pith papers
-
Can Gradient Descent Simulate Prompting?
A MAML-style meta-training objective makes a single gradient step on new text recover part of the performance that prompting achieves, on reversal-curse and passage-QA tasks.
-
Implicit Reasoning Steering via Concept Chaining
Reinforcement-learning-optimized concept-chain paragraphs covertly steer language-model multiple-choice preferences after continued pretraining, with far lower detectability than direct paraphrases.
-
Towards a Principled Evaluation of Knowledge Editors
The choice of evaluation metric, generation length, and edit batch size changes the ranking of knowledge editors, and exact string matching produces false positives.
-
Improving LLM-Based Fault Localization with External Memory and Project Context
MemFL gives an LLM static project summaries and dynamic debugging tips, and reports a 12.7% Top-1 accuracy gain over LLM fault localization baselines on Defects4J with lower time and cost.
-
Benchmarking and Rethinking Knowledge Editing for Large Language Models
Under autoregressive and sequential editing, parameter-based knowledge editing methods perform poorly, while the retrieval-based SCR baseline consistently outperforms them across datasets and models.
Discussion (0). Sign in to comment.