REVIEW 6 cited by
Model Editing Harms General Abilities of Large Language Models: Regularization to the Rescue
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
read the original abstract
Model editing is a technique that edits the large language models (LLMs) with updated knowledge to alleviate hallucinations without resource-intensive retraining. While current model editing methods can effectively modify a model's behavior within a specific area of interest, they often overlook the potential unintended side effects on the general abilities of LLMs such as reasoning, natural language inference, and question answering. In this paper, we raise concerns that model editing's improvements on factuality may come at the cost of a significant degradation of the model's general abilities. We systematically analyze the side effects by evaluating four popular editing methods on three LLMs across eight representative tasks. Our extensive empirical experiments show that it is challenging for current editing methods to simultaneously improve factuality of LLMs and maintain their general abilities. Our analysis reveals that the side effects are caused by model editing altering the original model weights excessively, leading to overfitting to the edited facts. To mitigate this, a method named RECT is proposed to regularize the edit update weights by imposing constraints on their complexity based on the RElative Change in weighT. Evaluation results show that RECT can significantly mitigate the side effects of editing while still maintaining over 94% editing performance.
Forward citations
Cited by 6 Pith papers
-
Aligned but Blind: Alignment Increases Implicit Bias by Reducing Awareness of Race
Alignment on Llama 3 reduces explicit bias but amplifies implicit bias, because aligned models no longer represent 'black' and 'white' as racial concepts in ambiguous contexts.
-
Moir: Let the Model Direct Its Own Story for Robust Cross-Domain Knowledge Editing
MOIR: estimating the preservation covariance from a model's own random-token generations reduces collapse of math/code capabilities in some knowledge-editing settings, but the claimed consistency is not supported by t...
-
COMPKE: Complex Question Answering under Knowledge Editing
COMPKE is a new benchmark with 11,924 complex questions that tests knowledge editing through one-to-many relations and logical operations, where existing editing methods often fail.
-
Reproducibility Study of "AlphaEdit: Null-Space Constrained Knowledge Editing for Language Models"
Reproducibility study confirms AlphaEdit on original setups but finds performance degrades at high edit counts, fails to generalize to newer models, and harms downstream tasks.
-
SOS-LoRA: Static Orthogonal-Subspace Low-Rank Adaptation with Fixed Multi-Scale Scaling
A LoRA update split into several fixed, differently-scaled low-rank experts with orthogonal input directions improves fine-tuning accuracy at the same parameter count.
-
AUTOLAW: Enhancing Legal Compliance in Large Language Models via Case Law Generation and Jury-Inspired Deliberation
AutoLaw's verifier-ranked legal-role jury with a similar-case demonstration beats majority voting for violation detection on three law and policy benchmarks.
Discussion (0). Sign in to comment.