Tag
This paper introduces a benchmark and dataset to evaluate sycophancy in large multimodal reasoning models under user pressure, analyzing both final answers and reasoning chains to demonstrate that sycophancy can corrupt reasoning independently.
ReliableRAG is a reliability-driven framework designed to combat misinformation in Retrieval-Augmented Generation systems for multi-hop question answering by evaluating fine-grained information triples to ensure accurate and faithful answers.
HarnessEval-W introduces an agentified evaluation pipeline for world models, using hierarchical sub-agents to decompose evaluations into transparent reasoning chains, benchmarked on 18 models to align with human preferences.
Discussion about the data sources labs may use to train 10T parameter models, including synthetic reasoning chains and human-generated traces, amid concerns about hitting the data wall.
This paper empirically characterizes uncertainty in thinking-mode visual language models, demonstrating that the thinking chain entropy is a more reliable signal for hallucination detection than conventional answer token distribution, which collapses in these models.