M³Att poisons medical multimodal RAG by pairing covert textual misinformation with query-agnostic visual perturbations that increase retrieval of the bad content, causing LLMs to generate clinically plausible but incorrect responses.
Title resolution pending
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
baseline 1
citation-polarity summary
years
2026 8verdicts
UNVERDICTED 8roles
baseline 1polarities
baseline 1representative citing papers
OpenSafeIntent benchmark shows models fail to calibrate safety across intent shifts in matched dual-use prompts, indicating current evaluations are insufficient.
Public healthcare agent skills emphasize workflow automation over clinical diagnostics and treatments, with uneven lifecycle coverage and weak alignment between technical and clinical risk.
Quantile tokens inserted into LLM inputs combined with neighbor retrieval enable direct prediction of full distributions, yielding lower MAPE and narrower intervals than baselines on Airbnb and StackSample tasks.
Partial harnesses for LLM agents, specifying only initial execution steps, achieve higher pass rates than fully decomposed workflows, as analyzed through trajectory alignment and validated in synthetic and terminal benchmarks.
SCURank ranks multiple summary candidates with Summary Content Units to outperform ROUGE and LLM-based methods in summarization distillation.
SIREN identifies safety neurons via linear probing on internal LLM layers and combines them with adaptive weighting to detect harm, outperforming prior guard models with 250x fewer parameters.
Regression analysis of 53k astrophysics papers finds open data, open access, and open code each linked to higher citation counts after controlling for grants, authors, length, date, and subfield.
citing papers explorer
-
Knowledge Poisoning Attacks on Medical Multi-Modal Retrieval-Augmented Generation
M³Att poisons medical multimodal RAG by pairing covert textual misinformation with query-agnostic visual perturbations that increase retrieval of the bad content, causing LLMs to generate clinically plausible but incorrect responses.
-
OpenSafeIntent: Evaluating Intent-Calibrated Safe Completion Across Dual-Use Prompt Sets
OpenSafeIntent benchmark shows models fail to calibrate safety across intent shifts in matched dual-use prompts, indicating current evaluations are insufficient.
-
An Empirical Study of Agent Skills for Healthcare: Practice, Gaps, and Governance
Public healthcare agent skills emphasize workflow automation over clinical diagnostics and treatments, with uneven lifecycle coverage and weak alignment between technical and clinical risk.
-
Text-to-Distribution Prediction with Quantile Tokens and Neighbor Context
Quantile tokens inserted into LLM inputs combined with neighbor retrieval enable direct prediction of full distributions, yielding lower MAPE and narrower intervals than baselines on Airbnb and StackSample tasks.
-
Harnesses for Inference-Time Alignment over Execution Trajectories
Partial harnesses for LLM agents, specifying only initial execution steps, achieve higher pass rates than fully decomposed workflows, as analyzed through trajectory alignment and validated in synthetic and terminal benchmarks.
-
SCURank: Ranking Multiple Candidate Summaries with Summary Content Units for Enhanced Summarization
SCURank ranks multiple summary candidates with Summary Content Units to outperform ROUGE and LLM-based methods in summarization distillation.
-
LLM Safety From Within: Detecting Harmful Content with Internal Representations
SIREN identifies safety neurons via linear probing on internal LLM layers and combines them with adaptive weighting to detect harm, outperforming prior guard models with 250x fewer parameters.
-
Open Science in Astrophysics: Citation Benefits of Open Code, Open Data, and Open Access
Regression analysis of 53k astrophysics papers finds open data, open access, and open code each linked to higher citation counts after controlling for grants, authors, length, date, and subfield.