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.
hub
ACM Transactions on Information Systems , volume=
17 Pith papers cite this work. Polarity classification is still indexing.
17
Pith papers citing it
hub tools
citation-role summary
background 3
citation-polarity summary
years
2026 17verdicts
UNVERDICTED 17roles
background 3representative citing papers
OptiVerse is a new benchmark spanning neglected optimization domains that shows LLMs suffer sharp accuracy drops on hard problems due to modeling and logic errors, with a Dual-View Auditor Agent proposed to improve performance.
Uncertainty and correctness in LLMs are encoded by distinct feature populations, with suppression of confounded features improving accuracy and reducing entropy.
SG-RAG frames retrieval as subgraph matching to ensure LLMs meet every condition in factual queries and reports large gains over baselines on a new 120k-pair ERQA dataset.
Audit of ChatGPT, Copilot, Gemini and Perplexity finds ~16% of cited sources are AI-generated across 712 queries on politics, health and environment.
SAGE is a self-evolving agentic graph-memory engine that dynamically constructs and refines structured memory graphs via writer-reader feedback, yielding performance gains on multi-hop QA, open-domain retrieval, and long-term agent benchmarks.
Theoretical analysis of continual factual knowledge acquisition shows data replay stabilizes pretrained knowledge by shifting convergence dynamics while regularization only slows forgetting, leading to the STOC method for attention-based replay selection.
LegalCiteBench reveals that current LLMs achieve under 7% accuracy on closed-book legal citation retrieval and completion tasks, with misleading answer rates above 94% for nearly all tested models.
LiteGUI trains 2B/3B-scale GUI agents via SFT-free guided on-policy distillation and multi-solution dual-level GRPO to reach SOTA lightweight performance and compete with larger models.
CAGE uses common-agency games and an EPEC algorithm to compute equilibrium policies that balance multiple conflicting objectives for test-time LLM alignment.
Verbal-R3 uses a verbal reranker to generate analytic narratives that guide retrieval and reasoning in LLMs, achieving SOTA results on complex QA benchmarks.
LUCID detects hallucinations in LLM-KG reasoning by extracting node/edge features from attention and semantics then integrating them with KG structure in a GNN, achieving SOTA on nine new benchmark datasets versus 15 baselines.
A small set of sparse autoencoder features in LLMs drives shifts between generous and selfish allocations in dictator games, with causal patching and steering confirming their role and generalization to other social games.
NeuFS selects active few-shot samples for LLMs by representing samples via neuron activation patterns and applying a dual-criteria strategy of diversity and neuron consensus to identify informative examples.
CR4T is a model-agnostic framework using lightweight risk detection and domain-conditioned rewriting to convert unsafe or refusal-style LLM responses into developmentally appropriate guidance for adolescents.
AInterviewer is an open-source multi-agent platform for AI-led qualitative interviews that integrates controlled question administration with LLMs and supports local models via a web GUI.
DCM-Agent improves LLM optimization solving by 11–21% on seven benchmarks via dual-cluster memory of Approaches, Checklists, and Pitfalls plus adaptive path switching.
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.
-
OptiVerse: A Comprehensive Benchmark towards Optimization Problem Solving
OptiVerse is a new benchmark spanning neglected optimization domains that shows LLMs suffer sharp accuracy drops on hard problems due to modeling and logic errors, with a Dual-View Auditor Agent proposed to improve performance.
-
Are LLM Uncertainty and Correctness Encoded by the Same Features? A Functional Dissociation via Sparse Autoencoders
Uncertainty and correctness in LLMs are encoded by distinct feature populations, with suppression of confounded features improving accuracy and reducing entropy.
-
Structure Guided Retrieval-Augmented Generation for Factual Queries
SG-RAG frames retrieval as subgraph matching to ensure LLMs meet every condition in factual queries and reports large gains over baselines on a new 120k-pair ERQA dataset.
-
Synthetic Sources?: Auditing Generative Search Engine Citations for Evidence of AI-Generated Sources
Audit of ChatGPT, Copilot, Gemini and Perplexity finds ~16% of cited sources are AI-generated across 712 queries on politics, health and environment.
-
SAGE: A Self-Evolving Agentic Graph-Memory Engine for Structure-Aware Associative Memory
SAGE is a self-evolving agentic graph-memory engine that dynamically constructs and refines structured memory graphs via writer-reader feedback, yielding performance gains on multi-hop QA, open-domain retrieval, and long-term agent benchmarks.
-
Towards Understanding Continual Factual Knowledge Acquisition of Language Models: From Theory to Algorithm
Theoretical analysis of continual factual knowledge acquisition shows data replay stabilizes pretrained knowledge by shifting convergence dynamics while regularization only slows forgetting, leading to the STOC method for attention-based replay selection.
-
LegalCiteBench: Evaluating Citation Reliability in Legal Language Models
LegalCiteBench reveals that current LLMs achieve under 7% accuracy on closed-book legal citation retrieval and completion tasks, with misleading answer rates above 94% for nearly all tested models.
-
LiteGUI: Distilling Compact GUI Agents with Reinforcement Learning
LiteGUI trains 2B/3B-scale GUI agents via SFT-free guided on-policy distillation and multi-solution dual-level GRPO to reach SOTA lightweight performance and compete with larger models.
-
Common-agency Games for Multi-Objective Test-Time Alignment
CAGE uses common-agency games and an EPEC algorithm to compute equilibrium policies that balance multiple conflicting objectives for test-time LLM alignment.
-
Verbal-R3: Verbal Reranker as the Missing Bridge between Retrieval and Reasoning
Verbal-R3 uses a verbal reranker to generate analytic narratives that guide retrieval and reasoning in LLMs, achieving SOTA results on complex QA benchmarks.
-
Detecting Hallucinations for Large Language Model-based Knowledge Graph Reasoning
LUCID detects hallucinations in LLM-KG reasoning by extracting node/edge features from attention and semantics then integrating them with KG structure in a GNN, achieving SOTA on nine new benchmark datasets versus 15 baselines.
-
Understanding the Mechanism of Altruism in Large Language Models
A small set of sparse autoencoder features in LLMs drives shifts between generous and selfish allocations in dictator games, with causal patching and steering confirming their role and generalization to other social games.
-
Neuron-Aware Active Few-Shot Learning for LLMs
NeuFS selects active few-shot samples for LLMs by representing samples via neuron activation patterns and applying a dual-criteria strategy of diversity and neuron consensus to identify informative examples.
-
CR4T: Rewrite-Based Guardrails for Adolescent LLM Safety
CR4T is a model-agnostic framework using lightweight risk detection and domain-conditioned rewriting to convert unsafe or refusal-style LLM responses into developmentally appropriate guidance for adolescents.
-
AInterviewer: A Platform for Designing and Conducting AI-led Qualitative Interviews
AInterviewer is an open-source multi-agent platform for AI-led qualitative interviews that integrates controlled question administration with LLMs and supports local models via a web GUI.
-
Dual-Cluster Memory Agent: Resolving Multi-Paradigm Ambiguity in Optimization Problem Solving
DCM-Agent improves LLM optimization solving by 11–21% on seven benchmarks via dual-cluster memory of Approaches, Checklists, and Pitfalls plus adaptive path switching.