LibEvoBench benchmark shows LLMs are version-oblivious on evolving APIs, with documentation helping but version specification not.
hub
Knowledge conflicts for llms: A survey
17 Pith papers cite this work. Polarity classification is still indexing.
17
Pith papers citing it
hub tools
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
Probes predicting future behaviors from intermediate steps enable Future Probe Controlled Generation for steering large reasoning models with minimal quality degradation.
TRACER uses token reassignment for concept-related items plus a coherence regularizer to unlearn specific concepts in generative recommendation while preserving utility better than baselines.
RAGognizer adds a detection head to LLMs for joint training on generation and token-level hallucination detection, yielding SOTA detection and fewer hallucinations in RAG while preserving output quality.
OR-VSKC provides 28,190 synthetic operating room images plus an expert subset to expose and reduce visual-semantic knowledge conflicts in multimodal models for surgical risk detection.
SHIFT reformulates neuron editing as learnable gate modulation on under 0.01% parameters to let LLMs adaptively balance contextual and parametric knowledge during RAG generation.
LLMs identify fabricated statistics in isolation (rates 0.76-1.00) but ignore numeric validity during synthesis, relying on a methodology-register representation that transfers across domains.
Persona prefixes reduce brand recommendation Jaccard similarity by 0.12-0.20, with mid-market brands swapping up to 75% of recommendations while category leaders remain ~80% consistent across OpenAI and Anthropic models.
RAG models exhibit a monitoring-control gap: they acknowledge epistemic conflicts in accumulating documents yet fail to constrain unsafe recommendations, with single-turn tests overestimating safety.
A three-regime framework resolves contradictions in LLM context vs. parametric knowledge conflicts by distinguishing single-source updating, competitive integration, and task-appropriate selection, with empirical confirmation of certainty gradients and task effects across five models.
Knowledge conflicts in hypernetwork LLM adaptation stem from constant adapter margins losing to frequency-dependent pretrained margins; selective layer boosting and conflict-aware triggering raise deep-conflict accuracy to 71-72.5% on Gemma-2B and Mistral-7B.
ConflictRAG introduces a conflict-aware RAG pipeline with two-stage detection (MLP + selective LLM), Entropy-TOPSIS credibility assessment, and a new CARS metric, reporting 88.7% F1 and 5.3-6.1% gains on benchmarks.
MeMo encodes new knowledge into a separate memory model that integrates with frozen LLMs, showing strong performance on QA benchmarks while avoiding catastrophic forgetting and working without access to model weights.
LLMs generally fail to maintain stable worldviews under adversarial conversational pressure, indicating they lack core beliefs akin to those in human cognition.
Decomposing long-context reasoning into atomic skills, synthesizing targeted pseudo-datasets, and applying RL improves LLM performance on long-context benchmarks by an average of 7.7%.
CRVA-TGRAG combines parent-document segmentation, ensemble retrieval, and teacher-guided fine-tuning to mitigate knowledge conflicts and improve accuracy in LLM-based CVE vulnerability analysis.
Two major AI providers diverge in which brands they recommend but converge on classifying the failure reasons, especially for low-prominence brands.
citing papers explorer
No citing papers match the current filters.