Cover Image

Safeguarding RAG Pipelines with GMTP: A Gradient-based Masked Token Probability Method for Poisoned Document Detection

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Kim, San, Kim, Jonghwi, Jeon, Yejin, Lee, Gary Geunbae
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Merging Triggers, Breaking Backdoors: Defensive Poisoning for Instruction-Tuned Language Models
by: Kim, San, et al.
Published: (2026)

Multi-Facet Blending for Faceted Query-by-Example Retrieval
by: Do, Heejin, et al.
Published: (2024)

Adversarial DPO: Harnessing Harmful Data for Reducing Toxicity with Minimal Impact on Coherence and Evasiveness in Dialogue Agents
by: Kim, San, et al.
Published: (2024)

An Investigation Into Explainable Audio Hate Speech Detection
by: An, Jinmyeong, et al.
Published: (2024)

Mixture-of-Experts with Intermediate CTC Supervision for Accented Speech Recognition
by: Lee, Wonjun, et al.
Published: (2026)

Learning When to Translate for Multilingual Reasoning
by: Kang, Deokhyung, et al.
Published: (2026)

Autoregressive Score Generation for Multi-trait Essay Scoring
by: Do, Heejin, et al.
Published: (2024)

Retrieval-Augmented Fine-Tuning With Preference Optimization For Visual Program Generation
by: Kang, Deokhyung, et al.
Published: (2025)

Key-Element-Informed sLLM Tuning for Document Summarization
by: Ryu, Sangwon, et al.
Published: (2024)

Multimodal Cognitive Reframing Therapy via Multi-hop Psychotherapeutic Reasoning
by: Kim, Subin, et al.
Published: (2025)

MIRROR: Multimodal Cognitive Reframing Therapy for Rolling with Resistance
by: Kim, Subin, et al.
Published: (2025)

TPA: Next Token Probability Attribution for Detecting Hallucinations in RAG
by: Lu, Pengqian, et al.
Published: (2025)

Leveraging the Interplay Between Syntactic and Acoustic Cues for Optimizing Korean TTS Pause Formation
by: Jeon, Yejin, et al.
Published: (2024)

Denoising Table-Text Retrieval for Open-Domain Question Answering
by: Kang, Deokhyung, et al.
Published: (2024)

Cross-lingual Back-Parsing: Utterance Synthesis from Meaning Representation for Zero-Resource Semantic Parsing
by: Kang, Deokhyung, et al.
Published: (2024)

Ontology-Free General-Domain Knowledge Graph-to-Text Generation Dataset Synthesis using Large Language Model
by: Kim, Daehee, et al.
Published: (2024)

K-COMP: Retrieval-Augmented Medical Domain Question Answering With Knowledge-Injected Compressor
by: Cho, Jeonghun, et al.
Published: (2025)

PanicToCalm: A Proactive Counseling Agent for Panic Attacks
by: Lee, Jihyun, et al.
Published: (2025)

GuRE:Generative Query REwriter for Legal Passage Retrieval
by: Kim, Daehee, et al.
Published: (2025)

HybridRAG: A Practical LLM-based ChatBot Framework based on Pre-Generated Q&A over Raw Unstructured Documents
by: Kim, Sungmoon, et al.
Published: (2025)

Adaptive Planning for Multi-Attribute Controllable Summarization with Monte Carlo Tree Search
by: Ryu, Sangwon, et al.
Published: (2025)

Behavior-Aware Item Modeling via Dynamic Procedural Solution Representations for Knowledge Tracing
by: Seo, Jun, et al.
Published: (2026)

Multi-Dimensional Optimization for Text Summarization via Reinforcement Learning
by: Ryu, Sangwon, et al.
Published: (2024)

Revisiting Early Detection of Sexual Predators via Turn-level Optimization
by: An, Jinmyeong, et al.
Published: (2025)

Why Do Multilingual Reasoning Gaps Emerge in Reasoning Language Models?
by: Kang, Deokhyung, et al.
Published: (2025)

First Token Probability Guided RAG for Telecom Question Answering
by: Chen, Tingwei, et al.
Published: (2025)

Leveraging What's Overfixed: Post-Correction via LLM Grammatical Error Overcorrection
by: Park, Taehee, et al.
Published: (2025)

Teach-to-Reason with Scoring: Self-Explainable Rationale-Driven Multi-Trait Essay Scoring
by: Do, Heejin, et al.
Published: (2025)

Autoregressive Multi-trait Essay Scoring via Reinforcement Learning with Scoring-aware Multiple Rewards
by: Do, Heejin, et al.
Published: (2024)

Speak & Spell: LLM-Driven Controllable Phonetic Error Augmentation for Robust Dialogue State Tracking
by: Lee, Jihyun, et al.
Published: (2024)

Exploring Iterative Controllable Summarization with Large Language Models
by: Ryu, Sangwon, et al.
Published: (2024)

STACK: Adversarial Attacks on LLM Safeguard Pipelines
by: McKenzie, Ian R., et al.
Published: (2025)

PSY-STEP: Structuring Therapeutic Targets and Action Sequences for Proactive Counseling Dialogue Systems
by: Lee, Jihyun, et al.
Published: (2026)

A Theoretical Analysis of Why Masked Diffusion Models Mitigate the Reversal Curse
by: Jeon, Moongyu, et al.
Published: (2026)

Towards Prompt Generalization: Grammar-aware Cross-Prompt Automated Essay Scoring
by: Do, Heejin, et al.
Published: (2025)

Understanding LLM Behavior in Multi-Target Cross-Lingual Summarization
by: Ryu, Sangwon, et al.
Published: (2026)

PicPersona-TOD : A Dataset for Personalizing Utterance Style in Task-Oriented Dialogue with Image Persona
by: Lee, Jihyun, et al.
Published: (2025)

Focus on the Core: Efficient Attention via Pruned Token Compression for Document Classification
by: Yun, Jungmin, et al.
Published: (2024)

Latent Preference Modeling for Cross-Session Personalized Tool Calling
by: Yoon, Yejin, et al.
Published: (2026)

LP Data Pipeline: Lightweight, Purpose-driven Data Pipeline for Large Language Models
by: Kim, Yungi, et al.
Published: (2024)