Similar Items: Grounded or Guessing? LVLM Confidence Estimation via Blind-Image Contrastive Ranking
- The First Token Knows: Single-Decode Confidence for Hallucination Detection
- Beyond Confidence: Rethinking Self-Assessments for Performance Prediction in LLMs
- From Unstructured Recall to Schema-Grounded Memory: Reliable AI Memory via Iterative, Schema-Aware Extraction
- Grounded Satirical Generation with RAG
- MatryoshkaLoRA: Learning Accurate Hierarchical Low-Rank Representations for LLM Fine-Tuning
- Correct Is Not Enough: Training Reasoning Planners with Executor-Grounded Rewards