ThoughtFold applies introspective redundancy detection within correct CoT trajectories to create sub-trajectory spectra, then uses masked preference optimization to penalize redundant explorations, yielding 56% token reduction on DeepSeek-R1-Distill-Qwen-7B while preserving accuracy.
Mask-dpo: Generalizable fine-grained factuality alignment of llms
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
representative citing papers
Fine-tuning on new knowledge induces propagating hallucinations in LLMs by weakening attention to key entities, with mitigation via reintroducing known knowledge during later training stages.
SENTINEL reduces MLLM object hallucinations by over 90% via sentence-level early intervention with detector-bootstrapped preference data and C-DPO loss, outperforming prior SOTA on hallucination and capability benchmarks.
Proposes bidirectional token-wise KL regularizer and visual-contrastive grounding objective to create fine-grained on-policy preference pairs for medical LVLMs by minimally editing model outputs.
citing papers explorer
-
ThoughtFold: Folding Reasoning Chains via Introspective Preference Learning
ThoughtFold applies introspective redundancy detection within correct CoT trajectories to create sub-trajectory spectra, then uses masked preference optimization to penalize redundant explorations, yielding 56% token reduction on DeepSeek-R1-Distill-Qwen-7B while preserving accuracy.
-
Understanding New-Knowledge-Induced Factual Hallucinations in LLMs: Analysis and Interpretation
Fine-tuning on new knowledge induces propagating hallucinations in LLMs by weakening attention to key entities, with mitigation via reintroducing known knowledge during later training stages.
-
Mitigating Object Hallucinations via Sentence-Level Early Intervention
SENTINEL reduces MLLM object hallucinations by over 90% via sentence-level early intervention with detector-bootstrapped preference data and C-DPO loss, outperforming prior SOTA on hallucination and capability benchmarks.
-
Analyzing and Improving Fine-grained Preference Optimization in Medical LVLMs
Proposes bidirectional token-wise KL regularizer and visual-contrastive grounding objective to create fine-grained on-policy preference pairs for medical LVLMs by minimally editing model outputs.