TrustMargin arbitrates between direct and RAG answers from a frozen LLM by combining a parametric-prior margin and an evidence-binding margin computed from model likelihoods, improving results on 2WikiMQA and CWQA.
Making retrieval- augmented language models robust to irrelevant context
11 Pith papers cite this work. Polarity classification is still indexing.
11
Pith papers citing it
citation-role summary
background 4
citation-polarity summary
roles
background 4polarities
background 4representative citing papers
ArbGraph resolves conflicts in RAG evidence by constructing a conflict-aware graph of atomic claims and applying intensity-driven iterative arbitration to suppress unreliable claims prior to generation.
Repurposing competency questions as runtime executable plans creates a controlled neuro-symbolic RAG architecture that produces evidence-closed stories from knowledge graphs.
A survey that defines Compound AI Systems, proposes a multi-dimensional taxonomy based on component roles and orchestration strategies, reviews four foundational paradigms, and identifies key challenges for future research.
Sentence-level contextual entrainment exists across LLMs, weakens with scale, and is localized to 2-4% of attention heads whose deactivation removes the effect without performance loss.
CoRM-RAG uses a cognitive perturbation protocol to simulate biases and trains an Evidence Critic to retrieve documents that support correct decisions even under adversarial query changes.
Self-RAG trains LLMs to adaptively retrieve passages on demand and self-critique using reflection tokens, outperforming ChatGPT and retrieval-augmented Llama2 on QA, reasoning, and fact verification.
SingGuard introduces a policy-adaptive multimodal LLM guardrail with dynamic reasoning regimes and SingGuard-Bench, reporting SOTA F1 scores across 35 datasets and improved policy-following accuracy under runtime shifts.
LLM safety judges resist adjusting evaluations when given contradictory context or new safety definitions, despite some ability to learn from new information.
PRA-RAG is a new aggregation algorithm for RAG that claims provable robustness bounds against poisoned retrieved texts and reduces attack success rate to 1% while keeping 71% accuracy.
A survey of RAG paradigms, components, benchmarks, and challenges for improving LLMs on knowledge-intensive tasks.
citing papers explorer
-
TrustMargin: Training-Free Arbitration between Parametric Memory and Retrieved Evidence in Large Language Models
TrustMargin arbitrates between direct and RAG answers from a frozen LLM by combining a parametric-prior margin and an evidence-binding margin computed from model likelihoods, improving results on 2WikiMQA and CWQA.
-
ArbGraph: Conflict-Aware Evidence Arbitration for Reliable Long-Form Retrieval-Augmented Generation
ArbGraph resolves conflicts in RAG evidence by constructing a conflict-aware graph of atomic claims and applying intensity-driven iterative arbitration to suppress unreliable claims prior to generation.
-
Competency Questions as Executable Plans: a Controlled RAG Architecture for Cultural Heritage Storytelling
Repurposing competency questions as runtime executable plans creates a controlled neuro-symbolic RAG architecture that produces evidence-closed stories from knowledge graphs.
-
From Standalone LLMs to Integrated Intelligence: A Survey of Compound Al Systems
A survey that defines Compound AI Systems, proposes a multi-dimensional taxonomy based on component roles and orchestration strategies, reviews four foundational paradigms, and identifies key challenges for future research.
-
Sentence-Level Contextual Entrainment in Large Language Models
Sentence-level contextual entrainment exists across LLMs, weakens with scale, and is localized to 2-4% of attention heads whose deactivation removes the effect without performance loss.
-
Beyond Semantic Relevance: Counterfactual Risk Minimization for Robust Retrieval-Augmented Generation
CoRM-RAG uses a cognitive perturbation protocol to simulate biases and trains an Evidence Critic to retrieve documents that support correct decisions even under adversarial query changes.
-
Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection
Self-RAG trains LLMs to adaptively retrieve passages on demand and self-critique using reflection tokens, outperforming ChatGPT and retrieval-augmented Llama2 on QA, reasoning, and fact verification.
-
SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning
SingGuard introduces a policy-adaptive multimodal LLM guardrail with dynamic reasoning regimes and SingGuard-Bench, reporting SOTA F1 scores across 35 datasets and improved policy-following accuracy under runtime shifts.
-
Safety is Contextual, LLM-Judges Are Not: Navigating the Rigid Priors of Evaluators
LLM safety judges resist adjusting evaluations when given contradictory context or new safety definitions, despite some ability to learn from new information.
-
PRA-RAG: Provably Robust Aggregation in Retrieval-Augmented Generation against Retrieval Corruption
PRA-RAG is a new aggregation algorithm for RAG that claims provable robustness bounds against poisoned retrieved texts and reduces attack success rate to 1% while keeping 71% accuracy.
-
Retrieval-Augmented Generation for Large Language Models: A Survey
A survey of RAG paradigms, components, benchmarks, and challenges for improving LLMs on knowledge-intensive tasks.