Cordyceps poisoning induces an information hiding scheme in LLMs via semantic associations, enabling covert control attacks with 40% higher success than prior methods and up to 98% survival against defenses.
org/abs/2004.09813
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
roles
method 1polarities
use method 1representative citing papers
HTEB introduces dynamic, multi-axis evaluation of text embedding robustness using LLM transformations, finding decoupled profiles across models and that scaling does not close all robustness gaps.
A code-and-comment analysis method detects semantic clones in Solidity functions with 59% overall precision (84% for same-name functions) and 97% recall on 300k contracts, plus LLM summaries for uncommented code.
SeedER uses initial dense seeding followed by RL-driven selective expansion to improve recall on compositional KG queries while limiting candidate set size.
LLaVA-Video-178K is a new synthetic video instruction dataset that, when combined with existing data to train LLaVA-Video, produces strong results on video understanding benchmarks.
A 200M-parameter Turkish sentence embedding model is adapted from a multilingual teacher via tokenizer pruning, mean-composition initialization, and offline cosine distillation, achieving 77.55% Pearson correlation on STSbTR and 7th place on TR-MTEB.
STAF applies sentence embeddings from transformers to classify SCA findings, reaching 89% F1 and beating prior filters by 11% within projects and 6% across projects.
Introduces Tree Generation (TG-SFT) to generate synthetic instruction-tuning data from LLMs, reducing catastrophic forgetting when fine-tuning MLLMs on domain-specific or multimodal data.
citing papers explorer
-
Towards Better Static Code Analysis Reports: Sentence Transformer-based Filtering of Non-Actionable Alerts
STAF applies sentence embeddings from transformers to classify SCA findings, reaching 89% F1 and beating prior filters by 11% within projects and 6% across projects.