Papers by Yaniv Nemcovsky
Representing LLMs in Prompt Semantic Task Space (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models (LLMs) achieve impressive results over various tasks, and public repositories contain an abundance of pre-trained models. |
| Approach: | They propose an efficient, training-free approach to representing LLMs as linear operators within the prompts’ semantic task space. |
| Outcome: | The proposed representations achieve state-of-the-art results on success prediction and model selection tasks with notable performance in out-of sample scenarios. |
REMIND: Memorization and Unlearning in LLMs Through the Lens of Input Loss Landscapes (2026.acl-long)
Copied to clipboard
| Challenge: | REMIND is a framework that diagnoses residual memorization states by probing local ILL curvature over semantically coherent neighborhoods. |
| Approach: | They propose a framework that diagnoses memorization states by probing local ILL curvature over semantically coherent neighborhoods. |
| Outcome: | The proposed framework outperforms baseline models with 82% multi-class ROC-AUC and 2 higher AUC at 1% FPR. |
Jailbreak Attack Initializations as Extractors of Compliance Directions (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Safety-aligned LLMs respond to prompts with compliance or refusal, each corresponding to distinct directions in the model’s activation space. |
| Approach: | They propose an initialization framework that aims to project unseen prompts further along compliance directions. |
| Outcome: | The proposed initialization framework achieves an increased attack success rate and reduced computational overhead, highlighting the fragility of safety-aligned LLMs. |